Interviewing Experts

Research Methods Series

General Editors: Bernhard Kittel, Professor of Social Science Methodology,
Department of Social Sciences, Carl von Ossietzky Universität Oldenburg, Germany
and Benoît Rihoux, Professor of Political Science, Université catholique de Louvain
(UCL), Belgium.
In association with the European Consortium for Political Research (ECPR), Palgrave
Macmillan is delighted to announce the launch of a new book series dedicated to
producing cutting-edge titles in Research Methods. While political science currently
tends to import methods developed in neighbouring disciplines, the series contrib-
utes to developing a methodological apparatus focusing on those methods which are
appropriate in dealing with the specific research problems of the discipline.
The series provides students and scholars with state-of-the-art scholarship on
methodology, methods and techniques. It comprises innovative and intellectually
rigorous monographs and edited collections which bridge schools of thought and cross
the boundaries of conventional approaches. The series covers both empirical-analytical
and interpretive approaches, micro and macro studies, and quantitative and qualita-
tive methods.
Titles include:
Alexander Bogner, Beate Littig and Wolfgang Menz (editors)
INTERVIEWING EXPERTS
Audie Klotz and Deepa Prakash (editors)
QUALITATIVE METHODS IN INTERNATIONAL RELATIONS
A Pluralist Guide
Lane Kenworthy and Alexander Hicks (editors)
METHOD AND SUBSTANCE IN MACROCOMPARATIVE ANALYSIS

Research Methods Series

Series Standing Order ISBN 978–0230–20679–3 hardcover
Series Standing Order ISBN 978–0230–20680–9 paperback
(outside North America only)
You can receive future titles in this series as they are published by placing a standing order.
Please contact your bookseller or, in case of difficulty, write to us at the address below with your
name and address, the title of the series and one of the ISBNs quoted above.
Customer Services Department, Macmillan Distribution Ltd, Houndmills, Basingstoke, Hampshire
RG21 6XS, England
Interviewing Experts

Edited by

Alexander Bogner
Austrian Academy of Sciences, Austria

Beate Littig
Institute of Advanced Studies, Austria

and
Wolfgang Menz
Institute for Social Science Research, Germany
Selection and editorial matter © Alexander Bogner, Beate Littig and
Wolfgang Menz 2009
Chapters © their authors 2009
English Language Translation © respective authors 2009
All rights reserved. No reproduction, copy or transmission of this
publication may be made without written permission.
No portion of this publication may be reproduced, copied or transmitted
save with written permission or in accordance with the provisions of the
Copyright, Designs and Patents Act 1988, or under the terms of any licence
permitting limited copying issued by the Copyright Licensing Agency,
Saffron House, 6-10 Kirby Street, London EC1N 8TS.
Any person who does any unauthorized act in relation to this publication
may be liable to criminal prosecution and civil claims for damages.
The authors have asserted their rights to be identified as the authors
of this work in accordance with the Copyright, Designs and Patents Act 1988.
First published 2009 by
PALGRAVE MACMILLAN
Palgrave Macmillan in the UK is an imprint of Macmillan Publishers Limited,
registered in England, company number 785998, of Houndmills, Basingstoke,
Hampshire RG21 6XS.
Palgrave Macmillan in the US is a division of St Martin’s Press LLC,
175 Fifth Avenue, New York, NY 10010.
Palgrave Macmillan is the global academic imprint of the above companies
and has companies and representatives throughout the world.
Palgrave® and Macmillan® are registered trademarks in the United States,
the United Kingdom, Europe and other countries.
ISBN: 978–0–230–22019–5 hardback
ISBN: 0–230–22019–3 hardback
This book is printed on paper suitable for recycling and made from fully
managed and sustained forest sources. Logging, pulping and manufacturing
processes are expected to conform to the environmental regulations of the
country of origin.
A catalogue record for this book is available from the British Library.
Library of Congress Cataloging-in-Publication Data
Interviewing experts / edited by Alexander Bogner, Beate Littig, and
Wolfgang Menz.
p. cm.—(Research methods series)
Includes bibliographical references and index.
ISBN 978–0–230–22019–5 (alk. paper)
1. Interviewing. 2. Specialists – Interviews – Methodology. I. Bogner,
Alexander. II. Littig, Beate. III. Menz, W. (Wolfgang)
BF637.I5I57 2009
001.4!32—dc22 2009013661
10 9 8 7 6 5 4 3 2 1
18 17 16 15 14 13 12 11 10 09
Printed and bound in Great Britain by
CPI Antony Rowe, Chippenham and Eastbourne
Contents

List of Illustrations vii

Notes on Contributors viii

Introduction: Expert Interviews – An Introduction

to a New Methodological Debate 1
Alexander Bogner, Beate Littig and Wolfgang Menz

Part I Theoretical Concepts:

Methodology of Expert Interviews
1 The Expert Interview and Changes in Knowledge Production 17
Michael Meuser and Ulrike Nagel
2 The Theory-Generating Expert Interview:
Epistemological Interest, Forms of Knowledge, Interaction 43
Alexander Bogner and Wolfgang Menz
3 At Eye Level: The Expert Interview – a Talk between
Expert and Quasi-expert 81
Michaela Pfadenhauer
4 Interviewing the Elite – Interviewing Experts:
Is There a Difference? 98
Beate Littig

Part II Methodological Practice: Generating Data

5 On Interviewing “Good” and “Bad” Experts 117
Jochen Gläser and Grit Laudel
6 Interviewing Experts in Political Science:
A Reflection on Gender and Policy Effects Based on
Secondary Analysis 138
Gabriele Abels and Maria Behrens
7 Expert Interviews on the Telephone: A Difficult Undertaking 157
Gabriela B. Christmann
8 Expert versus Researcher:
Ethical Considerations in the Process of Bargaining a Study 184
Vaida Obelenė

v
vi Contents

Part III Fields of Application:

Applications of Expert Interviews in
Different Fields of Research
9 How to Interview Managers? Methodical and Methodological
Aspects of Expert Interviews as a Qualitative Method in
Empirical Social Research 203
Rainer Trinczek
10 Expert Interviews in Interpretive Organizational Research 217
Ulrike Froschauer and Manfred Lueger
11 Between Scientific Standards and Claims to Efficiency:
Expert Interviews in Programme Evaluation 235
Angela Wroblewski and Andrea Leitner
12 The Delphi Method: Eliciting Experts’ Knowledge in
Technology Foresight 252
Georg Aichholzer

Index 275
Illustrations

Tables

2.1 Typology of interaction situations and interview strategies 68

5.1 Quality-dependent communication situations in
interviews with researchers 122
5.2 Differentiation of experimental physicists in
Australia and Germany according to their research
performance and acquisition of external funds 133
6.1 Typical effects in conversational interaction 144
12.1 The Austrian Technology Delphi:
organization, roles and tasks 257
12.2 Size of the two main Expert Delphi rounds 262

Figures

4.1 Differentiating between experts and the elite 108

5.1 Area of overlap of the two definitions of expert interviews 119
5.2 Relationship between performance levels and
the adaptation of the research content 134
7.1 The research design in the context of the extended project 158

vii
Contributors

Gabriele Abels is a professor for comparative politics and European

integration at the Institute of Political Science, University of Tübingen.
Her research interests are European integration, political participation,
biotechnology policy, gender studies. She is the co-editior of “femina
politica: Zeitschrift für feministische Politik-Wissenschaft.”
Her most recent publications include: Abels, G. and Lepperhoff, J. (forth-
coming 2009) “Frauen-, Geschlechter- und Intersektionalitätsforschung.
Methodologische Entwicklungen und offene Fragen” in Barbara
Friebertshäuser et al. (eds) Handbuch Qualitative Forschungsmethoden in der
Erziehungswissenschaft, 2nd edn (Weinheim, München: Juventa) and Abels, G.
(2007) “Trade and Human Rights: Inter- and Supranational Regulation of
GMOs and ART” in Montpetit, E. et al. (eds) The Politics of Biotechnology in
North America and Europe: Policy Networks, Institutions and Internationalization
(Lanham, MD: Lexington Books), pp. 35–59.
Georg Aichholzer is project director and senior researcher at the Institute
of Technology Assessment (ITA) of the Austrian Academy of Sciences
and senior lecturer at Vienna University of Economics and Business
Administration. Trained in Social Sciences (PhD in Sociology) his research
interests focus on technology assessment, interrelations of information
technology and society, and related policies. A major field of study are
technological innovations in government and governance (electronic pub-
lic services, electronic participation).
Among his publications are: Aichholzer, G. and Burkert, H. (eds) (2004)
Public Sector Information in the Digital Age. Between Markets, Public Management
and Citizens? (Rights, Cheltenham, UK, and Northampton, MA: Edward
Elgar Publishing) and Aichholzer, G. (2007) “Opening the Black Box:
Economic and Organisational Effects of e-Government” in Remenyi, D.
(ed.) Proceedings of the 3rd International Conference on e-Government, University
of Quebec at Montreal, Canada, 26–28 September 2007 (Dublin: Academic
Conferences International), pp. 1–10.
Maria Behrens is professor at the Department of Political Science at the
University Wuppertal (Germany). Research Areas: political coordination
and regulation of conflicts in international politics (global governance);
the relationship of intergovernmental and transnational forms of political
decision-making (private governance).
Publications: Behrens, M. (forthcoming 2009) “Beyond the competi-
tive race: US and EU foreign trade policy” in Wynn, N. (ed.) Conflict and

viii
 Contributors ix

Consensus: Transatlantic Perspectives (Cambridge: Cambridge University

Press); Behrens, M. (2007) “Global Governance” in Benz, A. et al. (eds)
Handbuch Governance (Wiesbaden: VS-Verlag), pp. 311–24 and Behrens, M.
(2005) Globalisierung als politische Herausforderung, Global Governance
zwischen Utopie und Realität (Wiesbaden: VS-Verlag).
Alexander Bogner is a sociologist by training and researcher at the Institute
for Technology Assessment of the Austrian Academy of Sciences in Vienna.
His main research interests are science and technology studies, sociology of
expertise and methods of empirical research.
His publications include: Bogner, A. (2005) “How Experts Draw
Boundaries. Dealing with non-knowledge and uncertainty in prenatal
testing” in Science, Technology and Innovation Studies 1, 17–37 and Bogner, A.
and Menz, W. (2006) “Science crime. The Korean cloning scandal and the
role of ethics” in Science and Public Policy 33, 601–12.
Gabriela B. Christmann is head of the research department “Dynamics
of Communication, Knowledge and Spatial Development” at the Leibniz
Institute for Regional Development and Structural Planning (IRS) in Erkner
(near Berlin). Her main research interests include the sociology of know-
ledge and culture, urban sociology, communication analysis, and methods
of qualitative research.
Recent publication: Christmann, G. (2008) “The Power of Photographs
of Buildings in the Dresden Urban Discourse. Towards a Visual Discourse
Analysis” in Forum Qualitative Sozialforschung / Forum: Qualitative Social
Research 9(3), Art. 11, http://nbnresolving.de/urn:nbn:de:0114-fqs0803115
(accessed on 16 May 2009).
Ulrike Froschauer is Assistant Professor at the Institute of Sociology,
University of Vienna, Austria. Her main research interests are in the field
of the sociology of organisations, sociology of organisational consulting,
methodology and methods of qualitative research and evaluation.
Among her publications are: Froschauer, U. (2006) “Veränderungsdynamik
in Organisationen” in Tänzler, D., Knoblauch, H., Soeffner, H.-G. (eds) Zur
Kritik der Wissensgesellschaft (Konstanz: UVK), pp. 157–83 and Froschauer, U.,
Lueger, M. (2003) Das qualitative Interview. Zur Praxis Interpretativer Analyse
sozialer Systeme (Wien: WUV-UTB).
Jochen Gläser, PD (Free University Berlin) Dr (Humboldt-University Berlin)
is a senior researcher at the Technical University of Berlin’s Center for
Technology and Society. His major research interests are in the sociology of
science, sociological theory and qualitative methods.
Recent publications include: Gläser, J. and Whitley, R. (eds) (2007)
The Changing Governance of the Sciences: The Advent of Research Evaluation
Systems (Berlin: Springer); Gläser, J. and Laudel, G. (2007) “Interviewing
Scientists” in STI-Studies 3/2 and Gläser, J. and Laudel, G. (2008) “Creating
x Contributors

Competing Constructions by Re-Analysing Qualitative Data” in Historical

Social Research 33/3.
Grit Laudel, Dr (University of Bielefeld) is a senior researcher at the Rathenau
Institute in The Hague, Netherlands. She is interested in the sociology of
science and the methodology and methods of qualitative research.
Recent publications are: Gläser, J. and Laudel, G. (2008) “From apprentice to
colleague: the metamorphosis of Early Career Researchers” in Higher Education
56/3; Gläser, J. and Laudel, G. (2007) “Interviewing Scientists” in STI-Studies
3/2; and Gläser, J. and Laudel, G. (2008) “Creating Competing Constructions
by Re-Analysing Qualitative Data” in Historical Social Research 33/3.
Andrea Leitner is a member of the academic staff at the Institute for Advanced
Studies in Vienna. Her main research interests are labour market processes,
especially interfaces between paid and unpaid work as well as between labour
market and education, gender studies and evaluation research.
Her publications include: Leitner, A. (2007) Frauenförderung im Wandel.
Gender Mainstreaming in der österreichischen Arbeitsmarktpolitik (Frankfurt/
New York: Campus) and Leitner, A. and Wroblewski, A. (June, 2006) “Welfare
States and Work-Life Balance. Can Good Practices Be Transferred from the
Nordic Countries to Conservative Welfare States?” in European Societies 8(2),
295–317.
Beate Littig is a sociologist, head of the Sociology Department at the
Institute for Advanced Studies (IHS) in Vienna and a permanent lecturer at
the University of Vienna. Her research interests include qualitative method-
ology, gender studies, sociology of practice and sustainable work.
Recent publications include: Littig, B. (2006) Book review: Lewis A. Dexter
(2006). “Elite and Specialized Interviewing.” With a New Introduction
by Alan Ware and Martín Sánchez-Jankowski (16 Paragraphs) in Forum
Qualitative Sozialforschung / Forum: Qualitative Social Research 9(1), Art. 5,
http://nbn-resolving.de/urn:nbn:de:0114-fqs080151 (accessed on 16 May
2009) and Hildebrandt, E. and Littig, B. (eds) (2006) “Concepts, Approaches
and Problems of Work-Life-Balance,” European Societies 8/2, Special Issue.
Manfred Lueger is Associate Professor at the Institute for Sociology
and Social Research, Vienna University of Economics and Business
Administration. His research has focused on methodology and methods of
interpretative social research, the analysis of organisations, and entrepre-
neurship education.
Publications include: Lueger, M. (2001) Auf den Spuren der sozialen Welt.
Methodologie und Organisierung interpretativer Sozialforschung (Frankfurt,
Berlin, Bern, Bruxelles, New York, Oxford, Wien: Peter Lang Verlag) and
Hammerschmid, G., Lueger, M., Meyer R., Sandner, K. (2005) “Contextualizing
Influence Activities. An Objective Hermeneutical Approach” in Organization
Studies 2005/8 (26), 1145–68.
 Contributors xi

Wolfgang Menz, Dr, a researcher at the Institute for Social Science Research,
ISF München. He studied Sociology und Political Sciences at the Universities
of Marburg/Germany, Frankfurt/Germany and Edinburgh/UK. Holder of
a PhD-Scolarship of the Hans-Böckler-Stiftung (2003–2006), Researcher
at the Johann Wolfgang Goethe-Universität, Frankfurt (2001–2003) and
the Institut für Sozialforschung, Frankfurt (2003–2007), lecturer at the
Universities of Vienna/Austria and Frankfurt/Germany. His research inter-
ests include the sociology of work and organisation, science studies and
methods of qualitative research.

Michael Meuser is professor for the sociology of gender relations at the

University of Dortmund, Germany. His research interests are: sociology of
gender, sociology of knowledge, qualitative methods, political sociology,
sociology of the body.
Recent publications: Meuser, M. (forthcoming) “Gender Competence?
Gender Mainstreaming, Managing Diversity and the Professionalisation of
Gender Politics in Germany” in Mustafa Özbilgin (ed.) Diversity, Equalitiy
and Inclusion at Work: a Research Companion (Cheltenham/New York: Edward
Egar Press) and Klein, G. and Meuser, M. (eds) Ernste Spiele. Zur politischen
Soziologie des Fußballs (Bielefeld: transcript 2008).

Ulrike Nagel, Dr. phil. habil., Senior Lecturer of Microsociology,

Otto-von-Guericke-Universität Magdeburg, Department of Sociology.
Research interests: micro-sociology, sociology of professions, European
identity, biography analysis, qualitative methods of social research.
Publications include: Nagel, U., Teipen, Ch. and Velez, A. (2005) “Die
Macht der Verhältnisse und die Stärke des Subjekts. Eine Studie über
Ostdeutsche Manager vor und nach 1989. Zugleich eine biographietheo-
retische Erklärung für Stabilität und Instabilität der DDR” in ZBBS 2/2005,
pp. 277–302 and Nagel, U. (1997) Engagierte Rollendistanz. Professionalität in
biographischer Perspektive (Opladen: Leske & Budrich, Reihe Biographie und
Gesellschaft).

Vaida Obelenė, a sociologist by training, works on social stratification in

post-communist Europe. Currently she is a PhD researcher at the European
University Institute (Florence) in the department of Social and Political
Sciences. Her thesis is entitled: “Discontinuity in Elite Formation: Former
Komsomol Functionaries in the Period of Post-communist Transition in
Lithuania and Belarus.”
Publications: Jasiukaityte, Vaida and Reiter, H. (2004) “Jugendpolitik in
Ländern des Übergangs – welchen Beitrag kann sie zur Zivilgesellschaft
in Europa leisten?” in Lauritzen, P. and Otten, H. (eds) Jugendarbeit und
Jugendpolitik in Europa (Wiesbaden: Verlag für Sozialwissenschaften),
pp. 181–93 and Jasiukaityte, Vaida (2004) Making a difference with minority
youth in Europe (Strasbourg: Council of Europe).
xii Contributors

Michaela Pfadenhauer is Professor for Sociology at Karlsruhe University.

Diplom at Bamberg University (Sociology and Political Science); PhD
at Dortmund University (Sociology). She is Coordinator of the German
Sociological Association research network Sociology of Profession. Her
work domains are Sociology of Knowledge, Sociology of Everyday Culture
(Posttraditional Communities), Consumerism and Professionalism.
Publications: Pfadenhauer, M. (2006) “Crisis or Decline? Problems of
legitimation and loss of confidence of modern professionalism” in Current
Sociology 54 (4/2006), 565–78 and Pfadenhauer, M. “Ethnography of Scenes.
Towards a Sociological Life-world Analysis of (Post-traditional) Community-
building” (31 paragraphs) in Forum: Qualitative Social Research [On-line
Journal] 6(3), Art. 43. Available at: http://www.qualitative-research.net/
fqs-texte/3–05/05–3-43-e.htm (Date of Access: 09 23, 2005).
Rainer Trinczek is now professor at the Friedrich-Alexander-Universität
Nürnberg-Erlangen. His main research interests are industrial relations,
sociology of management, sociology of work, and qualitative research
methods.
His publications include: Trinczek, R. (2006) “Work-Life Balance and
Flexible Working Hours – The German Experience” in Blunsdon, B., Blyton, P.,
Dastmalchian, A. and Reed, K. (eds) Work-Life Integration. International
Perspectives on the Balancing of Multiple Roles (Basingstoke/New York: Palgrave
MacMillan), pp. 113–34, and Artus, I., Böhm, S., Lücking, S., Trinczek, R. (eds)
(2006) Betriebe ohne Betriebsrat. Informelle Interessenvertretung in Unternehmen
(Frankfurt/New York: Campus).
Angela Wroblewski studied Sociology in Vienna, since 1998 member of the
academic staff at the HIS, lecturer at the Vienna University of Economics
and Business Administration and the University of Vienna. Research inter-
ests: evaluation of social programmes, gender and education research.
Recent publication: Cook, T.D., Steiner, P.M. and Wroblewski, A. (forthcoming)
“Randomised experiments and quasi-experimental designs in educational
research” in Cousins, B., Ryan, K. (eds) International Handbook of Educational
Evaluation (Thousand Oaks, CA: Sage).
Introduction: Expert
Interviews – An Introduction
to a New Methodological Debate
Alexander Bogner, Beate Littig and Wolfgang Menz

Before we go any further, we would like to begin by providing the reader

with a step-by-step introduction to the methodological debate surround-
ing expert interviews. In doing so, we will start with a brief discussion of
the generally accepted advantages and risks of expert interviews in research
practice (1). We will follow this by outlining current trends in the socio-
logical debate regarding experts and expertise, since expert interviews are –
at least on the surface – defined by their object, namely the expert (2). We
will then conclude with a look at the current methodological debate regard-
ing expert interviews, an overview of the layout and structure of this book,
as well as summaries of the 12 articles it contains (3).

I.1 Expert interviews: easy to manage?

The debate surrounding expert interviews is a recent one. The article

published in 1991 by Meuser and Nagel was instrumental in launch-
ing an initial systematic debate on expert interviews in Germany. But it
would be another ten years before the debate actually gained significant
momentum, in an upward trend that is also reflected in current method-
ology handbooks (for example Flick, 2006, Przyborski and Wohlrab-Sahr,
2008).1 The focus of this interest lies primarily on issues of what con-
stitutes an expert, the differences between the various forms of expert
interviews and their role in research design, as well as the specifics of
interviewing and interaction in comparison to other qualitative inter-
view forms.
The use of expert interviews has long been popular in social research. The
actual role of the expert interviews in individual research design, their form
and the methods used to analyse the results might vary from case to case,
but there are still a number of common, practical reasons for their popular-
ity in research.

1
2 Alexander Bogner, Beate Littig and Wolfgang Menz

Firstly, in relative terms, talking to experts in the exploratory phase of a

project is a more efficient and concentrated method of gathering data than,
for instance, participatory observation or systematic quantitative surveys.
Conducting expert interviews can serve to shorten time-consuming data
gathering processes, particularly if the experts are seen as “crystallization
points” for practical insider knowledge and are interviewed as surrogates for a
wider circle of players. Expert interviews also lend themselves to those kinds
of situations in which it might prove difficult or impossible to gain access to
a particular social field (as is the case, for instance, with taboo subjects).
The economic aspect also extends to the broad, practical matter of initiat-
ing and conducting such interviews. The organizational structures behind
the experts in institutions (for example their secretaries or press offices) can
often serve as an easy point of entry to the field of research. Furthermore,
if the targeted expert is not only willing to participate, but also holds a key
position in the organization, opportunities for expanding the researcher’s
access to the field may well also be unearthed in the interview. Sometimes,
the expert will even indicate additional potential interviewees with expert-
ise in a particular field during the interview itself. Equipped with the added
bonus of the support of an expert in a key position, the researcher may then
often find it easier to gain access to an extended circle of experts.
Beyond the direct benefits, it is also evident that expert interviews offer
researchers an effective means of quickly obtaining results and, indeed, of
quickly obtaining good results. Frequently, the fact that the interviewer and
the interviewee share a common scientific background or relevance system
can increase the level of motivation on the part of the expert to participate in
an interview. A shared understanding of the social relevance of the research
can then often be assumed, largely eliminating the need for further justifica-
tion. A number of secondary motivating factors also make it comparatively
easy to encourage and motivate experts to participate in such interviews:
the professionalism of people familiar with being in the public eye; silent
awareness of the scientific and/or political relevance of their field of activity
or personal achievements; the desire to help “make a difference” – no matter
how small; professional curiosity about the topic and field of research; an
interest in sharing one’s thoughts and ideas with an external expert.
Regardless of what might be myth and what is reality, the anticipated
promise of rapid and unproblematic access to objective data makes expert
interviews an extremely appealing option for empirical social researchers.
But is the expert interview method really quite so simple and uncomplicated?
If so, does this then render methodological considerations superfluous? Or
are expert interviews in some ways just too tempting? Do they not – in their
naïve belief in the totality of expert knowledge – harbour the danger of advo-
cating a pre-reflexive definition of what constitutes an expert? Or the risk
of granting the undisputed relevance of expert knowledge a standing that
would ultimately constitute the non-validated confirmation and, thus, the
 An Introduction to a New Methodological Debate 3

legitimization of social hierarchies? A critical look at the current social sci-

ence debate on experts and expertise could have a corrective effect here, and
we will examine these questions in more detail in the following section.

I.2 Trends in social science research into expertise

As far as the expert and expertise are concerned, recent social science
research trends have proved relatively stable. The “expert” has edged into the
centre of theoretical interest from both a theory of society and a democratic
theory perspective as well as from the sociology of knowledge, scientific or
technical research standpoints (cf. Jasanoff and others, 1995, Bogner and
Torgersen, 2005). Yet the literature on expert interview methods remains
largely unmoved by this trend.
In scientific and technical circles, researchers are currently rethinking
what really constitutes an expert and where the “relevant” knowledge for
political decisions actually lies. In this context, Collins and Evans (2002)
maintain that the sociology of expertise is currently entering a third wave.
Based on their timeline, the first wave is embodied in the golden age of the
expert with its clear and recognized horizontal division between experts
and lay people. The expert as agent of truth and authority encounters a
political system that uses its power to enforce expertise (“truth speaks to
power”). The second wave is characterized by social constructivism in its
prime, with its focus on demystifying science: knowledge is deciphered
as a social activity and the validity of expert knowledge as a construction
process is decoded. Likewise, challenging the boundary between experts
and lay people accelerates the debate on the democratization of expertise
(cf. Maasen and Weingart, 2005). To counter the constructivist breaking of
the expert’s spell – which it is ultimately claimed would lead to epistemic
anarchy – Collins and Evans (2007) propose a “realist approach” as the third
wave. “The realist approach (...) starts from the view that expertise is the
real and substantive possession of groups of experts and that individuals
acquire real and substantive expertise through their membership of those
groups” (Collins and Evans, 2007, p. 3). If some form of institutionalized and
autonomous area of science is to remain, Collins and Evans maintain that
there is a need for genuine expertise based on expert knowledge and, thus,
participation limits. The fact that they only consider “technical expertise”
contributes to the suggestiveness of their ideas and, at the same time, reveals
the cognitivistic constraints of their analysis of expertise. Specific prob-
lem framing is used to tailor questions in a way that makes this “technical
expertise,” that is the expert’s factual knowledge (and not some form of lay
practical knowledge), relevant. Expertise is relevant not (solely) by virtue of
its own intrinsic quality, but also as a result of external conditions.
From the political science and democratic theory perspectives, expertise
is viewed primarily as a challenge for democracy. Is it not the case – as is
4 Alexander Bogner, Beate Littig and Wolfgang Menz

commonly feared – that the worldviews of experts and the things they see
as relevant preform parliamentary decisions because they are – by virtue
of their authority – considered to be the factual basis behind the political
debate? The tension between expertise as necessary basis and ideological
preformation of political decisions was already a point of discussion in the
technocratic debate of the 1960s. Stephen Turner (2002) was the latest to
question whether expertise represents a fundamental threat to liberal dem-
ocracy. He bases his own trust in the democratic compatibility of expertise
on the basic dubitability of the latter, realized not least in public protests
and scientific self-criticism – effectively turning scientific criticism into an
empirical sign of a functioning democracy.
Similar positions regarding the democratizing side effects of the expertise
boom can also be found in the theory of society debate. For example, in the
“reflexive modernization” debate – with its explicit socio-critical emphasis –
expert knowledge is seen as the driving force and crystallization point for
social conflict and as the stimulus and medium for an emancipated battle
for the conditions of definition. Seen from this perspective, the side effects
of the modernization process (for example global warming, ecological devas-
tation, genetic manipulation) turn enforcement of the Enlightenment ideal
of perfect control over society through expert knowledge into a moment of
social self-enlightenment (cf. Beck, 1992). Since these risks and dangers are
abstract in nature, scientific knowledge (that is critical experts) is required
to turn them into social fact. In this regard, the expert is becoming more
diversified. A new actor is taking to the stage in the battle for rationality –
in the form of the counter- or anti-expert who advises critical NGOs in
risk controversies. However, experts as agents of different rationality models
will only be effective insofar as they actually succeed in influencing pub-
lic awareness through the media. Consequently, an ability to put specific
knowledge to use for political gain is a constitutive characteristic of this
type of “post-traditional” expert. The key from this perspective is to inter-
pret the world in a high-profile and influential (but not necessarily new)
manner and thus – as (counter-)expert – become a powerful voice in the
battle for the conditions of definition.
Expert knowledge is also accorded a central role in Giddens’ moderniza-
tion theory. He discusses the changes in the modern world from the point
of view of knowledge dynamics (not the side effects). Expert knowledge is
part of the “institutional reflexivity” (Giddens, 1991, p. 20) which supposes
that all premises of individual and organizational activity will be routinely
examined in the light of new information about such practices. Furthermore,
experts become important when people find themselves having to deal
with abstract systems (whose internal workings they do not understand). It
is up to the expert here to convince them to trust such (primarily technical)
abstract systems, for example by means of appropriate self-staging strategies
(Giddens, 1990). This is by no means an easy task, because in late modernity
 An Introduction to a New Methodological Debate 5

the growth in relevance of expert knowledge is paradoxically accompanied

by a crisis of recognition on the part of the experts. Even though progressive
detraditionalization in many areas pushes us to base our decisions on expert
knowledge, we are nonetheless also experts at always getting a “second
opinion.” Particularly in cases where problems are categorized and handled
as issues of values not knowledge (for example in genetic counselling), the
previously rigid hierarchy between experts and lay people tends to give way
to more flexible and situative structures of interaction (Bogner, 2005).
Giddens does not, however, relate this trend to the problem framing
aspect; instead he puts it in the context of an ambivalent attitude to special-
ization: specialization might well safeguard the continued existence of the
expert and the development of new forms of expertise, but it also reduces
experts to mere representatives of specialized knowledge, knowledge they
find increasingly difficult to keep abreast of. The expert becomes, in many
aspects, increasingly the lay person. This “laicization” of the expert changes
the relationship of trust between lay people and experts: expertise must
be increasingly stage-managed to gain acceptance. Beck places greater
emphasis on the aspect of the acceptance or validity of knowledge claims
than Giddens, because his focus on risk controversies (which are character-
ized by expert dissent and rivalry between kindred forms of knowledge),
causes him to think more in terms of the expert/counter-expert distinction
than that of the expert/lay person. This sensitizes us to the fact that some-
thing must come into play as far as the acceptance of expert knowledge is
concerned, a point that will be discussed later in the context of political effi-
cacy and practical effectiveness (cf. Bogner and Menz, in this volume).

I.3 The articles in this book:

methodology and method in expert interviews

What lessons can we draw from this brief foray into social science research
into expertise for our own methodological debate? First and foremost
the realization that the naïve image of the expert as source of objective
information – on which one or the other simplified notion of successful
expert interviews is based – has long become problematic. In our context,
this confirms a need for increased reflection on expert interviews and on
methodology behind them. Expert interviews are by no means simply just
“information gathering meetings” used primarily for collecting facts and
knowledge. To clarify any misunderstandings here: expert interviews are,
of course, not only a popular way of gathering information, they are also
a totally legitimate method for some forms of research. But as Gläser and
Laudel’s article in this book clearly shows, some basic methodological rules
still apply when conducting and evaluating such information gathering
expert interviews. However, the level of consideration that must be given
to the methodology increases proportionally when such interviews are not
6 Alexander Bogner, Beate Littig and Wolfgang Menz

intended primarily to establish a sound factual basis, but instead follow the
goal that lies at the heart of qualitative research: the reconstruction of latent
content of meaning. Expert interviews intended for this purpose – like all
other accepted methods of gathering data – require careful validation and
a solid theoretical basis. By now it should be quite clear that there is no
such thing as the expert interview. The spectrum ranges from quantitative
measures through to the use of experts as a form of information source (for
example as in Schmid, 1995; frequently also encountered in text books, for
example Lamnek, 2005) and the theoretically demanding, resolutely quali-
tative approach taken by Michael Meuser and Ulrike Nagel (1991, also in
this volume).
So it is with good reason that the contributing authors in this book also
seek plausible answers in their own specific contexts to the basic questions
that offer justification for the existence of expert interviews as an inde-
pendent research method: What constitutes an expert? What distinguishes
expert knowledge from other forms of knowledge? What are the different
types of expert knowledge? Which type of interview is best (for the actual
goal of the research and purpose of the expert interview)? What strategies
are available for analyzing the results (again in light of the form and func-
tion of the expert interview in the actual research design)?
All these questions have arisen since the dawn of the debate on the meth-
odology of expert interviews (cf. the article by Bogner and Menz, in this
volume), and we view the collection of articles included in this book as a
study and continuation of this methodological debate.
The book itself is divided into three parts. The theoretical or concep-
tual articles in the first part examine what lies behind the methodology
of expert interviews. Our aim with these articles is to offer a more precise
outline of the purpose and form of such interviews and examine what actu-
ally constitutes an expert interview. The key question that must first be
answered here is: What special characteristics does a person have to have to
constitute an “expert?” How is “expert knowledge” obtained? What types
of information and knowledge should be gathered? Can the expert inter-
view be justifiably singled out as a separate form of interview and clearly
differentiated from other forms of qualitative and quantitative interviews?
What is the difference between expert interviews and the elite interviews
encountered in English-speaking countries? And finally: What effect do
such considerations have on data gathering methods, interaction strategies
and the analysis of results?
Michael Meuser and Ulrike Nagel’s methodological concept of the expert
interview goes far beyond their earlier groundbreaking articles in the expert
interview debate (cf. Meuser and Nagel, 1991). This can be attributed not
least to their reformulation of the definition of the expert, which draws
on the reception of both sociology of knowledge and modernization the-
ory approaches. In this regard, Meuser and Nagel incorporate, in particular,
 An Introduction to a New Methodological Debate 7

current thinking on the changes in knowledge production, which can, in

part, be seen as an indication of a new relationship between science, politics
and the general public. This refers, in essence, to the emergence of a new type
of research that is characterized by its practical relevance, project-like nature
and transdisciplinarity, that is the inclusion of the knowledge spread across
a range of very different actors (“Mode 2”). These considerations prompted
Meuser and Nagel to extend their definition of the expert. Whereas their
previous publications restricted the circle of experts to members of the pro-
fessional functional elite, they now extend it in light of new (global) network-
like negotiation processes of knowledge production to include the people
actively involved in shaping public affairs. These include, for example, NGO
representatives who have (often) acquired their expertise outside their pro-
fessional role. In the course of their voluntary or professional activities, these
people have acquired specialized problem-solving and analytical knowledge
that is of relevance in expert interviews. As far as the analysis of expert inter-
views is concerned, Meuser and Nagel advocate a six-step process that (in
contrast to earlier concepts) should also examine the possible influence of
experience gained outside the professional realm.
In their article, Alexander Bogner and Wolfgang Menz contribute to shaping
the debate by differentiating between various forms of expert interview.
Their typology identifies three different types of expert interview each
intended for a different purpose: the exploratory expert interview used pri-
marily to provide orientation, the systematizing expert interview targeted
at the systematic retrieval of information and the theory-generating expert
interview aimed – in the spirit of qualitative social research – at reconstruct-
ing social interpretative patterns and subjective action orientation criteria.
They follow this by presenting a detailed definition of what constitutes an
expert. Based on a classification of various dimensions of expert knowledge
(technical knowledge, process knowledge, interpretive knowledge/“know-
why”), they propose a reformulation of the sociology of knowledge based
definition of the expert. This redefinition sees expert knowledge as an “ana-
lytical construction” and, at the same time, incorporates the expert’s “for-
mative power”: expert interviews are neither characterized by an interest
in limited special or specialized knowledge (as suggested in the sociology
of knowledge debate), nor can they be adequately defined by separating
the private sphere from the (generally occupational) functional context.
Experts are generally of research interest above all because they are in a
position to actually put their own interpretations into practice. Drawing on
this, Bogner and Menz call for an expert interview “interaction model” in
which the so-called interaction effects (normally considered as interfering
variables) are seen as constitutive and productive elements in the data pro-
duction process.
Michaela Pfadenhauer presents her theoretical support for the expert inter-
view as independent qualitative method from an ethnographic design
8 Alexander Bogner, Beate Littig and Wolfgang Menz

perspective. In her opinion, expert interviews are a particularly appropriate

method in research aimed at reconstructing explicit expert knowledge. She
also points out that an extraordinary level of prior knowledge of the subject
matter – obtained essentially through an ethnographic “inventory” of the
field of research – is required to guarantee their productiveness. Since the
expert’s impression of the interviewer influences the type of knowledge he/
she will communicate in the interview, relevant expert knowledge can only
be obtained through professional reference to the expert’s actual relevance
system. Pfadenhauer considers this specificity of the interaction, which
requires that the interviewer become a “quasi expert” to successfully carry
out an expert interview, as a central constitutive element of such interviews.
She also advises strongly against mistaking expert interviews to be a com-
paratively unproblematic and “economic” way of obtaining data or “short-
cut strategy.”
If we compare the predominantly German-language literature on expert
interviews with international articles on interviewing the elite as Beate Littig
does in her first article in this book, more commonalities than differences
are revealed. Similar issues are discussed in both, such as the problems of
gaining access to these groups and the specifics of interaction and interview-
ing. Although not identical, even the respective target group definitions
(experts and the elite) for such interviews overlap. This article discusses the
commonalities and differences in these two methodological approaches,
thereby contributing to a more detailed specification of the methodology of
expert interviews. It concludes with a sociology of knowledge based appeal
that the (professional) functional elite – given their positions of power – be
considered as a specific group of experts. From a methodological perspec-
tive and as a result of their specific interpretive knowledge (“know-why”)
and procedural knowledge (“know-how”), experts (and thus also the elite)
are of interest to social and political sciences research. Consequently, inter-
views with the elite aimed at generating explicit, tacit, professional or occu-
pational knowledge should be seen as expert interviews.
The second part of the book focuses on methodological practices and the
considerations that accompany them. What kind of data can be gathered
from expert interviews above and beyond that obtained from the customary
qualitative, guideline based individual interviews? How can the quality of
the data be guaranteed? What determines and characterizes communication
in expert interviews? Which personal skills, competences and attributes are
beneficial in this form of interaction? What role does “gender” play here?
How can the particular interaction structures be used to benefit the data
gathering process? And, last but not least: Which research ethics issues have
to be considered?
Jochen Gläser and Grit Laudel examine an issue hitherto largely ignored in
the expert interview debate, namely the “quality” of the interview partner
and the information and knowledge he/she provides in the course of the
 An Introduction to a New Methodological Debate 9

interview. In their opinion, neither the methodological debate nor, indeed,

sociological studies take into consideration the fact that different experts
with different levels of knowledge and different quality requirements in
their work can be expected to provide correspondingly different informa-
tion. Using their own research into the influence of institutional research
conditions on the production of scientific knowledge as an example, they
illustrate that, depending on the “quality” of the expert, not only can the
same social phenomena be described in the interview in different ways, but
it is also possible for diverging descriptions to conceal similar phenomena.
To reconstruct this information and draw benefit from it in the data gather-
ing and analysis processes, additional information above and beyond the
factual information provided in the interviews must be gathered on the
interview subjects. Gläser and Laudel advocate the systematic inclusion of
the “expert quality” issue in the individual research steps – from the selec-
tion of the experts through to the interviewing method and the analysis
of the results. In this respect, Gläser and Laudel’s article can be seen as an
impulse for an (overdue) debate on the validation criteria of expert inter-
views in general.
In their article, Gabriele Abels and Maria Behrens demonstrate the advan-
tages of analyzing gender-specific interaction effects in expert interviews
and putting them to productive use in the interview setting for the collec-
tion of practical and factual knowledge. In doing so, they make the assump-
tion that the person of both the expert and the researcher (in their case, a
female researcher) are also always present in the interview. Correspondingly,
the interaction is influenced by various non-circumventable subject-related
factors of influence. Abels and Behrens single out the category “gender”
and analyze its implications for interviewing experts and for the success of
expert interviews. They draw on their own empirical research to demon-
strate that different “interaction effects” in the interview can be attributed
to gender-related assumptions and prejudices. The results show – and this
is their central hypothesis – that, for all intents and purposes, both posi-
tive and discriminating effects can serve to generate productive data. At the
same time, the authors add two constraints: firstly that not all interaction
effects are of the same use in producing relevant data and, secondly that cer-
tain interaction effects can significantly impair the validity of the data. In a
second step, they verify their gender-related conclusions by means of a sec-
ondary analysis of their own interview notes from the associated research
projects. This analysis reveals that subjective postscripts relating to inter-
view situations can provide insightful material for a subsequent reflection
on one’s own self-reflexivity in the data production phase.
Gabriela B. Christmann discusses telephone-based interviews, a variation
on the expert interview theme in which the possibilities and limitations
are determined by technology. Although telephone interviews with experts
have long become established research practice (not least for economic
10 Alexander Bogner, Beate Littig and Wolfgang Menz

reasons), little reflection has been given to the methodology behind them.
Christmann refers in her methodological considerations to experience
drawn from a methodologically diverse German university project. Fourteen
of the interviewed experts in leading university positions were interviewed
in a face-to-face setting, while eight were interviewed by telephone for eco-
nomic reasons. Based on a comparison of the two approaches and a review
of the relevant methodological literature, Christmann’s assessment of
telephone-based expert interviews is sceptical. Even if the telephone inter-
views in her example research project did produce important information,
methodological concerns prevail. Telephone interviews are neither easier to
organize, nor is there any guarantee that the expert will be able to devote
his/her full attention to the interview: since there is no face-to-face contact,
the interviewer cannot predict or control distractions, lapses in concentra-
tion or interruptions by third parties. Reducing the interaction to a purely
linguistic level makes it more difficult to interpret, and the interviewee has
far less room for development – an aspect that perhaps carries less weight in
information gathering expert interviews than in those intended for recon-
structive social research theory building purposes. All in all, telephone-
based expert interviews prove a difficult and taxing undertaking – both for
the interviewee and for the interviewer.
Vaida Obelenė addresses the question of research ethics in the context of
the expert interview. In this context she discusses the extent to which the
propositions of the literature on democratic research practices are relevant
for an expert researcher. By drawing on her research experiences of study-
ing former communist functionaries, who established themselves in rela-
tion to new forms of knowledge and power in a post-communist society,
Obelenė reflects such practices in terms of choices that may undermine the
researcher’s purposes including his/her commitment to the ethicalness of
the study. Furthermore, this chapter aims to explore the tension between,
on the one hand, the need of assertiveness on the part of the researcher in
defending the study’s purpose vis-à-vis the powerful expert, and the need of
the researcher’s sensitivity to the interests and vulnerabilities of the expert
on the other. Against this background expert interviews can be understood
as a form of ‘bargained research’ where the interests of both parties have to
be considered.
The third part of the book contains a selection of articles that deal with
the importance of expert interviews, the way they are conducted and the
particular specifics of interaction in such interviews in concrete fields of
application and social science sub-disciplines (industrial sociology, inter-
pretative organizational research, labour market research and technology
foresight). One particular question comes to the fore here, namely the meth-
odological consequences that result from the structures peculiar to each
respective field of research and their consequences for the success of the
interviewing techniques used in an expert interview setting.
 An Introduction to a New Methodological Debate 11

Rainer Trinczek uses an industrial sociology case study to illustrate that

such interviews must by no means generally – as is a common preconception
in qualitative research – adhere to the principles of neutrality and restraint.
Although appropriate interviewing strategies can be found in methodo-
logical suppositions, the actual structure of such an interview has to follow
the rules of normal communication, interpreting these everyday communi-
cation rules in line with the actual situation and the purpose of the inter-
view. In the case study analysed by Trinczek (interviews with managers), the
rules of communication and, thus, the expectations of the interviewees, are
based on their everyday work situation; in general, the interviewees expect
the interview to follow the question and answer structure predominantly
encountered in everyday work situations. However, the researcher also has
to take the actual subject matter into consideration when deciding whether
to orient the entire interview on these expectations. Trinczek illustrates this
using two thematically heterogeneous research projects as examples. An
argumentative/discursive interview approach is suitable if the subject mat-
ter addresses a work-related topic. But the situation is different in research
projects that deal with the “private world,” where (and in line with the vari-
ous communication structures encountered in everyday life) a narrative-
based interview structure is appropriate.
In their article, Manfred Lueger and Ulrike Froschauer illustrate the import-
ance of expert interviews in interpretative social research based organiza-
tional analysis. By distinguishing different levels of observation (first and
second order), they propose the reconstruction of analytically different types
of knowledge based on a heuristic of distinct arenas and expertise reflection
levels. The interviewees (as experts in the organizational lifeworld) have
internal organizational experience and know-how. From an expanded obser-
vation perspective, they are in a position to provide qualified information
on internal knowledge structures and constructions. To ensure the different
knowledge forms are contrasted with appropriate complexity, those persons
whose professional profile qualifies them as relevant internal or external
from the point of view of the actual research should ultimately be integrated
into the research. Pursuant to their analytical perspective, Froschauer and
Lueger illustrate the specific individual data gathering and interpretation
requirements raised by the different types of contrasting expertise in organ-
izational action.
Andrea Leitner and Angela Wroblewski deal in their article with the stand-
ing and specifics of expert interviews in an evaluation research context.
They begin by providing an overview of the development of this branch of
research and then go on to examine the possibilities and prerequisites of
the use of expert interviews in a “responsive evaluation” approach. They
consider the role of experts as “stakeholders” with a specific interest in the
results of the evaluation to be a central characteristic of expert interviews
aimed at evaluating socio-political measures. According to Leitner and
12 Alexander Bogner, Beate Littig and Wolfgang Menz

Wroblewski, this “stakeholder” problem needs to be assessed from a meth-

odology perspective. In the approach they describe, this validity problem is
counteracted by references to information from other data sources during
the interview. Examples from labour market policy evaluation research are
used to present the determining factors in the expert interview interaction
process and discuss the possibilities available for dealing with this bias.
The Expert Delphi method presented by Georg Aichholzer plays an import-
ant role in the rapidly growing field of technology foresight. Against the
background of some serious uncertainties in modern societies, this rela-
tively tightly structured group communication process basically aims to
“rationalize the future” through a methodologically controlled generation
of expert knowledge. In many cases, foresight processes also aspire to offer
social networking, voting and consensus building functions that improve
the performance of innovation systems. Aichholzer’s article looks at the
methodology modifications and combinations that have accompanied the
growing use of the Expert Delphi and illustrates them using international
examples. Following an introduction to the basic elements of the Delphi
method, Aichholzer explains its use in the innovative Austrian Technology
Delphi, which incorporates a number of modifications to the classic Delphi.
The different strategies used to capture expert knowledge in a further
example – an international foresight process on the future of European
transport systems – allows an interesting comparison between the Expert
Delphi and cross-impact analysis. The subsequent applications discussed in
the article, such as a Finnish approach geared towards balancing consensus
and diversity and the use of internet assistant Expert Delphis, demonstrate
other application-specific adaptations to and combinations of this method.
This book should not, of course, be seen as an exclusive compendium
and the final word on a fully sanctioned research method. Far more, it is an
invitation to others to reflect on and examine the different forms of expert
interview in more detail. Indeed, we expect and hope that an intensified,
critical debate on this topic will further increase the practical benefits of
this method. The internationalization of the debate is an important step
in this direction. Consequently, our intention with this book is to build a
stable platform on which “experts” from different research traditions, sci-
entific and language cultures can participate in an intensified and fertile
debate on expert interviews. This would ultimately benefit not only a small
community of expert researchers, but also qualitative social research as a
whole.

Note
1. For a long time even representatives of the qualitative paradigm were undecided
as to whether expert interviews actually did represent a discrete method of data
collection that could be differentiated from other interview forms. No reference is
 An Introduction to a New Methodological Debate 13

made, for example, to expert interviews in “A Companion to Qualitative Research”

(Flick and others, 2004), the English edition of a key German handbook.

References
Beck, U. (1992) Risk Society. Towards a New Modernity. (London: Sage Publications).
Bogner, A. (2005) “How Experts Draw Boundaries. Dealing with Non-Knowledge
and Uncertainty in Prenatal Testing” in Science, Technology and Innovation Studies 1 ,
pp. 17–37, <www.sti-studies.de/articles/2005–01/bogner.htm>
Bogner, A., Torgersen, H. (eds) (2005) Wozu Experten? Ambivalenzen der Beziehung von
Wissenschaft und Politik. (Wiesbaden: Verlag für Sozialwissenschaften).
Collins, H. M., Evans, R. (2002) “The Third Wave of Science Studies: Studies of
Expertise and Experience” in Social Studies of Science 32 , pp. 235–96.
Collins, H. M., Evans, R. (2007) Rethinking Expertise (Chicago and London: University
of Chicago Press).
Flick, U. (2006) Introduction to Qualitative Research, 3rd edn (London: Sage Publications).
Flick, U., Kardorff, E. von and Steinke, I. (eds) (2004) A Companion to Qualitative
Research (London and Thousand Oaks/CA: Sage Publications).
Giddens, A. (1990) The Consequences of Modernity (Cambridge: Polity Press).
Giddens, A. (1991) Modernity and Self-Identity – Self and Society in the Late Modern Age
(Stanford/CA: Stanford University Press).
Jasanoff, S., Markle, G. E., Petersen, J. C. and Pinch, T. (eds) (1995) Handbook of Science
and Technology Studies (Thousand Oaks/CA: Sage Publications).
Lamnek, S. (2005) Qualitative Sozialforschung: Lehrbuch, 4th edn (Weinheim: Beltz).
Maasen, S. and Weingart, P. (eds) (2005) Democratization of Expertise? Exploring Novel
Forms of Scientific Advice in Political Decision-Making (Dordrecht: Springer).
Meuser, M. and Nagel, U. (1991) “ExpertInneninterviews – vielfach erprobt, wenig
bedacht. Ein Beitrag zur qualitativen Methodendiskussion” in Garz, D. and
Kraimer, K. (eds) Qualitativ-empirische Sozialforschung. Konzepte, Methoden, Analysen
(Opladen: Westdeutscher Verlag), pp. 441–71.
Przyborski, A. and Wohlrab-Sahr, M. (2008) Qualitative Sozialforschung. Ein Arbeitsbuch
(München: Oldenbourg).
Schmid, J. (1995) “Expertenbefragung und Informationsgespräch in der
Parteienforschung: Wie föderalistisch ist die CDU?” in Alemann, U. v. (ed.)
Politikwissenschaftliche Methoden. Grundriss für Studium und Forschung (Opladen:
Westdeutscher Verlag), pp. 293–325.
Turner, S. (2002) “What is the Problem with Experts?” in Social Studies of Science 31 ,
pp. 123–49.
This page intentionally left blank
Part I
Theoretical Concepts:
Methodology of Expert Interviews
This page intentionally left blank
1
The Expert Interview and
Changes in Knowledge Production
Michael Meuser and Ulrike Nagel

1.1 Introduction

The expert interview as a method of qualitative empirical research, designed

to explore expert knowledge, has been developed considerably since the
early 1990s. A number of readers has been published1 and thus a gap in the
methods’ literature has been dealt with, much to the benefit of many dis-
ciplines and fields of research in the social sciences. It can be assumed that
through increased reflection on methodical issues research into experts’
knowledge has gained in professionalism and quality.2
At the same time, the expert interview has become very popular as a
“streamlined” method (Bogner and Menz, 2002a, pp. 9–10, see also Bogner,
Littig and Menz, in this volume): recruiting informants does not seem to
cause any difficulties, as a method the expert interview appears to be “quick,
easy, and safe” in its application, and it promises to be of good practical value.
However, the authors continue, it is easily overlooked that the expert inter-
view is an ambitious method which cannot be considered to be on a sound
footing either in terms of modernization theory or methodologically.3
In this article we shall deal with the notion of expert and expert knowledge
as outlined in the sociology of knowledge, reflecting on expert knowledge
as embedded in processes of modernization, and hence in the light of mod-
ernization theory. We shall argue that expert knowledge is a knowledge sui
generis with its own characteristic traits necessitating a particular methodo-
logical approach. We shall outline the distinctive features of this approach
taking a closer look both at data collection and interpretation of data.4
Furthermore, we shall discuss the concept of knowledge in the context of
changes in the production of knowledge or rather new forms of knowledge
production as they are being reflected in the debates on post-modernity
and the Knowledge Society. Set forth by the regime of accumulation inher-
ent in globalization, modern societies are facing changing conditions of
the production of knowledge in general and more so of knowledge-based
expert systems. Following Welsch (1993), these conditions can be described

17
18 Michael Meuser and Ulrike Nagel

as constituting a basic plurality, a plurality which cannot simply be dealt

with as an intensified coexistence of spheres of knowledge and points of
view of experts, but – besides the diversification of knowledge – must also
be understood as a trend towards transgressing the borderlines between
spheres of knowledge. Some of these characteristics of societal change shall
be discussed with a view to their impact on the expert and expert systems,
highlighting the difficulties and dilemmas imposed on the expert’s action
and interaction processes. Against the background of modernization theory
and a diagnosis of our time we shall ask if and to what extent the changing
forms of knowledge production will have to be taken into account when
analysing expert knowledge, and to what new foci the researcher’s attention
is drawn.5

1.2 Expert, expert knowledge and societal change

1.2.1 Sociology of knowledge as frame of reference

In a pragmatic perspective – focusing on the local context of knowledge
production, the status of expert could be understood as ascribed by the
researcher: a person is attributed as expert by virtue of his role as informant
(Walter, 1994, p. 271). Who is identified as expert and who not depends on
the researcher’s judgement. However, this definition remains insufficient
since it does not provide the researcher with criteria to distinguish between
experts and non-experts. The term “expert” might end up being used infla-
tionarily and finally anybody might be seen as an expert – at least as an
“expert of her or his own life.” Following this line the expert interview
would no longer be distinguishable from other techniques of interviewing,
like, for example, the narrative or ethnographic interview.6
A definition of the term expert can best be arrived at by reviewing the fea-
tures distinguishing expert knowledge from other forms of knowledge like
everyday knowledge and common-sense knowledge. In scientific research
an individual is addressed as an expert because the researcher assumes – for
whatever reason – that she or he has knowledge, which she or he may not
necessarily possess alone, but which is not accessible to anybody in the field
of action under study. It is this advantage of knowledge which the expert
interview is designed to discover, and it is an exclusive realm of knowledge
which is highly potential because and in as far as it is linked with the power
of defining the situation. What comes into sight when we combine the prag-
matic definition of the expert interview with the sociology of knowledge-
perspective, is the distinction between expert and lay person.
It is the researcher who according to his research objective decides who
she or he wants to interview as an expert; but we have to add that this is
not an arbitrary choice but is related to the recognition of an expert as
expert within his own field of action. At the same time not every person
recognized as an expert in a particular setting is necessarily addressed as a
 Experts and Changes in Knowledge Production 19

potential informant. We would like to conclude that a person is considered

an expert if she or he possesses an “institutionalized authority to construct
reality” (Hitzler, Honer and Maeder, 1994). Expert knowledge is character-
ized by the chance “to become hegemonial in a certain organizational and
functional context within a field of practice” and, thus, “to be influential
in structuring the conditions of action for other actors [....] in a relevant
way” (Bogner and Menz, 2002b, p. 46, our translation, cf. also Liebold and
Trinczek, 2002, p. 36).
This definition starts from the distinction between three ideal types of
knowledge given by Alfred Schütz (1964): the “expert,” the “man on the
street,” and the “well-informed citizen.” The distinguishing feature for
Schütz is the extent of the person’s “readiness to take things for granted”
(p. 123). While the “man on the street” “lives ... naively in his own and his
in-group’s intrinsic relevances” (p. 129), the expert is at home in a system
of relevances “imposed ... by the problems pre-established within his field”
(p. 130). These relevances to him, however, are not like a fate he blindly
resigns himself to, “but by his decision to become an expert he has accepted
the relevances imposed within his field as the intrinsic, and the only intrin-
sic, relevances of his acting and thinking” (p. 130).
The discussion within the sociology of knowledge referring to Schütz
mainly focuses on the distinction between expert and layperson. This
distinction is based on the specialized knowledge of the expert. Not every
special knowledge, however, already is expert knowledge, as Sprondel
(1979, p. 141) has shown, but only that one which can be grasped as
a “socially institutionalized expertise.” In a society where division of
labour is an organizing principle, expert knowledge, Sprondel continues,
is “special knowledge considered necessary” (p. 148) in relation to prob-
lems, which are defined as special problems. Sprondel views this know-
ledge as linked to the role of the professional. Defining the term expert
with reference to the professions does make sense in so far as expertise –
seen historically – has become differentiated in the context of the profes-
sions, and until today and to a large extent is basically acquired on this
basis.
Yet more recent analyses of societal change convincingly show that new
forms of knowledge production have developed. It is argued that these new
forms of knowledge production do not replace the traditional specific rele-
vances established within the various fields of action (disciplines, profes-
sions, and so on) but, rather, complement them. It seems that with these
new forms of knowledge production the link between expert knowledge and
professional role is loosened, the sharp distinction between expert and lay
person weakened, and the professional’s claim of exclusiveness for the rele-
vances of her or his discipline, is fading. In the following chapter we will
deal with these processes as implying some more characteristics of expert
knowledge.
20 Michael Meuser and Ulrike Nagel

1.2.2 Societal change and new forms of knowledge production

The definition of expert and expert knowledge proposed by Schütz and
Sprondel is an apt representation of the form of knowledge production
dominant in modernity. In modernization theory, this form is referred to
as Mode 1 (Gibbons and others, 1994). In Mode 1, knowledge is generated
according to the cognitive and social norms of a disciplinary context and, in
so far, is autonomous (Gibbons and others, 1994, p. 1). These norms are the
relevances “imposed” upon the expert, and it is them – and them alone –
which, following Schütz, determine her or his behaviour. In the expert’s
practice, these norms can and have to be interpreted in such a way that
the requirements of situation and context are taken into account in each
case. Mode 1 serves as a point of reference for the analysis of changes in
knowledge production towards Mode 2. In this process the characteristics
of Mode 1 – for example, disciplinarity and the distinction between expert
and lay person – lose their discriminating power, and knowledge produc-
tion is observed as taking place within wider, transdisciplinary contexts
(ibid.). It is commonly agreed that with the emergence of Mode 2, the form
of knowledge production characteristic of modernity, Mode 1, has not dis-
appeared, but “has continued, come to a head and transformed itself. It has
hardly reached its own objectives; its aporia have become even clearer, if
not more incontestable, and this is exactly why in dealing with modernity
shifts of perspective can be seen to be emerging” (Bonss, 1998, p. 969).
In a historical perspective, the beginnings of the changes from Mode 1 to
new forms of knowledge production – as in Mode 2 – can be located in the
1960s. This decade reflects the onset of a process in which the problematic,
ambivalent consequences of modernization of industrial society and of the
belief in the logics of progress (as defined by functional differentiation, sci-
entification, and disciplinarity) are clearly becoming visible. The logics of
progress are increasingly criticized, accompanied by a growing scepticism
towards progress, particularly towards solutions in the fields of science/
technology and social engineering. The indisputability of the ambivalent
side-effects of modernization in the 1970s, as exemplified by the problems
related to the civil use of atomic energy, speeds up the establishment of a
reflexive attitude towards the plausibility structures of modernity, and in
the 1980s this leads to novel – globalized – perceptions of problems: to the
emergence of new authorities of interpretation, new systems of knowledge,
new concepts of production, new ways of life and so on, the situation being
difficult to read and the future development still concealed, Habermas
(1985) coins it as a new “Unübersichtlichkeit” (concealment).
There are two ways of looking at societal change crystallizing from this
process. On the one hand, there is the discourse on risks which addresses
the risks and uncertainties of contemporary society (Beck, 1992; Beck,
Giddens and Lash, 1994), and “which is centered on how to control the new
side-effects and the power of the side-effects” (Bonss, 1998, p. 973). On the
 Experts and Changes in Knowledge Production 21

other hand, expressing the ambivalences of modernity (Baumann, 1991)

the constructionist and postmodern discourse is taking shape emphasizing
the pluralization of knowledge and reason gone plural (Welsch, 1993).
In this context of complex and concealed problems of societal change expert
knowledge is confronted with the following dilemma: on the one hand, due
to an increasing demand of expert knowledge and growing expectations for
interpretation and orientation, expert knowledge and expert systems multi-
ply and proliferate. On the other hand, however, being the source of the risks
claimed to be under control, the “conservative” – Mode 1 – scientific expert
knowledge itself slides into a legitimation crisis. Various trends mixing
together are making up a paradoxical situation: the questioning of Mode1-
knowledge production; the emergence of innovative trandsdisciplinary
arrangements of knowledge production in large parts of society; and the
development of new authorities of interpretation with their new relevances,
network-based communication cultures, and integrated forms of knowledge
production which take into account local knowledge and knowledge of lay
people, thereby establishing a world of counter- experts, counter-expertise,
and alternative publics.7 In the course of time the latter do not lose their
impact but undergo a process of institution building of their own, and very
often their personal is found in positions of power. This can be illustrated
by the new social movements, NGOs, and networks of civil society – whose
members when advancing in years tend to take over positions in the polit-
ical, social, or cultural system, that is in the institutions of society.
It is of interest for our discussion that in the course of modernity becom-
ing reflexive, and due to the criticism of the expertocrization of society,
expert knowledge is also generated outside professional contexts. But it
should not be mistaken with lay people’s knowledge or the knowledge of
the well-informed citizen, but has to be understood as linking up with sci-
entific expertise, thus developing hybrids and producing knowledge differ-
ently from Mode 1; it is generated in pluralized networks, negotiated by
experts and publics of lay people, and applied to concrete societal prob-
lems (and possibly organized as counter-power). As it is, the criticism of the
expertocrization of modernity itself and of its side-effects is seen to be ini-
tiating the development of integrated counter-expert systems which com-
bine non-scientific and scientific knowledge, different knowledge cultures,
counter-publics, and alternative structures of participation.
The emergence of new knowledge systems and knowledge cultures; of
uncommon commentators of deficits of modernization and responses to
problems; of new (collective) actors and publics and their network-based
negotiations in the process of knowledge production, has consequences:
The superior problem-solving rationale (Mode 1) as claimed by the scientific
system is exposed to systematic doubt and put into competition with the
expertise originating from heterogeneous knowledge systems and spheres
of interest. The latter themselves, of course, are science-based, but instead
22 Michael Meuser and Ulrike Nagel

of being disciplinary they are transdisciplinary and heterogeneous with

regard to the actors and their relevances, that is incorporating also local
knowledge, experiential knowledge, viewpoints of lay people and others
concerned. In marking their belongingness to an alternative culture some
of these knowledge systems are surfacing even a new type of aesthetic,
new symbols (sneakers, rainbow, sunflower). The new mode of knowledge
production by negotiating relevances in pluralistic and transdisciplinary
networks is more and more acknowledged as “socially acceptable.” What
is more, while problems and imperatives of innovation are globalized and
novel in many cases – expert knowledge as created in the plural worlds of
counter-experts turns into a demanded source. Mode 1-knowledge produc-
tion does not become outdated. It is, as is stated time and again, indispens-
able as to some of its characteristics, but it is complemented by Mode 2.

Mode 1 problems are set and solved in a context governed by the, largely
academic, interests of a specific community. By contrast, Mode 2 know-
ledge is carried out in a context of application. Mode 1 is diciplinary
while Mode 2 is transdiciplinary. Mode 1 is characterised by homogen-
ity, Mode 2 by heterogenity. Originally, Mode 1 is hierarchical and tends
to preserve its form, while Mode 2 is more heterarchical and transient.
Each employs a different kind of quality control. In comparison with
Mode 1, Mode 2 is more socially accountable and reflexive. It includes a
wider, more temporary and heterogeneous set of practitioners, collabor-
ating on a problem defined in a specific and localized context. (Gibbons
and others, 1994, p. 3)

The universities, too, are affected by these developments. Nowotny (1997)

states that with the changes in knowledge production the universities lose
their privileged status as legitimate sites of the production of knowledge.
The increasing number of academically trained individuals can be held
responsible for this process since it is their research competence, which
is influencing a variety of institutions. It is stated that this, on the other
hand, was having repercussions on the “largely disciplinary organized sys-
tem of knowledge production.” In the case of Mode 2 knowledge production
already “the definition of the scientific problem” were “carried out in the
context of concrete and therefore changing applications.” Mode 2 was open
to societal demands and expectations. Disciplinarity as a characteristic fea-
ture of Mode 1-knowledge production was more and more complemented
and also partially replaced by transdicipinarity, that is an integration not
only of other scientific disciplines, but of institutions and organizations
beyond the field of the sciences as well (cf. Gibbons and others, 1994). As
to further development of Mode 2-theory, this aspect of opening-up the
process of knowledge production for new circles of experts from the public
realm, of “the closer and closer interaction of science and society,” becomes
 Experts and Changes in Knowledge Production 23

a central issue and is discussed as an “indicator of the emergence of a new

kind of science” (Nowotny, Scott and Gibbons, 2004, p. 7).
The new knowledge regime is interpreted by Rammert (2003, p. 483)
as “the regime of heterogeneity in the distribution of knowledge production”
(emphasis in original). According to Rammert, knowledge production is car-
ried out in a “fragmentary” way. Knorr-Cetina (2002, pp. 12–13, 18) speaks
of “epistemic cultures” which she conceives of as a structural feature of
Knowledge Societies, and which, fragmented as they are, can no longer be
grouped in a homogeneous model of science. Rammert, quite like Nowotny,
states an extension of the circle of producers of “legitimate” knowledge.
“The quality (of knowledge; MM/UN) is no longer assessed by colleagues
within a discipline, the ‘peers’ only but by heterogeneous groups of experts,
a mixture of epistemic cultures” (Rammert, 2003, p. 488). Rammert asserts
that also non-scientific, political and moral, aspects play a role here, and
that alternative groups such as environmental activists, NGOs, and for
instance the “Chaos Computer Club” are being integrated into the process
of knowledge production.8

1.2.3 Relevances of expert knowledge revisited

Even though the analysis of the changes in knowledge production towards
Mode 2 is not at all undisputed (Bender, 2001; Hettlage, 2004), it is never-
theless supported by the debate to a great extent. We assume that today
expertise both within and outside the professions is constituted in a more
complex way than described in “classical” works of the sociology of know-
ledge. In defining the term expert this recent development should be taken
into account, much to the benefit of the method of expert interview as
well. What is needed is a definition of the status of expert going beyond
the traditional understanding of the professional role, and linking it with a
widened perspective on the “relevances imposed upon” the expert. At issue
here are new forms of knowledge production and their conditions inside
and outside the professions, as well as the newly emerging relevances which
have to be taken into account as important aspects while analysing expert
knowledge. The new conditions of knowledge production reflect the intri-
cate demands impinged upon processes of defining and solving problems
by the late-modern Complex Societies (Schütze, 2002), Knowledge Societies
and the Global Risk Society (Beck, 1998), conditions, which transform the
ambivalences of modernity (Baumann, 1991).
Gibbons and others (1994) and other Mode 2-theorists have derived their
hypotheses from the analysis of new, pluralistic ways of knowledge produc-
tion in cooperation between different spheres of society (science and its
disciplines, industry, politics, public administration). In comparison with
this approach our starting point is the sociology of knowledge-perspective
which we have reviewed in the light of the constructionist-postmodernist
or Knowledge Society brand of modernization on the one hand, and with
24 Michael Meuser and Ulrike Nagel

a view to the process of differentiation of extra-professional fields of expert

knowledge as illustrated by the new social movements and publics on the
other hand. It is not surprising that both approaches, even though within
different language games, lead to quite similar findings regarding the trans-
formation of expert knowledge. It must be pointed out again, however,
that the Mode 2-phenomenon is not at all unanimously confirmed by the
debate on modernization; much to the contrary, empirical researchers have
expressed considerable doubt about the mode of transformation as claimed
by Gibbons and others.9
Regarding the analysis of expert knowledge it has to be ascertained that
not only professional knowledge is treated as expert knowledge. We are
following Gordon (1975, p. 199) who – referring to research on local politics –
understands experts (“special respondents”) as individuals “who are active in
community affairs regardless of their position in the social status system.”
Active participants in community affairs, for example, are members of
citizens’ groups, relief organizations, and self-help groups, as well as volun-
teers in welfare, social work, and similar fields. The activities of these types
of actors are not restricted to local (municipal or regional) contexts. Some of
them are active at the global level, for example within the scope of NGOs. All
of these actors acquire a special knowledge through their activity – and not
necessarily through their training – because they have privileged access to
information. Their expertise, too, is socially institutionalized and linked to a
specific context and its functional requirements, even though in a different
way from the expertise grounded on the professional role.10
The decoupling of the term expert from the professional role does not imply
an inflationary extension of the circle of individuals eligible for an expert
interview. The frequently used phrase “expert on one’s own behalf” is out of
place here. Of those who have something to report on account of personal
observation and experience, not each and every individual is to be interviewed
as an expert. Schmid-Urban and others (1992, p. 85) rank among experts, for
example, also pub-owners and kiosk-operators, since these “often are in touch
with certain groups.” They are informants who, in our view, can definitely
contribute important background information, but do not give an expert
opinion. Such individuals do not meet the criterion of active participation.
They have not acquired their knowledge of a particular problem (for example
housing shortage, poverty, health risks) through an activity which is aimed
at the problem and, therefore, with a view to analyzing and/or helping to
solve the problem in any way. The contacts and observations from which they
derive their knowledge are embedded in structures of relevance, which are not
focused on a problem but, for example, determined by economic motives. The
definition of experts as active participants emphasizes the specific functions
such individuals have with regard to problems – whether by virtue of a profes-
sional role, or as a volunteer. Special knowledge acquired through carrying out
such functions is the subject matter of the expert interview.
 Experts and Changes in Knowledge Production 25

It is reasonable to decouple the term expert from holding a formal pos-

ition in a hierarchy of occupational status positions and to argue, accord-
ingly, that the expert interview shall be used also with interviewees who
are active participants. While integrating the active participants into the
group of potential experts we are avoiding the risk involved in a narrow
definition of the term expert – the risk that “professionalisation of a solu-
tion for a problem [is] connected with a momentous narrowing of a problem
definition” (Sprondel, 1979, p. 143). Certainly, also an expertise issued by a
citizens’ group or an NGO is knowledge- and science-based, however, it is a
different view, different with regard to the determinative relevances as com-
pared to those of a head of a public authority or a minister of state.11
Viewing the status of the expert and the extra-professional social settings
of production of expertise from the wider perspective just outlined, and tak-
ing into account the reflections on Mode 2-knowledge production (Gibbons
and others, 1994), a more differentiated account of relevant aspects of expert
knowledge comes into sight. In the following, we shall take a closer look at
some aspects we believe to be particularly important for the data analysis of
expert interviews.

(a) Socio-cultural conditions of the production of expert knowledge: In the tran-

sition to a pluralized, heterogeneous mode of knowledge production
expert knowledge is sensitized regarding stocks of knowledge generated
outside the scientific world and acquired, experienced (and suffered) in
extra-professional practice. A co-mingling between expert knowledge
of active participants and that of professional-scientific experts takes
place, resulting in the formation of hybrids between formerly separated
fields of knowledge and symbolic orders; in transgression of boundar-
ies between periphery and centre, everyday experience and systematic
knowledge, local and global expertise.12 Under these circumstances
experts become sensitized for the context within which their know-
ledge has to prove itself as useful (Gibbons and others, 1994; Nowotny
and others, 2004), while the active participants’ advantage is the posses-
sion of local knowledge.
Accordingly, while exploring expert knowledge with the tool of
the expert interview, the socio-cultural conditions of the production
of knowledge become relevant topics of data collection and analysis.
Attention is drawn to the embeddedness of the expert in circumstances
and milieus; to the heterogeneity of relevant others; to membership in
global communities and local networks; to arenas and circles the expert
is involved in and orientation is derived from (Schütze, 2002). While
following this lead we are exploring if and to what extent the expert
is open to mixing different knowledge bases in the sense of transdis-
ciplinarity and hybridity – to a mingling between the perspectives of
experts, laymen, and active participants –; whether she or he has an
26 Michael Meuser and Ulrike Nagel

open awareness context with regard to other knowledge orders and

a readiness to overcome an “imperialistic” expert position (Welsch,
1993); altogether we are following up the question of plurality, transdis-
ciplinarity, and reflexivity of the production of expert knowledge and
the definitions it provides. The focus of analysis, therefore, is directed
towards the modes and mechanisms of producing knowledge, that is
to an understanding of the process, by which knowledge is generated,
checked, communicated, reflected upon, modified, and finalized. The
socio-cultural conditions of knowledge production are reviewed as a
core dimension of expert knowledge as much as of expert interview-
ing. We suggest they should be taken into account at all phases of the
research process starting from designing the interview schedule, con-
ducting the interview, and analyzing the data.13 As to methodology, the
analysis of expert knowledge thus yields to a process-oriented interpret-
ive approach.
At this point we are going to revise an earlier standpoint regarding
the relevances of the expert. We were then arguing that the expert’s
relevances were defined exclusively by the responsibilities attached to his
or her position and function in the field of action under study, that is
by these relevances alone (Meuser and Nagel, 1991). With regard to the
expert’s relevancies we are taking on a different view now. The produc-
tion of expert knowledge being influenced more and more by a plur-
ality of different spheres – including knowledge acquired beyond and
independent of the scientific-professional and economic spheres –, it
has to be assured that the responsibilities connected with the expert’s
position are not the only ones to be focused on but also the relevances
derived from other realms of experience, p.e. personal life experience
as a private person. This reflects a point at issue in the discussion about
the expert interview, the question of whether and in which way the
biographical motivation and milieu-specific embeddedness of know-
ledge (-acquisition) should be a subject matter of the interview. In an
earlier article (Meuser and Nagel, 1991) we suggested to ignore the
private aspects of expert behaviour. Bogner and Menz (2002b, p. 44),
however, are in favour of a “methodical integration of the expert as ‘pri-
vate person’.” The authors argue that there would hardly be a chance
in distinguishing between an interviewee as expert and as private per-
son. As it were only the analysis of the case would show the extent to
which “private” relevances were important for the reconstruction of the
expert’s knowledge and behaviour.
We find this a convincing statement. In addition, a revision of our
earlier position is suggested by more recent research in which we used
the method of the expert interview. The private person’s relevances,
however, are examined to see how they affect those relevances, which
are of primary importance for the expert’s position and responsibilities.
 Experts and Changes in Knowledge Production 27

To give an example, in a study on dual career couples (Behnke and

Meuser, 2003, 2005) human resources managers of large companies
were interviewed as experts; interviewees who showed a commitment to
improve the work- life- balance also reported some concern about their
highly qualified daughters who, unlike the sons, had great difficulties
in finding employment appropriate to the level of their qualification.
This shows how experiences made in the private sphere influence the
perception and, possibly, the handling of work-related responsibilities.
This kind of impact should definitely be thematized in an expert inter-
view, at the same time the expert interview is not be confused with a
biographical interview. Biographical structures and processes being the
research objective – like the development of an expert’s knowledge base,
the dynamics of a career, and changes of identity – a narrative biograph-
ical interview should be chosen as a tool of data collection and analysis.
On the other hand, in exploring expert knowledge, the focus remains
on the institutional framework within which the expert moves and on
the individual actor involved, her or his position and responsibilities
within a particular context.

(b) Negotiating expert knowledge: practices of communication and organization:

With the development of new socio-cultural forms of knowledge pro-
duction further relevances of expert knowledge and for its analysis are
surfacing. Since knowledge increasingly emerges by way of negotiation
within plural, heterogeneous discourse communities, and networks of
experts, the patterns and practices of expert communication, of bar-
gaining over definitions and solutions are gaining importance. They
are of interest as a collective practice across institutional, professional,
and entrepreneurial boundaries – as well as the practice of organizing
such negotiation processes. In order to discover the typical traits of an
expert’s body of knowledge, attention was directed to discourse and
networking practices, and in particular to the patterns of establishing
consensus as well as differences between opinions, even more so to prac-
tices of working on minimum consensus, to negotiating procedures, cri-
teria of participation, and strategies of inclusion and exclusion.
Determining negotiation and networking as basic characteristics of
expert knowledge raises the question whether the analysis must yield to
an analysis of epistemic cultures – as is suggested by Knorr-Cetina (2002,
p. 12) or alternatively by Schütze (2002), the latter favouring the con-
cept of social worlds and arenas. According to Knorr-Cetina, Knowledge
Society as structured by epistemic cultures is adequately comprehended
by way of exploring the practices of the different knowledge cultures;
and she further remarks that in the context of Knowledge Society the
inside worlds of knowledge are becoming ever more crucial. Accordingly,
and in contrast to older studies focusing on disciplines and specialized
28 Michael Meuser and Ulrike Nagel

fields, the analysis of epistemic cultures is concerned with the proced-

ural rationality and the epistemic practice within fields of knowledge
and thus with the complex life-worlds at the insides of modern institu-
tions. As a result, the subject matter of research has changed from the
construction of knowledge towards the “construction of the machineries
by which knowledge is being constructed” (Knorr-Cetina, 2002, p. 13).14
This raises the question of the relation between the expert interview
and ethnographic methods. While stating that the expert interview
plays an important role as an independent method of empirical social
research, it can yet be agreed that it is as much open to an ethnographic
research design as are other tools of interpretive-qualitative research, for
example participant observation. Regarding our concept of the expert
interview, it should be added that a basic ethnographic attitude would
prove very useful.

(c) Expert knowledge as construction: Recognizing expert knowledge as a col-

lective project of producing knowledge by way of negotiation, cooper-
ation, networking and teamwork, it has to be acknowledge that the
production of knowledge is to be defined as an open ended process.
Throughout the actual making of the expertise, the outcome regard-
ing the definition of what is the case and the respective problem solving
strategies remains an open question, the productive process itself must
be seen as bearing an uncertainty regarding its result. In addition, the
practices guiding the production of expertise need to be conceived of
as being carried out with a view to options “in spe,” so to say, that is
options not yet to be apprehended but expected to materialize in future.
This practice is to be opposed to a mode of decision-making marked by
neglect of ambivalences and uncertainties, the expert relying on her or
his authority, freedom of judgement and power to define situations.15
Yet the new-mode-experts as much as their predecessors in Mode 2
are not likely to be acting in a social vacuum devoid of power struc-
tures. Their settings, too, are allocating insider and outsider positions,
as can be illustrated with regard to the definition of rules of conduct
and criteria of validity. In accordance with Foucault’s approach to the
knowledge-power-relationship and with the discourse analysis devel-
oped from his work, expert discourses are to be understood as orderly
practices of production of meaning patterns, particularly busy drawing
the line between favourable and unfavourable patterns and legitimizing
and delegitimizing potential speakers (Keller, 2005).
It has to be assured that a negotiative model of expertise does not
imply that procedures and results are arbitrary. An impressive example
is given by the Intergovernmental Panel on Climate Change. The
panel’s “Summary for Policymakers: Climate Change 2007: The Physical
Science Basis” (www.ipcc.ch), gives an insight into the production of
 Experts and Changes in Knowledge Production 29

knowledge by a worldwide network of experts. The network’s tasks are

defined as following: to systematically check the validity of studies on
climate change showing different results; to contrast divergent profiles;
to establish a minimal consensus by ranking indicators of qualitative
probability (in the sense of likelihood, not quantitative probability) and
modelling them in different scenarios. Reflecting on this example of
transnational and transdisciplinary knowledge production, one might
say that claims of validity of expertise can be measured by criteria of
plurality, plurality of negotiations, of actors, of knowledge bodies being
processed.
Knowledge production is an open-ended process, moving towards
unknown futures, taking into account unforeseen options and devel-
opments. Looking at the reciprocal and collective process of discourse
among experts with different outlooks and across boundaries of dis-
ciplines, professions, and spheres of knowledge, expert knowledge can-
not but be recognized as a social construction, as socially created in a
societal practice. Depending on its (social) location and other factors,
such practice may take on different forms, bring about different results
and hence must be seen as a contingency, as appearing not inevitably.
The constructiveness and contingency of expert knowledge and stand-
points is further underlined by the limited duration of expert networks
alluding to the temporality of knowledge bodies and therefore limited
claim of validity.
Bearing these characteristics in mind, expertise in the discourse on
post-modernity is reflected as an attitude, a non-imperialistic attitude
of acknowledging plurality and contingency, the otherness of other sys-
tems and practices of knowledge, while at the same time refusing to
give up ones’ own claim of validity. The practice of plurality as stated
by Welsch (1993, p. 322) is said to be an issue of hard work. Plurality
is endangered, on the one hand, through superficiality, that is neglect
of conflicts and oppositions between concepts and, on the other hand,
through indifference or arbitrariness in the sense of “anything goes.”
Consequently, when analysing expert interviews, attention is drawn to
the expert’s awareness of contingencies, to his habitus and practice of
controlling plurality and contingencies, particularly, following Welsch,
to the practices of acknowledgement and erasure of plurality and con-
tingencies, and to the use of irony.

(d) Explicitness of expert knowledge: Expert knowledge – in classical soci-

ology of knowledge constructed as an ideal type and not reconstructed
empirically – is defined as a special knowledge which the expert is clearly
and distinctly aware of. It is located in “discursive consciousness” – to
use the terminology of Giddens (1984).16 For both Schütz (1964) and
Sprondel (1979) this constitutes a crucial difference as compared with
30 Michael Meuser and Ulrike Nagel

everyday knowledge. Yet in view of the more recent discussion of the term
expert in the sociology of knowledge and of empirical research on expert
behaviour, this standpoint cannot be maintained. Contrary to Schütz’
and Sprondel’s assumption of the explicitness of expert knowledge, light
is shone to the tacit parts of expert knowledge, to relevances escaping the
experts’ awareness and reflection. Not only everyday knowledge, but also
expert knowledge comprises pre-theoretical experimental knowledge
(Bogner and Menz, 2002b, see also Bogner and Menz, in this volume,
Köhler, 1992; Meuser and Nagel, 1994; Schröer, 1994).17

The new forms of knowledge production appearing in a pluralized and

globalized world are to be considered as accompanied by highly detailed
expression and overstatement of viewpoints; by an increased necessity to
discuss options thoroughly; and by elaborating maximum contrasts between
approaches (Welsch, 1993, p. 323). Therefore, an increased level of explicit-
ness of expert knowledge might rightly be expected. However, together with
the change towards a knowledge-based society and continuous growth in
differentiation and reflection of knowledge (Schütze, 2002), an increase in
non-explicit knowledge is to be observed. According to Rammert (2003),
this is not a zero-sum game; the explicitness of knowledge does not neces-
sarily result in a decrease of implicit stocks of knowledge. Paradoxically,
it rather were the case “that in the intensified process of explicating and
formalizing growing stocks of knowledge [...] the relevance of non-explicit
knowledge emerges clearer and clearer” (Rammert, 2003, p. 484; emphasis in
original), namely, collectively shared experiences, incorporated knowledge,
rules of thumb, organizational routines, informal rules. Adopting Knorr-
Cetina’s understanding of epistemic cultures, a dynamic concept of know-
ledge needs to be applied: no longer should knowledge be conceptualized as
a fixed product, as an “intellectual or technological product” but considered
as a “process within definite contexts of production,” directing the ana-
lytic attention towards “knowledge as it is practiced” (Knorr-Cetina, 2002,
pp. 17–18). Yet the operational knowledge guiding and orienting a person’s
behaviour is difficult to be accessed consciously, it is hardly to “reeled off”
just like that in the interview. However, it is seizable in the empirical data
and open to reconstruction from what the interviewee tells.18 It is to be
achieved favourably on the basis of narrations of concrete problems, con-
flict and problem solutions taken from the expert’s experience (see below).
Summing up, what follows from the diagnosis of present-day society
regarding the analysis of expert knowledge and starting from the concept
of expert interview as an open method which is not guided by a script or
a preconceived sequential order but solely by topics to be covered in the
course of interviewing: Firstly, the diagnosis of our time leads to adopting a
process-oriented analytical view on expert knowledge. Questions to be posed
may regard reference groups, important spheres of knowledge; accounting
 Experts and Changes in Knowledge Production 31

of the plurality and globalization of knowledge. Relevant aspects refer to

the embeddedness of expert knowledge in milieus and socio-cultural set-
tings. Secondly, expert knowledge is defined by the communicative prac-
tice of insider groups and networks; the procedures of negotiating opinions;
and the strategies of inclusion and exclusion. Thirdly, another element is
the biographical mixture of the expert status within life-worldly, private as
well as public spheres of experience. Fourthly, in view of the complexity,
uncertainty, and ambiguity of expert knowledge, it is the expert’s habitus,
his awareness of contingencies, and his strategies of self-assurance, which
finally come into focus as an essential component of her or his knowledge.

1.3 Data collection

According to the procedural nature and the non-explicitness of considerable

parts of expert knowledge like tacit or pre-theoretical experiential know-
ledge, the expert interview is not likely to be conceptualized as a process
of extracting knowledge from the interviewee by asking questions in the
sense of a questionnaire. Referring to Giddens’ distinction between prac-
tical and discursive consciousness, the expert knowledge connected with
habitual ways of problem management can be located in between the two
poles. Neither is it a fully pre-reflexive knowledge at the level of basic rules
or ethno-methods as analysed by Cicourel (1972), nor is it comparable to the
knowledge of grammatical rules, which certainly most individuals intui-
tively have command of but are only partly capable of making explicit.
Experts will report cases of decision-making and state principles they keep
to; these are the data necessary for reconstructing the supra-individual,
field-specific patterns of expert knowledge.
Against this background, we consider an open interview based on a topic-
guide to be appropriate for data collection. As regards the reconstruction of
knowledge, which underlies expert behaviour, questionnaires would at best
allow for knowledge at the level of the discursive consciousness containing
rationalist reasoning corresponding with officially accepted standards. This
type of argument is to be found quite often in expert interviews, but apart
from the rare case in which the interviewee does not really cooperate, that
is answers with semi-official statements, experts do reveal a lot more about
relevances and maxims connected with their positions and functions: when
they carry on talking about their activities, extemporize, give examples,
or use other forms of exploration. The open interview provides the room
for the interviewee to unfold his own outlooks and reflections. As to data
collection interviewing should be based on general topics but avoid closed
questions and a prefixed guideline.
For the interviewer it is a must to prepare the interview topics thor-
oughly and build up a knowledge base of the field the experts are moving
in. Different from the narrative interview, in the expert interview naivety
32 Michael Meuser and Ulrike Nagel

would involve the risk of presenting oneself to the expert as an incompe-

tent interviewer. Such renunciation would also mean taking a wrong course
methodically, as the focus is not on the individual expert’s biography but
on action strategies and criteria of decision-making connected with a par-
ticular position.
The effort invested into the design of the topic guide provides the inter-
viewer with the thematic competence enabling him for productive interview-
ing. The motto “Ignorance as Method” (Hitzler, 1991) led the wrong way.
Most of the time the researcher cannot afford to be too naive and ignorant.
When collecting expert interviews with managers, Trinczek (1995, p. 65)
found that the readiness of the respondents “to bring up their knowledge
and viewpoints” is decisively influenced by the competency with which the
interviewer is able to present her or himself. It is important to be informed
in advance about rules, principles, statutes within the expert’s context, for
instance about collective agreements and the fundamentals of labour law
when making a study on human resources management. Much the same is
due to crucial events being published in the media.19 An uninformed inter-
viewer might cause doubts as much about her or his competence as her or
his involvement. To be short of knowledge of events inside the respective
institution is forgivable.
The expert interview and the biographical interview differing clearly
with regard to objective and methodical design, narrative passages are not
excluded from the expert interview. Narratives about episodes in the field
of the expert’s professional activity turn out to be key points of reference for
the reconstruction of orientations guiding conduct. Methodically, this can
be put to good use by eliciting narrations through the mode of interview-
ing. Narratives provide insight into the tacit aspects of expert knowledge,
which she or he is not fully aware of and which, on the contrary, become
noticed only gradually in the course of the narration.20 This may be illus-
trated by a narrative sequence drawn from an interview with a personnel
manager who, by tackling a case of decision-making over candidates for a
higher position, seeks to illustrate the criteria underlying the final result. 21

There was a woman, a mechanical engineer, who had been here for an
interview, and she should possibly be considered for a position as super-
visor of outside work, and then it was argued that it was likely she wouldn’t
be able to come out on top against a totally male-dominated outdoor
staff, where the going is a bit rough, you know, because no one is treated
with kid gloves out there. It’s unpleasant work, in all weathers, it partially
is dirty work, and also there are just distressing decisions to be made in
personell [sic] matters and orders to be given and so on, and then it was
said, no, she would probably not be able to do that, you know. And then
our employees’ representative spoke out in favor of that woman and said,
why not give her a chance, if one always raises objections, women can
 Experts and Changes in Knowledge Production 33

never prove themselves, and then the gentlemen in charge who should
possibly employ the lady felt more and more uneasy, and I really have
to say, and that now is something I meant earlier regarding one’s own
view, you can leave it on, I don’t mind [this refers to the tape recorder;
MM/UN], and when you realize, there are about eight or ten candidates,
and one of them was a woman, and she enters the room on very high
heels, she is wearing a skirt, she sits down and makes a good impres-
sion and tells you that her previous employment was in a construction
department, a drawing office, where she designed engine heads and was
responsible for production and so on, and now she would have to, or this
is what one should imagine, that now in this position she really tackles
her tasks, then I, too, have had my doubts, you know, and then I finally
said, well, all right, that man X certainly is the better choice. That’s about
how it goes. So despite all efforts to reject common prejudice, once again
it comes to the conclusion, that, alas, she wouldn’t be able to do the job,
after all.

Through this narrative about a particular case, general criteria of decision-

making in recruitment become as clear as it becomes apparent how criteria
of decision-making in an organization, in connection with gender-related
structures of prejudice, bring about a practice which is typical and not
unfamiliar. While narrating his story the interviewee becomes aware that
he is being recorded and by and by is tangled up in a narrative the unfolding
and outcome of which he could not clearly foresee from the start. In doing
so he reveals more about the relevance structures underlying his behaviour
than he would actually do – and would be able to do – if he were asked dir-
ectly. It can be presumed that the knowledge coming to light is located at
the border of practical and discursive consciousness.
Apart from narratives, reports about breaches of routine activities are
particularly instructive (Walter, 1994, p. 275). Mechanisms of orderly func-
tioning become apparent by looking at how breaches and conflicts are con-
trolled. This corresponds to the logic of breaching experiments in early
ethnomethodology (Garfinkel, 1967). It is obvious that the occurrence of
such reports cannot be anticipated when designing an interview schedule.
Therefore, it is all the more important to carry out the interviews in such a
way that (a) does not prevent the expert from addressing unforeseen aspects
of topics and (b) utilizes such aspects in subsequent interviews. From our
experience it is crucial for a successful outcome of an expert interview to
use the schedule in a flexible, non-bureaucratic way – that is as a thematic
guideline and not as if it were a questionnaire to be administered. It is the
relevance structures of the interviewees, which shall be elicited, not those
of the interviewer.
Questions should be focused on the how of decision-making and acting. In
this way, general principles and maxims can be grasped, and a reconstruction
34 Michael Meuser and Ulrike Nagel

of the logic underlying a decision is facilitated. Furthermore the wording of

questions should aim at the supra-personal level of knowledge and relate to
the institutions in question. Unless the interviewee retreats to his formal
role, his personal qualifications come up anyway. So it becomes clear what
the official institutionalized reality is as well as to what extent the expert
acts in a field of conflict between perceived institutional directives and his
personal interpretation of rules. Questions about decisions made or proc-
esses having taken place eventually allow a reconstruction of the logic and
procedure of the decision-making process. Follow-up questions should as far
as possible evoke accounts of concrete events or generate narrations.
The dynamics of the situation in expert interviewing, like with other
types of interviews and other reactive methods, is decisively determined by
the mutual perceptions of the participants. The impact of the interviewer’s
appearance as determined by factors like age, gender, and social status has
already been described in the literature on the (structured) interview. In
expert interviewing both the status relation and gender relation play a
prominent role.
Regarding the status relation it has been stated above that the outcome of
the interview is co-determined by whether the expert sees the interviewer as
a competent interlocutor. The ascription of competence to the interviewer
is based not only on his behaviour, but also on aspects of his formal status.
Interviewers who hold a high academic degree are often granted “credit,” so
to speak, as regards their competence. On the other hand the experts might
be sceptical about young researchers who do not yet hold a higher degree,
in particular in cases of a high status of the expert. This seems to play a role
in management research. In interviews with executive personnel, Trinczek
(2002, p. 219, see also Trinczek, in this volume) has observed expectations
of the following kind: “If the university hasn’t even bothered to send over a
‘real’ professor, the researcher should at least have got a PhD.”
The gender relation appears to be increasingly important in a society,
which is not only hierarchically structured but also showing an unequal
distribution of acknowledged expertise to the disadvantage of women. In
most fields of research the experts under study are males despite all changes
in gender-relations. This is to be understood in connection with the devel-
opment of professions and particularly applies to the functional elites of
society (Littig, 2002). In the historical development of occupations and
professions, expertise has become a field predominated by males, “a man’s
business,” so to speak (Wetterer, 1992). Methodical and methodological
consequences resulting from the dominating “manliness” of the expert sta-
tus, is an issue raised only rarely, as Littig (2002, p. 191) rightly remarks.
In situations of expert interviewing the female researchers’ professional
status is often neglected. Young female researchers who are dealing with a
male-dominated field are especially affected by the fact that in the experts’
understanding gender status dominates over professional status. If this is
 Experts and Changes in Knowledge Production 35

the case, the female researcher is considered to be “acceptably incompetent”

(Gurney, 1985). Abels and Behrens report how they utilized this constella-
tion in interviews with male experts:

In many talks with experts we claim to have received important pieces

of information just because men find it necessary to explain matters
through and through or to come up with facts which we as women in an
ascribed lower status are not believed to be capable of getting right. These
projections and the concomitant frankness can be further increased
through naïve and humble questions appealing to the expert’s readiness
to enlighten us. (1998, p. 86)

The ascription of features and character traits on the basis of gender may
foster expectations on the respondent’s side, which can become a burden
for the female researcher but can also be strategically instrumentalized with
regard to the research objectives.

1.4 Analysis22

Different from the analytic approach appropriate for case studies, in the
analysis of expert interviews attention is focused on thematic units, that
is passages with similar topics which are scattered about the interviews.
Sequentiality of statements within a single interview is not of interest.
Instead, what gains importance is the institutional-organizational context
within which the expert’s position is embedded and which provides the
actor with guiding principles. Right from the beginning of the analysis, the
context is taken into account in order to assess the meaning and signifi-
cance of the expert’s statements – no matter at what point in the course of
the interview they appear. The context as commonly shared by the experts
largely ensures the comparability of the interviews; in addition, comparabil-
ity is guaranteed through the use of the topic guide. This guideline reflects
the relevant topics against a horizon of other possible topics and serves to
focus the interviews.
Transcription: As a general rule interviews are being taped. Transcriptions
of thematically relevant passages are a prerequisite for the analysis. A tran-
scription of the whole recording – in contrast to working with biographical
interviews – is not standard. The transcription is also less detailed; prosodic
and paralinguistic elements are notated only to a certain extent.
Paraphrase: The sequencing of the text according to thematic units is eas-
ily done, as it were, in the manner of commonsense reasoning. In order to
rule out a narrowing of the thematic comparison of passages from the dif-
ferent interviews – the next but one step in the analysis – and to avoid to
“give away reality,” the paraphrase should follow the unfolding of the con-
versation and give account of the interviewee’s opinions.
36 Michael Meuser and Ulrike Nagel

Coding: The next step in condensing the material is to order the para-
phrased passages thematically. The interpreter keeps close to the text and
adopts the terminology of the interviewee. At best a term or phrase can be
used as it is. Whether one or more coding categories are attached to a pas-
sage depends on how many topics are addressed. It is allowed and necessary
to break up the sequentiality of the text also within passages, since the sub-
ject matter of the analysis is not the totality of the individual person’s life.
The frame of reference at this stage in the analysis still is the single inter-
view; condensations, typifications, abstractions remain within its horizon.
Thematic comparison: From this stage onward the analysis surpasses the
single passage in the text. The logic of the procedure corresponds to that
of coding, but now thematically comparable passages from different inter-
views are tied together (cf. Nagel, 1986). Category formation close to the
language of data has to be maintained; theoretical abstraction should be
refrained from, if possible. Since in the course of the thematic comparison
a large amount of data is condensed, it is essential to check and if necessary
revise coding decisions. The results of the thematic comparison have con-
tinuously to be checked in the light of the other relevant passages in the
interviews, to examine whether they are sound, complete and valid.
Sociological conceptualization: It is only now that a distant reviewing of the
texts and the terminology of the interviewees takes place. Features shared
and features differing from interview to interview are elaborated and cat-
egorized by drawing on the theoretical knowledge base. The specific charac-
teristics of the commonly shared knowledge of experts are condensed and
categorizations formulated. The process of category formation implies a sub-
sumption of phenomena under a term claiming general validity, on the one
hand, and a reconstruction of this term as valid for the reality under scru-
tinization, on the other hand. The level of abstraction is that of empirical
generalization. Statements refer to structures of expert knowledge. While
establishing links to the academic discourse, the generalizations remain
restricted to the empirical data, even though a terminology is used which
cannot be found in the material itself.
Theoretical generalization: The researcher arranges the categories accord-
ing to their internal relations. When representing the results of research
the empirically generalized findings are framed by a theoretically inspired
perspective. In this reconstructive process the meaning structures of
the field of action under study are connected to form typologies and
theories – overcoming loose ends and unconnected findings so far handled
pragmatically.
Regarding data analysis all stages of the analytical process should be
passed through and shortcuts be avoided. What is more, while the process
of interpretation is progressing it often proves necessary to go back to an
earlier stage in order to check the adequacy of generalizations as grounded
in data. This recursiveness is the typical merit of this approach.
 Experts and Changes in Knowledge Production 37

Notes
1. See, for example, Hitzler, Honer and Maeder (1994), Brinkmann, Deeke and Völkel
(1995), Schulz (1998), Bogner, Littig and Menz (2002, 2nd edition 2005), Mieg and
Näf (2006).
2. As we showed elsewhere, the expert interview is one of the most frequently
applied methods of empirical research, both as a method in its own right and
in the framework of a triangulation of methods. It is applied particularly in
industrial sociology, and organizational, educational, and policy research.
Although the expert interview has been used frequently for a long time it was
hardly ever reflected methodologically and with a view to the characteris-
tics as compared with those of other methods of interviewing. The result is a
certain “confusion and inconsistency” as to the understanding of the expert
interview (Mieg and Brunner, 2004, p. 199; for a similar view see also Bogner
and Menz, 2002a, p. 20). Correspondingly, on checking through relevant
handbooks on methods it turns out that the expert interview is at best men-
tioned as an exploratory method. Exceptions are Flick (1995), von Alemann
(1995), Friebertshäuser and Prengel (1997), Nohlen and Schultze (2002), Kühl
and Strodtholtz (2002), Bohnsack, Marotzki and Meuser (2003), Becker and
Kortendiek (2004). It is widely agreed to locate the expert interview in the con-
text of qualitative social research methodology. Our article is written in this
line of thought which however is not unchallenged but criticized as an unjus-
tified option. (Deeke, 1995).
3. Similar Pfadenhauer (2002, pp. 127–8, our translation), for whom the expert inter-
view is “a method of data collection which is highly ambitious involving great
effort.” See also Pfadenhauer in this volume.
4. Our comments are based on previous articles on expert interview and expert
knowledge (Meuser and Nagel, 1991, 1994, 1997, 2002). In this article our former
considerations are discussed in the light of changes in knowledge production and
extended with a view to the impact on the analysis of expert knowledge. Doubts
about our conceptualization of the expert interview as a method in its own right
are expressed by Deeke (1995), Kassner and Wassermann (2002).
5. In previous articles we treated this contextualization of the expert and expert
knowledge somewhat briefly; meanwhile the changes in knowledge production
have become more visible and require to be dealt with – which we shall do in this
article. It will be examined whether and in which way the processes of societal
change – discussed as newly emerging knowledge cultures and forms of know-
ledge production – are challenging a revision of concepts drawn from the soci-
ology of knowledge.
See Gibbons and others (1994) on the change of forms and cultures of know-
ledge production in different spheres of society (economy, education, the human-
ities and the social sciences). For an analysis of knowledge cultures in the sciences
and high-technology see Knorr-Cetina (2002), and in general Rammert (2003).
For critique cf. the contributions in Bender (2001) and a review of publications on
the Knowledge Society (Hettlage, 2004).
6. A confusion of terms can occasionally be observed also in the method’s literature
(cf., for example, Gläser and Laudel, 2004).
7. Cf. Pfadenhauer (2003, p. 171) who assumes that counter-experts have undergone
a marked change in status with their expertise being upgraded, and accordingly
are no longer labelled as “counter”-experts.
38 Michael Meuser and Ulrike Nagel

8. For example, not so long ago judges at the German Federal Constitutional Court
consulted the Chaos Computer Club, an association of hackers, with regard to
the legal processing of safety provisions for computers used in the registration of
votes (Kölner Stadt-Anzeiger, 29 August 2007).
9. Cf. the majority of the contributions in Bender (2001). Above all warnings are
also concerning the enthusiasm which seems to drive Gibbons and others in
view of Mode 2; cf. also Hettlage (2004).
10. An organization like Greenpeace is a good example illustrating the observation
that an increase in institutionalization and recognition usually is coupled with
a professionalization and scientification of also the non-professional or counter-
expertise (For the institutionalization of social movements in general cf. Rucht,
Blattert and Rink, 1997).
11. Pfadenhauer (2003, p. 172) points out that the expertise of counter experts is
as much opinionated as those of other experts implying that they are not at all
covering the role of disinterested third party. It should be mentioned here that
the expert interview like other interviewing methods is used to gather informa-
tion about a subject matter as seen from various perspectives, and although the
expert’s knowledge may be of a privileged nature it is nonetheless tied to a pos-
ition like any other knowledge.
12. In postmodern discourse it is claimed that the dividing line between insiders
and outsiders, insider and outsider knowledge becomes blurred; we do not share
this approach. In accordance with the perspective of the sociology of know-
ledge we maintain the distinction between expert and lay knowledge, though
not the clear-cut distinction between the professional expert and the layman. As
mentioned above, we place a fourth figure besides layman, well-informed citi-
zen and professional, namely the active participant with her or his input of new
relevances, fulfilling the function of a critical corrective against the side-effects
of the modernization process.
13. Thus the production of expert knowledge develops as an independent subject
matter to be studied in future research on different groups of experts.
14. We are not going into the question raised by Knorr-Cetina as to whether this
logically implies a departure from traditional definitions of knowledge.
15. In contrast to the model of negotiation discussed here cf. the study of Bogner
and Menz (2002a) on the rationality of decision-making in the political system.
16. Giddens (1984) distinguishes a discursive consciousness from a practical con-
sciousness. The first can be outlined by a person when asked to do so since it is
explicitly represented in the stock of knowledge; the second can only be recon-
structed from the implicit stock of knowledge.
17. Cf. Bogner and Menz (2002b, pp. 43–4) who distinguish between three dimen-
sions of expert knowledge to be hold analytically separate: “technical know-
ledge,” “process knowledge,” and “interpretive knowledge.” Technical knowledge
is expert knowledge in the narrower sense; it is explicit knowledge and can be
directly communicated in the interview. Process knowledge is conceived of as
“practical experiential knowledge” resulting from frequently and repeatedly
performed action patterns and interaction routines. Interpretive knowledge is
created not only in functional contexts but is additionally shaped by subjective
relevances and viewpoints. The aforementioned embeddedness of expert know-
ledge in the biographical sphere and in the life-world is bearing effect here.
18. Against the background of a tradition of ethnographic research, Pfadenhauer
(2003, p. 160) calls into question that the expert interview and interviewing are
generally suitable to gain insight into pre-theoretical experiential knowledge.
 Experts and Changes in Knowledge Production 39

She thereby touches upon issues of method and methodology under discussion
inside the world of qualitative social research which we would prefer to refrain
from in this article.
19. For more detailed comments see Pfadenhauer (2002) who even views the expert
interview as “a conversation between expert and quasi-expert.”
20. This is caused by the structural dynamics of narrations (cf. Schütze, 1982).
21. The interview was carried out for a study of implementation of equal treatment
policy in public administration (Meuser, 1989).
22. A detailed presentation of the analytical steps can be found in Meuser and Nagel
(1991, reprinted in Bogner, Littig and Menz, 2002).

Further readings
Miller, R. L. (2005) Biographical Research Methods, Vol. 4, Disputes and Concerns in
Biographical Research (London: Sage).
Nowotny, H., Scott, P. and Gibbons, M. (2001) Re-Thinking Science: Knowledge and the
Public in an Age of Uncertainty (Oxford: Polity Press).
Schütz, A. (1964) “The Well-Informed Citizen: An Essay on the Social Distribution of
Knowledge” in Collected Papers, Vol. 2, pp. 120–34 (The Hague: Nijhoff).

References
Abels, G. and Behrens, M. (1998) “ExpertInnen-Interviews in der Politikwissenschaft.
Das Beispiel Biotechnologie” in Österreichische Zeitschrift für Politikwissenschaft 27,
79–92.
Alemann, U. von (1995) Politikwissenschaftliche Methoden. Grundriss für Studium und
Forschung (Opladen: Westdeutscher Verlag).
Bauman, Z. (1991) Modernity and Ambivalence (Ithaca, NY: Cornell University Press).
Beck, U. (1992) Risk Society: Towards a New Modernity (London: Sage).
Beck, U. (1998) World Risk Society (Cambridge: Polity Press).
Beck, U., Giddens, A. and Lash, S. (1994) Reflexive Modernization. Politics, Tradition and
Aesthetics in the Modern Social Order (Cambridge: Polity Press).
Becker, R. and Kortendiek, B. (2004) Handbuch Frauen- und Geschlechterforschung
(Wiesbaden: VS-Verlag).
Behnke, C. and Meuser, M. (2003) “Doppelkarrieren in Wirtschaft und Wissenschaft”
in Zeitschrift für Frauenforschung und Geschlechterstudien 21, 62–74.
Behnke, C. and Meuser, M. (2005) “Vereinbarkeitsmanagement. Zuständigkeiten und
Karrierechancen bei Doppelkarrierepaaren” in Solga, H. and Wimbauer, C. (eds)
“Wenn zwei das Gleiche tun ...” – Ideal und Realität sozialer (Un-)Gleichheit in Dual
Career Couples (Opladen: Verlag Barbara Budrich), pp. 123–39.
Bender, G. (2001) Neue Formen der Wissenserzeugung (Frankfurt am Main and
New York: Campus).
Bogner, A., Littig, B. and Menz, W. (2002) Das Experteninterview. Theorie, Methode,
Anwendung (Opladen: Leske and Budrich).
Bogner, A. and Menz, W. (2002a) “Expertenwissen und Forschungspraxis: die
modernisierungstheoretische und methodische Debatte um die Experten” in
Bogner, A., Littig, B. and Menz, W. (eds) Das Experteninterview. Theorie, Methode,
Anwendung (Opladen: Leske and Budrich), pp. 7–29.
Bogner, A. and Menz, W. (2002b) “Das theoriegenerierende Experteninterview.
Erkenntnisinteresse, Wissensformen, Interaktion” in Bogner, A., Littig, B. and
40 Michael Meuser and Ulrike Nagel

Menz, W. (eds) Das Experteninterview. Theorie, Methode, Anwendung (Opladen: Leske

and Budrich), pp. 33–70.
Bohnsack, R., Marotzki, W. and Meuser, M. (2003) Hauptbegriffe qualitativer
Sozialforschung (Opladen: Leske and Budrich).
Bonss, W. (1998) “Uneindeutigkeit, Unsicherheit, Pluralisierung, Postmoderne. Eine Bilanz”
in Merkur 52, 968–75.
Brinkmann, C., Deeke, A. and Völkel, B. (1995) “Experteninterviews in der
Arbeitsmarktforschung. Diskussionsbeiträge zu methodischen Fragen und
praktischen Erfahrungen” in Beiträge zur Arbeitsmarkt- und Berufsforschung 191
(Nürnberg: IAB).
Cicourel, A. V. (1972) “Basic and Normative Rules in the Negotiation of Status and
Role” in Sudnow, D. (ed.) Studies in Social Interaction (New York: The Free Press),
pp. 229–58.
Deeke, A. (1995) “Experteninterviews – ein methodologisches und forschung-
spraktisches Problem. Einleitende Bemerkungen und Fragen zum Workshop”
in Brinkmann, C., Deeke, A. and Völkel, B. (eds) Experteninterviews in der
Arbeitsmarktforschung. Diskussionsbeiträge zu methodischen Fragen und praktischen
Erfahrungen. Beiträge zur Arbeitsmarkt- und Berufsforschung 191 (Nürnberg: IAB),
pp. 7–22.
Flick, U. (1995) Qualitative Forschung. Theorie, Methoden, Anwendung in Psychologie und
Sozialwissenschaften (Reinbek: Rowohlt).
Friebertshäuser, B. and Prengel, A. (1997) Handbuch Qualitative Forschungsmethoden in
der Erziehungswissenschaft (Weinheim and München: Juventa).
Garfinkel, H. (1967) Studies in Ethnomethodology (Englewood Cliffs: Prentice Hall).
Garz, D. and Kraimer, K. (1991) Qualitaiv-empirische Sozialforschung (Opladen:
Westdeutscher Verlag).
Gibbons, M., Limoges, C., Nowotny, H., Schwartzmann, S., Scott, P. and Trow, M.
(1994) The New Production of Knowledge: The Dynamics of Science and Research in
Contemporary Societies (London: Sage).
Giddens, A. (1984) The Constitution of Society: Outline of the Theory of Structuration
(Cambridge: Polity).
Gläser, J. and Laudel, G. (2004) Experteninterviews und qualitative Inhaltsanalyse
(Wiesbaden: VS-Verlag).
Gorden, R. L. (1975) Interviewing: Strategy, Techniques, and Tactics (Homewood: Dorsey
Press).
Gurney, J. N. (1985) “Not One of the Guys: The Female Researcher in a Male-
Dominated Setting” in Qualitative Sociology 8, 42–62.
Habermas, J. (1985) Die neue Unübersichtlichkeit (Frankfurt am Main: Suhrkamp).
Hettlage, R. (2004) “Die ‘Wissensgesellschaft’ im Verzauberungs-Entzauberungs-
Zirkel” in Soziologische Revue 27, 407–21.
Hitzler, R. (1991) “Dummheit als Methode. Eine dramatologische Textinterpretation”
in Garz, D. and Kraimer, K. (eds) Qualitativ-empirische Sozialforschung (Opladen:
Westdeutscher Verlag), pp. 295–318.
Hitzler, R., Honer, A. and Maeder, C. (1994) Expertenwissen. Die institutionalisierte
Kompetenz zur Konstruktion von Wirklichkeit (Opladen: Westdeutscher Verlag).
Kassner, K. and Wassermann, P. (2002) “Nicht überall, wo Methode draufsteht, ist
Methode drin. Zur Problematik der Fundierung von ExpertInneninterviews” in
Bogner, A., Littig, B. and Menz, W. (eds) Das Experteninterview. Theorie, Methode,
Anwendung (Opladen: Leske and Budrich), pp. 95–111.
Keller, R. (2005) Wissenssoziologische Diskursanalyse. Grundlegung eines Forschungsprogramms
(Wiesbaden: VS-Verlag).
 Experts and Changes in Knowledge Production 41

Knorr-Cetina, K. (2002) Wissenskulturen. Ein Vergleich naturwissenschaftlicher

Wissensformen (Frankfurt am Main: Suhrkamp) (Original: Epistemic Cultures. How
the Sciences make Knowledge (1999) Cambridge/London: Harvard University Press).
Köhler, G. (1992) “Methodik und Problematik einer mehrstufigen Expertenbefragung”
in Hoffmeyer-Zlotnik, J. H. P. (ed.) Analyse verbaler Daten. Über den Umgang mit
qualitativen Daten (Opladen: Westdeutscher Verlag), pp. 318–32.
Kühl, S. and Strodtholz, P. (2002) Methoden der Organisationsforschung. Ein Handbuch
(Reinbek: Rowohlt).
Liebold, R. and Trinczek, R. (2002) “Experteninterview” in Kühl, S. and Strodtholz, P. (eds)
Methoden der Organisationsforschung. Ein Handbuch (Reinbek: Rowohlt), pp. 33–70.
Littig, B. (2002) “Interviews mit Experten und Expertinnen. Überlegungen aus
geschlechtertheoretischer Sicht” in Bogner, A., Littig, B. and Menz, W. (eds) Das
Experteninterview. Theorie, Methode, Anwendung (Opladen: Leske and Budrich),
pp. 173–90.
Meuser, M. (1989) Gleichstellung auf dem Prüfstand. Frauenförderung in der
Verwaltungspraxis (Pfaffenweiler: Centaurus).
Meuser, M. and Nagel, U. (1991) “ExpertInneninterviews – vielfach erprobt, wenig
bedacht. Ein Beitrag zur qualitativen Methodendiskussion” in Garz, D. and
Kraimer, K. (eds) Qualitativ-empirische Sozialforschung (Opladen: Westdeutscher
Verlag), pp. 441–71.
Meuser, M. and Nagel, U. (1994) “ExpertInnenwissen und ExpertInneninterview”
in Hitzler, R., Honer, A. and Maeder, C. (eds) Expertenwissen. Die institutionalisi-
erte Kompetenz zur Konstruktion von Wirklichkeit (Opladen: Westdeutscher Verlag),
pp. 180–92.
Meuser, M. and Nagel, U. (1997) “Das ExpertInneninterview – Wissenssoziologische
Grundlagen und methodische Durchführung” in Friebertshäuser, B. and Prengel, A.
(eds) Handbuch Qualitative Forschungsmethoden in der Erziehungswissenschaft
(Weinheim and München: Juventa), pp. 481–91.
Meuser, M. and Nagel, U. (2002) “Vom Nutzen der Expertise. ExpertInneninterviews
in der Sozialberichterstattung” in Bogner, A., Littig, B. and Menz, W. (eds) Das
Experteninterview. Theorie, Methode, Anwendung (Opladen: Leske and Budrich),
pp. 257–72.
Mieg, H. A. and Brunner, B. (2004) “Experteninterviews. Reflexionen zur
Methodologie und Erhebungstechnik” in Schweizerische Zeitschrift für Soziologie
30, 199–222.
Mieg, H. A. and Näf, M. (2006) Experteninterviews in den Umwelt und
Planungswissenschaften (Lengerich: Papst).
Nagel (Matthes-), U. (1986) “Modelle und Methoden rekonstruktiver Theoriebildung”
in Gerhard, E. and others (eds) Subjektorientiertes Lernen und Arbeiten – Ausdeutung
einer Gruppeninteraktion (Bonn: Deutscher Volkshochschulverband), pp. 29–55.
Nohlen, D. and Schultze, R. O. (2002) Lexikon der Politikwissenschaft, Vol. 2 (München:
Verlag C. H. Beck).
Nowotny, H. (1997) Im Spannungsfeld der Wissensproduktion und Wissensvermittlung,
http://www.unicom.unizh.ch/unimagazin/archiv/1-97/wissensproduktion.html,
accessed on 11 September 2007.
Nowotny, H., Scott, P. and Gibbons, M. (2004) Wissenschaft neu denken. Wissen
und Öffentlichkeit in einem Zeitalter der Ungewissheit (Weilerswist: Velbrueck
Wissenschaft) (Original: Re-Thinking Science: Knowledge and the Public in an Age of
Uncertainty (2001) Oxford: Polity Press).
Pfadenhauer, M. (2002) “Auf gleicher Augenhöhe reden. Das Experteninterview –
ein Gespräch zwischen Experte und Quasi-Experte” in Bogner, A., Littig, B. and
42 Michael Meuser and Ulrike Nagel

Menz, W. (eds) Das Experteninterview. Theorie, Methode, Anwendung (Opladen: Leske

and Budrich), pp. 113–30.
Pfadenhauer, M. (2003) Professionalität. Eine wissenssoziologische Rekonstruktion insti-
tutionalisierter Kompetenzdarstellungskompetenz (Opladen: Leske and Budrich).
Rammert, W. (2003) “Zwei Paradoxien einer innovationsorientierten Wissenspolitik:
Die Verknüpfung heterogenen und die Verwertung impliziten Wissens” in Soziale
Welt 54, 483–508.
Rucht, D., Blattert, B. and Rink, D. (1997) Soziale Bewegungen auf dem Weg zur
Institutionalisierung. Zum Strukturwandel “alternativer” Gruppen in beiden Teilen
Deutschlands (Frankfurt am Main and New York: Campus).
Schmid-Urban, P. (1992) Kommunale Sozialberichterstattung (Frankfurt am Main:
Deutscher Verein für öffentliche und private Fürsorge).
Schröer, N. (1994) “Routiniertes Expertenwissen. Zur Rekonstruktion des struk-
turalen Regelwissens von Vernehmungsbeamten” in Hitzler, R., Honer, A. and
Maeder, C. (eds) Expertenwissen. Die institutionalisierte Kompetenz zur Konstruktion
von Wirklichkeit (Opladen: Westdeutscher Verlag), pp. 214–31.
Schütz, A. (1964) “The Well-Informed Citizen: An Essay on the Social Distribution of
Knowledge” in Collected Papers, Vol. 2 (The Hague: Nijhoff), pp. 120–34.
Schütze, F. (1982) “Narrative Repräsentation kollektiver Schicksalsbetroffenheit” in
Lämmert, E. (ed.) Erzählforschung (Stuttgart: Metzler), pp. 568–90.
Schütze, F. (2002) “Das Konzept der sozialen Welt im symbolischen Interaktionismus
und die Wissensorganisation in modernen Komplexgesellschaften” in Keim, I. and
Schütte, W. (eds) Soziale Welten und kommunikative Stile (Tübingen: Narr Verlag),
pp. 57–83.
Schulz, W. K. (1998) Expertenwissen. Soziologische, psychologische und pädagogische
Perspektiven (Opladen: Leske and Budrich).
Sprondel, W. M. (1979) “ ‘Experte’ und ‘Laie’: Zur Entwicklung von Typenbegriffen in
der Wissenssoziologie” in Sprondel, W. M. and Grathoff, R. (eds) Alfred Schütz und
die Idee des Alltags in den Sozialwissenschaften (Stuttgart: Enke), pp. 140–54.
Trinczek, R. (1995) “Experteninterviews mit Managern: Methodische und
methodologische Hintergründe” in Brinkmann, C., Deeke, A. and Völkel, B.
Experteninterviews in der Arbeitsmarktforschung. Diskussionsbeiträge zu methodischen
Fragen und praktischen Erfahrungen. Beiträge zur Arbeitsmarkt- und Berufsforschung
191 (Nürnberg: IAB), pp. 59–68.
Trinczek, R. (2002) “Wie befrage ich Manager? Methodische und methodolo-
gische Aspekte des Experteninterviews als qualitativer Methode empirischer
Sozialforschung” in Bogner, A., Littig, B. and Menz, W. (eds) Das Experteninterview.
Theorie, Methode, Anwendung (Opladen: Leske and Budrich), pp. 209–22.
Walter, W. (1994) “Strategien der Politikberatung. Die Interpretation der
Sachverständigen-Rolle im Lichte von Experteninterviews” in Hitzler, R.,
Honer, A. and Maeder, C. (eds) Expertenwissen. Die institutionalisierte Kompetenz zur
Konstruktion von Wirklichkeit (Opladen: Westdeutscher Verlag), pp. 268–84.
Welsch, W. (1993) Unsere postmoderne Moderne (Berlin: Akademie Verlag).
Wetterer, A. (1992) Profession und Geschlecht. Über die Marginalität von Frauen in
hochqualifzierten Berufen (Frankfurt am Main and New York: Campus).
2
The Theory-Generating Expert
Interview: Epistemological Interest,
Forms of Knowledge, Interaction
Alexander Bogner and Wolfgang Menz

2.1 The methodological ambiguity of the expert interview

2.1.1 Introduction: the debate about expert interviews

Expert interviews are a good example of the way in which the everyday
practice of social research and theoretical consideration of this practice
do not always run parallel to one another. The use of particular methods
sometimes precedes their general theoretical reflection. For many years, the
widely held view was that expert interviews were conducted frequently but
only rarely thought through (Meuser and Nagel, 1991). Only in recent years
has the debate about expert interviews gradually become more concrete
(see Bogner and Menz, 2008). However, this has certainly not led to a situ-
ation in which the different definitions and methodological conceptions of
expert interviews have moved closer together. Even today there are disputes
not only about how expert interviews can be placed on a secure methodo-
logical footing, but also about whether this is even possible in principle.
Different positions continue to be opposed to one another. Meuser and
Nagel (1991, 1994, 1997), for example, have advocated a form of expert
interview that is genuinely situated in the qualitative paradigm. They argue
that the long period in which methodological reflection was not well devel-
oped can be explained by the lack of recognition accorded to the specific
strengths of this kind of interview, and by the persistence of a research trad-
ition in which the expert interview is usually seen as having no more than
an exploratory function.
Against this view, it has been argued that there is no codified model of
“the” expert interview and that there cannot be any such thing; or that, if
such a canonization is developed, it is bound to lose any value it may have
because it will attribute exaggerated methodological significance to the
experience of specific cases (Deeke, 1995, Kassner and Wasserman, 2005).
It is argued that the attempt to turn the expert interview into a particular

43
44 Alexander Bogner and Wolfgang Menz

method overlooks the contextual nature of research. This leads to a relativ-

ity in this form of interview which places three kinds of obstacle in the way
of any attempted methodological generalization. Firstly, all research relies
on a relational concept of the expert, which depends on the topic being
investigated (Deeke, 1995, p. 7). Secondly, conversations with experts con-
stitute a particular social situation that is especially susceptible to interfer-
ences; this may not invalidate the basic principles of how interviews should
be conducted, but it sets narrow limits to the range of prescriptive methodo-
logical rules available (Vogel, 1995, p. 82). Thirdly and finally, it is argued
that one cannot stipulate that expert interviews should be carried out in a
certain way. Interviewers will always have their own particular interest in
the subject under investigation, and their own concrete question to which
they are seeking an answer; this inevitably leads to flexibility in the use
of this instrument of enquiry (see Trinczek, in this volume). Depending
on the interviewer’s interest and the research question under investigation,
there will be differences between conversations with experts in respect of
the extent to which they are pre-structured, the openness with which they
are conducted, and the ways in which they are processed, evaluated and
interpreted. This seems to mean that a proliferation of ways of proceeding
is unavoidable in the area of expert interviews. And this gives us very few
grounds for optimism that this form of interview could ever take on firm,
distinctive contours.
In addition, expert consultations are not restricted to qualitative inter-
views, and neither can they be considered typical representatives of the
qualitative paradigm when they are carried out as semi-structured inter-
views. The expert interview is, on the one hand, suspected of inadequate
methodological rigour and of producing little more than impressionistic
results because it lacks standardization and quantification of the data, while
appearing on the other hand to be too narrow as a way of bringing the
interviewee’s relevance structures into the open in a “pure” way because
the conversation is actively guided and the interviewer occasionally inter-
venes to redirect it. The expert interview is thus seen as a methodological
hybrid which, notwithstanding numerous indications that inter-paradigm
debates are rapidly becoming less relevant to the practice of research (see
Kelle and Erzberger, 1999), has clear weaknesses because it belongs to both
worlds.
This may all sound rather dramatic. In fact, though, the controversy
about the methodology of the expert interview is not, in our view, caused
by fundamental methodological difficulties. We suspect that the problem
lies elsewhere, and that the debate about the expert interview is being kept
alive by a lack of clarity in the systematization of the different epistemo-
logical interests and research designs.
In this article, we proceed in the first section by identifying three dom-
inant forms of expert interview. This is designed to make clear the specific
 The Theory-Generating Expert Interview 45

status of, and claims made on behalf of, the expert interview, aspects which
are frequently only dealt with implicitly in the debate about methodology.
It becomes apparent that methodological generalizations only make sense
in relation to expert interviews that are explicitly situated in the “inter-
pretative paradigm” (Wilson, 1970). In accordance with this methodo-
logical situation of the issue, we do not treat the object of investigation
as a social fact, and we do not treat knowledge about it as the result of an
objective comprehension or passive reception of the facts of the situation.
Instead, our research attitude is a perspective from the sociology of know-
ledge, which understands social reality as a construction created by acts
of interpretation (Berger and Luckmann, 1966). Academic research, which
analyzes the social order on the basis of meanings and structures of rele-
vance, appears as an active-constructive process (Schütz, 1962). Basic social
constructivist assumptions like these (see Knorr-Cetina, 1989, Flick, 2004)
are central points of reference for the discussion which follows about the
nature and knowledge of experts and about suitable interaction strategies
for use in interviews.
In the second section, we begin by proposing an analytic differentiation
between forms of expert knowledge in the framework of the discussion about
different approaches to the concept of the expert. Only an explicit concept
of expert knowledge as an act of construction performed by the researcher
is capable of paving the way for a fundamental change of perspective with
regard to the interpretation and conceptualization of the interaction situ-
ation. We then outline a concept of the expert, which takes into account the
power and the social effects of expert knowledge. This concept owes much
to critiques of the theoretical absolutization of the interactionist production
of meanings and rules, such as can be found in the work of Blumer (1969)
and others. Our own proposal is based on an assessment that the status
of experts is not just produced by subjective-situative processes of inter-
pretation, but has its pre-existence confirmed to an equal degree by these
processes. This enables us to incorporate into our analysis systematic asym-
metries and inequalities that are not limited to local interaction structures
and on which expert status essentially rests.1
In the third section, we discuss this methodologically oriented reconstruc-
tion of the expert with regard to interpretations of the interaction situation
and the practical consequences they have in interviews. In concrete terms,
we argue that what is known as interaction effects should be treated not as
variables that distort the situation, but as elements, which are constitutive of
the process of data production. This rejection of an “archaeological” model
of data production, which conceptualizes expert knowledge as a buried
treasure that must be dug up while being kept free from contamination as
far as possible, is combined with the outline of a model of typical interaction
situations in the expert interview which treats the production of data as a
social process.
46 Alexander Bogner and Wolfgang Menz

2.1.2 A typology of the expert interview

There is considerable variation in the concepts of the expert interview
employed in the methodological literature on this subject. Meuser and
Nagel’s (1991) classification of the expert interview as belonging to the
paradigm of interpretative social research, which followed the logic of their
research orientation, was seen by some critics as an over-hasty attempt to
lay claim to a methodologically “neutral” instrument of inquiry on behalf
of one particular paradigm (see Deeke, 1995). It is evident that this criticism
rests on a competing concept of the expert interview, but this concept is not
made explicit in any systematic way. The following discussion therefore pro-
poses a differentiation of the dominant forms of expert interview employed
in the methodological debate, in accordance with their epistemological
functions. Following suggestions made by Vogel (1995) and with the help of
ideas from the relevant works by Meuser and Nagel, we distinguish between
exploratory, systematizing, and theory-generating expert interviews.

(1) The expert interview owes its prominence in empirical social research,
as a frequently employed instrument for the collection of data, to its func-
tion as an exploratory tool. In both quantitatively and qualitatively oriented
research projects, expert interviews can serve to establish an initial orienta-
tion in a field that is either substantively new or poorly defined, as a way of
helping the researcher to develop a clearer idea of the problem or as a pre-
liminary move in the identification of a final interview guide. In this sense,
exploratory interviews help to structure the area under investigation and
to generate hypotheses. The experts interviewed may themselves belong to
the target group of the study as part of the field of action, but in many cases
experts are also deliberately used as a complementary source of informa-
tion about the target group that is the actual subject. In the latter case, the
expert’s role is that of someone who possesses “contextual knowledge.”2
Exploratory expert interviews should be conducted as openly as possible,
but purely on grounds of demonstrative competence it is advisable to struc-
ture in advance at least the central dimensions of the planned conversation
with reference to a topic guide. In this respect the exploratory expert inter-
view differs from the narrative or episodic interview, though this does not
mean that any spontaneous digressions or unexpected changes of subject
on the part of the expert should be nipped in the bud. The focus of the
exploratory interview, in terms of its subject matter, is on sounding out the
subject under investigation. The objective is not to compare data, acquire
as much information as possible, or standardize the data. There is thus a
fundamental distinction between the exploratory interview and the other
two types.
(2) The systematizing expert interview is related to the exploratory variant in
that it is oriented towards gaining access to exclusive knowledge possessed
by the expert. The focus here is on knowledge of action and experience,
 The Theory-Generating Expert Interview 47

which has been derived from practice, is reflexively accessible, and can be
spontaneously communicated. This kind of expert interview is an attempt
to obtain systematic and complete information. The expert enlightens the
researcher on “objective” matters. This means that the expert is treated here
primarily as a guide who possesses certain valid pieces of knowledge and
information, as someone with a specific kind of specialized knowledge that
is not available to the researcher. With the help of a fairly elaborate topic
guide, the researcher gains access to this knowledge.3 The systematizing
expert interview is probably the most widespread form of this interview
method found in research practice.
The only comprehensive textbook published so far on the expert inter-
view (Gläser and Laudel, 2004) also belongs to this type. It is true that this
type does not restrict the concept of the expert to the person who is in
possession of particular, specialized knowledge of exceptional quality. This
approach also treats knowledge derived from practical everyday experience
as a possible object of expert interviews. The main focus, though, is not on
the interpretative character of expert knowledge but rather on its capacity
to provide researchers with facts concerning the question they are investi-
gating. Experts are a source of information with regard to the reconstruc-
tion of sequences of events and social situations: “Experts are people who
have special knowledge about social facts, and expert interviews are a way
of gaining access to this knowledge” (ibid., p. 10). From this methodological
perspective it is not the experts themselves who are the object of the investi-
gation; their function is rather that of informants who provide information
about the real objects being investigated.
Even so, systematizing interviews are not necessarily open and qualita-
tive. Standardized surveys – such as those used in, for example, the Delphi
method (see Aichholzer, 2009) – are also possible here. The final point to
make in this connection is that in the case of systematizing expert inter-
views, unlike exploratory interviews, it is important for the data to be com-
parable in relation to the subject matter.
One significant aspect of the systematizing expert interview is the way it
has become an important tool for the collection of data in the framework
of multi-method approaches (triangulation), for example in organizational
sociology. We suspect that the dominance of this form of pure enquiry
about information has contributed to the restricted understanding of con-
versations with experts which leads to the perception of the systematizing
type as pars pro toto. It may also be a paradoxical consequence of the popu-
larity of the systematizing expert interview that the connection between
empirical practice and methodological reflection is so weak in the case of
the expert interview.
(3) We use the term “theory-generating” for the type of expert interview
that was methodologically established by Meuser and Nagel and has been
developed further by these authors. In this case the expert no longer serves
48 Alexander Bogner and Wolfgang Menz

as the catalyst of the research process, or to put it another way as the means
by which the researcher can obtain useful information and elucidation of
the issue being investigated. The essence of the theory-generating interview
is that its goal is the communicative opening up and analytic reconstruction
of the subjective dimension of expert knowledge. Here, subjective action
orientations and implicit decision making maxims of experts from a par-
ticular specialist field are the starting-point of the formulation of theory.4
The researcher seeks to formulate a theoretically rich conceptualization of
(implicit) stores of knowledge, conceptions of the world and routines, which
the experts develop in their activities and which are constitutive for the
functioning of social systems. In ideal terms, this procedure seeks to gener-
ate theory via the interpretative generalization of a typology – by contrast
with the representative statistical conclusions that result from standardized
procedures. Following considerations put forward by Glaser and Strauss
(1967) on databased theory, qualitative theory is drawn up here via theor-
etical sampling and comparative analysis as a process of inductive theory
formulation, at the conclusion of which, ideally, the researcher will have a
“formal” theory. It follows that the theory-generating expert interview must
be classified as part of the methodological canon oriented to the fundamen-
tal principles of interpretative sociology.

This means that we have settled the paradigmatic fate of the theory-
generating interview. However, we still have to answer the question of
whether this kind of expert interview involves a specific method that can
be clearly distinguished from those used in paradigmatically related forms
of interview.5 One could challenge our argument so far by objecting that
expert interviews have been inadmissibly defined in terms of the object of
the investigation or the person being interviewed, and so cannot be a dis-
tinct method (see Kassner and Wassermann, 2005). The following discus-
sion of the concept of the expert is a response to this objection.

2.2 Who counts as an expert? A method-oriented

reformulation of the concept of the expert

2.2.1 The concept of the expert in the discussion about methods

Three different ways of looking at the definition of an expert can be found
in the debate about the methodological foundations of the expert interview.
These approaches rely on different analytic and normative perspectives, and
in the discussion that follows we identify them as the voluntaristic, the con-
structivist, and the sociology of knowledge concepts of the expert. However,
because of the convergence between the constructivist and the sociology
of knowledge approaches, it comes as no surprise that what we encoun-
ter in practice is usually a mixture of different conceptualizations oriented
towards specific research interests. We argue that a definition which insists
 The Theory-Generating Expert Interview 49

exclusively on the knowledge dimension of expertise leads to weaknesses

in relation to method and to contradictions, and in the conclusion to this
section we outline a concept of the expert which makes use of ideas from
modernization theory concerning changes in the figure of the expert.

(1) The voluntaristic concept of the expert starts from the undeniable fact
that every human being is in possession of particular information, capaci-
ties and so on which equip them to deal with their own everyday life; one
can thus speak in a general sense of a specific advantage in terms of know-
ledge relating to personal arrangements. This would mean that in principle,
everyone is an expert – an expert on their own life, or as Mayring (1996,
p. 49) has put it in methodological terms, “experts on their own meanings.”
This concept of the expert is inseparable from an unspecific asymmetry
in knowledge, and has been criticized (see Meuser and Nagel, 1997, p. 484)
on the grounds that we can also enquire about the everyday knowledge
of people who are of interest as whole persons by using the methods of
narrative or problem-centred interviews. Considerations related to analytic
differentiation also mean that it is hard to see why we should extend the
concept of the expert in this way. In addition, if every individual is by def-
inition an expert, it becomes difficult to interpret situations in which expert
knowledge clearly has specific social effects. Of course, it does not seem
appropriate to treat the difference between laypersons and experts as an
absolute difference (this strict demarcation is increasingly being called into
question, especially in recent work on the sociology of knowledge and the
sociology of science – see Meuser and Nagel, in this volume), but it is no
more productive to adopt a voluntaristic approach which sees itself subject-
ively as emancipatory and critical of authority, but which finally does no
more than attempt to flatten out existing hierarchies by means of an effort
of conceptual will.
(2) The main characteristic of the constructivist definition is its focus
on the mechanisms involved in the ascription of the role of expert, and
it can be divided into a method-relational and a social-representational
approach. The first approach reflects the fact that every expert is also
to some degree the “construct” of a researcher’s interest. In conduct-
ing an investigation, the researcher assumes that the selected expert is
in possession of relevant knowledge about a certain subject (Meuser and
Nagel, 1997, Deeke, 1995). This perspective understands “being an expert”
as something that functions via the ascription of a role by actors who are
interested in information and elucidation, in knowledge of the “objective”
facts. The consequence of this approach for the practice of research is that
one can also look successfully for experts at lower levels of hierarchy within
organizations (see Froschauer and Lueger, in this volume). It is not always
the case (and one could even say that it is rarely the case) that leading figures
who are the public face of an organization are also the experts a researcher is
50 Alexander Bogner and Wolfgang Menz

looking for. In theoretical terms, the method-relational definition reminds

us that expertise is not a personal quality or capacity. However, this over-
looks the fact that there is always and unavoidably a “material subtext”
underlying the method-related construct, since the researcher is never at
liberty to select just anyone as an expert. As a rule, researchers fall back on
those who have established their reputation by publishing in the relevant lit-
erature, who are active in the corresponding associations and organizations,
and who have attained prestigious qualifications and occupy prestigious
positions. This provides researchers with a reasonable degree of certainty
that these are the experts who “really” possess a store of knowledge of rele-
vance to their research. These social ingredients indicate that the relational
approach and the social-representational approach are closely related. The lat-
ter approach states that an expert is anyone who is made into an expert by
societal processes, that is who is seen as an expert in social reality. From this
perspective, the expert can be described as a trained and specialized profes-
sional (an exponent of Weber’s “Fachmenschentum,” Weber, 1980, p. 576),
or as a member of the “functional elite” (Meuser and Nagel, 1994, p. 181).
The social-representational approach reflects the complex processes of def-
inition which underpin “being an expert,” together with all the prelimin-
ary assumptions these require, but its strict focus on the societal parameters
of expertise means there is a danger that a concept of the expert with elitist
implications will be uncritically accepted.
The method-relational and social-representational approaches are so
closely interconnected with one another that any distinction between them
is of primarily analytic value. In research practice, who is to count as an
expert (and who is sought after as an expert) is always defined via specific
research interests and simultaneously through the social representativity of
the expert. The limits of the constructivist position become clear when one
considers that researchers selecting an expert are always practically guided
by the form in which they find the social world, the meanings that have
been structured into that world before researchers engage with it.
(3) The focusing of the expert in terms of the sociology of knowledge has
become very influential in the debate about methods. One reason for this
is to be found in the paradigmatic orientation of the scholars who have
set out the idea of the theory-generating interview as a special form of the
qualitative interview. A second reason is the fact that the debate within soci-
ology about the figure and function of the expert was at a fairly early stage
initiated by sociologists of knowledge, and it has been largely dominated by
these scholars. This certainly does not mean, however, that any consistent
concept of the expert has been employed. What unites those who have been
engaged in this debate is the way they conceptualize experts in terms of the
specific structure of their knowledge.
Schütz (1964) sees the expert as a scientist who acts on the basis of cer-
tain, unambiguous knowledge that can be communicated and reflected on
 The Theory-Generating Expert Interview 51

at any time. However, because the characteristics of the layperson, in the

figures of the “well-informed citizen” who emancipates himself from a “nat-
ural world view” with the help of a reflexive way of thinking, and of the
“straightforward man in the street”, are not very well defined, the figure of
the expert becomes unclear and contradictory. As the well-informed citi-
zen weighs up, in a rational way, the arguments put forward by different
experts involved in a dispute, we can observe a new way of dealing with
knowledge and science in which the first cracks appear in the notion of the
expert as an objective and neutral guardian of the truth. In the following
discussion, both the sociology of science-related aspects of this question
and its implicit relevance to politics, and the theory of democracy were
taken up. Sprondel (1979) tries to characterize expert knowledge as “special
knowledge” in order to make Schütz’s concept of the expert more specific.
This special knowledge is unlike general knowledge in that it includes com-
plex, integrated stores of knowledge and is also related constitutively to the
pursuit of a profession. This enables Sprondel to differentiate between the
expert and the knowledgeable layperson who may have special knowledge
(for example, an amateur DIY enthusiast) but who, because his or her func-
tional context is restricted, is unable to constitute “structurally significant
social relations” (Sprondel, 1979, p. 149).
Criticism has been voiced of the narrowing of the concept of the expert
so as to restrict it to professional activity, in view of the fact that experts
participate in extra-parliamentary social movements on a voluntary basis
(Meuser and Nagel, 1997). Critics have also objected to the idea that what
distinguishes expert knowledge is its reflexivity and explicitness. Sprondel
argues, as does Schütz, that the expert’s special knowledge is ‘immediately
accessible’ while everyday knowledge is diffuse. Meuser and Nagel (1994,
p. 182f.), on the other hand, point out that the basic orientations of the
expert, his or her implicit knowledge, that is to say the unwritten laws and
decision-making maxims that operate in his or her specific functional area
of expertise, are not directly accessible. In terms of method, this means
that one cannot enquire directly about the implicit rules of routine action,
the expert’s habits and traditions; these things have to be reconstructed. In
terms of theory, it means that the definition of an expert which argues via
a differentiation of forms of knowledge needs to be expanded, though not
fundamentally corrected.

Introducing the idea of a special structure or form of knowledge that is

available to the expert is neither theoretically satisfactory nor productive
in terms of method. The first problem here is that an expert who is con-
ceptualized as possessing a specific (additional) store of knowledge seems to
have been detached from the societal conditions of recognition of his or her
expert status. Because this recognition depends on social parameters which
can themselves change, we would argue that it is not in the first instance
52 Alexander Bogner and Wolfgang Menz

actually existing differences in competence that characterize the (sought

after) expert, but rather the social relevance of his or her knowledge. We
shall return to this point later. Secondly, conceiving of expert knowledge
as a “homogeneous body of knowledge”, albeit one that is only implicitly
present, means ignoring statements that can be recognized as those of the
“private person.”6 Thirdly and finally, this static model is combined with a
sociologically unsatisfactory conceptualization of the interaction situation
which sets up an ideal communication process that is “natural” and “free of
distortion” and in which this “entity” is supposed to be made transparent. It
therefore seems to be appropriate to proceed via an analytic differentiation
of expert knowledge in order to solve these problems of method.

2.2.2 Expert knowledge as an “analytic construction”

The idea there is such a thing as “special knowledge” rests on the theoret-
ically problematic premise that the validity and generation of subjective
meanings and orientations can be thought of as strictly divided into the
categories of lifeworld and system. If, on the other hand, we proceed via
a process of analytical differentiation that cuts across the traditional dis-
tinction between everyday and expert knowledge, we can identify three
central dimensions of expert knowledge, which also converge with the dif-
ferent method-related and theoretical claims made by the expert interview:
(1) “technical knowledge,” which contains information about operations and
events governed by rules, application routines that are specific to a field,
bureaucratic competences, and so on. This “technical” knowledge remains
most closely related to the area of knowledge where knowledge provides
a specific advantage, where expert knowledge can be distinguished from
everyday knowledge because it is more systematic and more specific in
its content. This can be distinguished from (2) “process knowledge,” which
relates to the inspection of and acquisition of information about sequences
of actions, interaction routines, organizational constellations, and past or
current events, and where the expert, because of his or her practical activity,
is directly involved or about which she or he at least has more precise know-
ledge because these things are close to his or her field of action. This process
knowledge, unlike technical knowledge, is not really specialized knowledge
in the narrow sense (something one can acquire through educational quali-
fications), but is more a matter of knowledge based on practical experience
acquired from one’s own context of action. The theory-generating interview
seeks to gather (3) “interpretative knowledge,” that is to say the expert’s sub-
jective orientations, rules, points of view and interpretations, which suggest
a picture of expert knowledge as a heterogeneous conglomeration. As we
reconstruct this interpretative knowledge we enter, to put it in old-fashioned
terms, the sphere of ideas and ideologies, of fragmentary, inconsistent con-
figurations of meaning and patterns of explanation. This analytic differ-
entiation makes it possible to describe more precisely the epistemological
 The Theory-Generating Expert Interview 53

interest of the theory-generating expert interview, and it also makes clear

that expert knowledge cannot be understood satisfactorily if it is thought of
as a “homogeneous body of knowledge.”
Expert knowledge only becomes identified as interpretative knowledge
as the data are gathered and the principles of evaluation are applied; its
interpretative character is not a quality of the stores of knowledge them-
selves. In this respect it does not exist as an entity prior to interpretation,
and in this sense interpretative knowledge is always the result of an act
of abstraction and systematization performed by the researcher, an “ana-
lytic construction.”7 This means that the differentiation we have proposed
between three kinds of expert knowledge is not really based on any charac-
teristics of the knowledge itself, but is primarily a construction of the social
scientist doing the interpretation. It is almost impossible to tell, on the basis
of something said in an interview, whether this statement should be con-
sidered “technical knowledge” and so in no need of further interpretation,
or “interpretative knowledge,” that is as the expression of a subjective con-
struction of meaning on the part of the interviewee.
In general terms, “interpretative knowledge” does not constitute a spe-
cific advantage in terms of knowledge available to the expert. In order for
a substantively rich investigation of this “interpretative knowledge” to be
possible, it is therefore necessary to integrate the expert methodologically as
a “private person.” It is only in the phase when the data are evaluated that
it becomes clear whether the relevance structures and patterns of orienta-
tions used by the expert can be reconstructed exclusively by using his or
her explanations given from within the professional context, or whether it
is also necessary to incorporate comments made from the personal sphere.
It is frequently the case that those very passages in an interview where com-
monplaces and pithy sayings from everyday life are mobilized, or arguments
put forward which rely on metaphors from the “private” sphere, are of par-
ticular interest. One can hardly distinguish in practice between the inter-
viewee as “expert” and the interviewee as “private person”, and it makes no
methodological sense to attempt to do this.

2.2.3 The social relevance of expert knowledge

Before examining in more detail the problems of method that are implied
by the sociology of knowledge concept of the expert, we need to raise the
question of the criteria used in practice to single out experts who inter-
est the researcher. In order to answer this question, one can start by pro-
ceeding “reconstructively,” so to speak, and asking why certain people are
interviewed as experts in empirical investigations. In investigations where
the only goal is the gathering of “useful information” (Deeke, 1995), the
criterion of “knowledge” does indeed suffice to determine who is con-
sidered an expert (for example, in the case of Gläser and Laudel, 2004).
However, this does not apply to investigations where the main concern is
54 Alexander Bogner and Wolfgang Menz

the “interpretative knowledge” of the actors interviewed. This is of interest

for a social-scientific investigation not because the expert is, for example,
able to demonstrate access to this knowledge in an especially systematic
or reflexive way or because it provides a particularly accurate reflection of
reality, but because it affects practice to a significant degree. In theory-generat-
ing expert interviews, we consult experts because their action orientations,
knowledge and assessments decisively structure, or help to structure, the
conditions of action of other actors, thereby showing that expert knowledge
has a socially relevant dimension. It is not the exclusive nature of his or her
knowledge that makes an expert interesting for the purposes of an interview
oriented towards interpretative knowledge, but the fact that this knowledge
has the power to produce practical effects. As Beck (1992) has shown in his
analysis of the radical transformation in the meanings of knowledge and
science in conditions where the concepts of nature and society can hardly
be kept apart, what constitutes a post-traditional expert is that she or he is,
by virtue of his or her specific knowledge, politically influential. Experts
can thus be understood as people who, on the basis of specific knowledge
that is derived from practice or experience and which relates to a clearly
demarcated range of problems, have created a situation where it is possible
for their interpretations to structure the concrete field of action in a way
that is meaningful and guides action.8
“Expert” remains a relational concept inasmuch as the selection of per-
sons to be interviewed depends on the question at issue and the field being
investigated by the researcher. When carrying out empirical investigations,
we are not usually dealing with people whose activities are “relevant to the
whole of society,” who shape complex public discourses in a decisive way
and have the potential to exercise a great deal of power and influence in
ways that determine structural conditions spanning society as a whole. This
kind of expert can be found at the centre of discussions about the “Second
Modernity.” Case studies, though, usually involve a concrete investigation
of a smaller unit – a specific organization, company, educational or medical
institution, and so on. This means that the question of who is to count as an
expert for purposes of method always has to be answered in relation to the
concrete field of operation in which the expert acts, and with reference to
the investigative spectrum of the empirical study being carried out.
We are now in a position to offer the following approximate methodo-
logical definition of the concept of the expert. An expert has technical,
process and interpretative knowledge that refers to a specific field of action,
by virtue of the fact that the expert acts in a relevant way (for example,
in a particular organizational field or the expert’s own professional area).
In this respect, expert knowledge consists not only of systematized, reflex-
ively accessible knowledge relating to a specialized subject or field, but also
has to a considerable extent the character of practical or action knowledge,
which incorporates a range of quite disparate maxims for action, individual
 The Theory-Generating Expert Interview 55

rules of decision, collective orientations and patterns of social interpret-

ation. An expert’s knowledge, his or her action orientations and so on, also
(and this is decisive) point to the fact that she or he may become hegemonic
in terms of practice in his or her field of action (for example, in a certain
organizational-functional context). In other words, the possibility exists
that the expert may be able to get his or her orientations enforced (at least
in part). As the expert’s knowledge has an effect on practice, it structures the
conditions of action of other actors in the expert’s field in a relevant way.
This definition emphasizes that the goal of the theory-generating expert
interview is the reconstruction and analysis of a specific configuration of
knowledge. This means that in terms of method, it cannot be treated as noth-
ing more than a qualitative interview with a particular social group. The need
for a methodologically adequate concept of the expert is closely connected
with the problem of the concrete selection of people to be consulted in
the empirical investigation. The broader definition of the expert does not
just leave us with the problem of localizing the specific knowledge that is
relevant to our investigation. It also means that we need to identify those
potentials for power and influence, which mean that the expert’s interpret-
ative knowledge may succeed in becoming hegemonic. In the first case it
is not always leading figures within an organization that have access to
the knowledge that is relevant to the investigation, and real influence is
not automatically correlated with a person’s formal position. Since we often
have little idea in advance either of the distribution of relevant knowledge
or of the power structures within the field of investigation, the selection of
people to be consulted must inevitably be an iterative process. After we have
carried out the first round of interviews, we will have further information
that may help us in selecting our next group of interviewees. However, the
assignment of expert status often cannot be done on the basis of imma-
nent findings from the expert interviews alone; we need to bring into play
further sources of data and methods of investigation (for example further
interviews) before we are in a position to answer the question of how far the
expert knowledge has practical affects and is efficacious.9
We must now see whether the concept of the expert we have proposed
can demonstrate its usefulness in the framework of method-related con-
siderations. This is the subject of the next section.

2.3 Interaction structures in the expert interview

2.3.1 From the archaeological to

the interaction model of the interview
If one compares the relevant contributions to the debate about methods
in expert interviews in respect of what they say about the form of con-
sultation and the style of the interview, one can identify parallel features
which appear repeatedly even when there are differences of detail. As a
56 Alexander Bogner and Wolfgang Menz

consequence of the conceptualization of expert knowledge as a clearly

demarcated, stable and homogeneous body of knowledge, all these con-
tributions orient themselves towards a certain standardized ideal of how
to conduct an interview successfully, a kind of “one best way” of consult-
ing interviewees which follows an “archaeological model” of the interview.
There is a tacit assumption that attitudes, definitions of situations and
action orientations exist which are “true” and “real” and are independent
of context. These phenomena, it is implicitly assumed, have their roots in
a deep stratum of the human psyche, and the task of the researcher is to
use appropriate interview techniques to bring them to the surface in a form
that is as pure as possible. The interviewer is an essential instrument, but no
more than that, and she or he is also a source of mistakes and disturbances,
which obstruct or distort the process of “valid” evocation and reconstruc-
tion of the “genuine” values.10 In accordance with this conception, reflec-
tion on the interview as a social situation within the debate about the
expert interview essentially revolves around the concept of “interaction
effects,” the term used to describe whatever endangers the interaction struc-
ture being striven for and the distortions of and deviations from the ideal
kind of interview that is sought after (see Meuser and Nagel, 1991, p. 449ff.,
Vogel, 1995, Krafft and Ulrich, 1995). The very concept of an “interaction
effect” points to the analytic orientation here towards the model of the
“archaeological interview.” Interactions bring about effects (without them
we would have no interaction, just an indistinct noise without meaning),
and this is seen as a pathology of communication, a defect of the ideal,
distortion-free interview which should be avoided. But the situative effects
in the expert conversation diagnosed as “mistakes” and “distortions” can
be used productively (see Abels and Behrens, 2009); not only that, they are
constitutive for any process in which data are produced.
In the final analysis, the generally propagated ideal of openness in con-
versations and neutrality on the part of the interviewer is also, notwith-
standing the indisputably good reasons that we have for adhering to this
ideal, unable to move beyond the idea that the interviewer can in some
way remain “invisible” and, by not influencing what the interviewee says,
provoke action orientations, attitudes and so on that are as “pure,” context-
independent and situation-independent, as possible. All too often scholars
forget that anything that is said, both in everyday situations and in the
particular situation of the interview, is not just said about some subject in a
social vacuum, but is also said for the concrete interaction partner. Nothing
that is said can be detached from its social dimension, and so is always a
situative statement made in the framework of a specific interaction situ-
ation. And this situation is reflected on by the interviewee, either directly
and consciously or at the level of “practical consciousness” (Giddens, 1984);
the interviewee reacts to this situation and plays an active role in helping
to constitute it.
 The Theory-Generating Expert Interview 57

We therefore need an “interaction model of the interview” as a contrast

with the “archaeological model.” In the following discussion, we would like
to single out one particular dimension of the complex interaction mech-
anisms at work in the expert interview – though this dimension is, in our
opinion, a vital one. If statements are always made in relation to another
person who is perceived in a certain way, the structuring of the interaction
situation is decisively shaped by the “expected expectations,” that is the
conceptions the interviewee forms, on the basis of various indications and
pre-existing knowledge, and also of the experience of communication as it
takes place during the interview itself, of the possible expectations of the
interviewer/researcher. We therefore want to analyze the interaction struc-
tures in the expert interview by looking at the interviewee’s perception of
the interviewer and the competences ascribed to the latter. In doing this,
we assume that whatever is said by the interviewee is in essential respects
guided by what she or he believes or suspects to be the case in relation to
the interviewer’s competence, professional background, normative orien-
tations and attitudes, and possible influence within the relevant field of
investigation. This perspective takes as its object of analysis the so-called
“interaction effects,” which it treats as (necessary) components of the com-
munication structure under examination. This structure is always consti-
tuted with reference to a specific situation; it may be reproduced in the
course of interaction within the interview, but it may also change. Certain
communication structures may, depending on the epistemological interest
and goal of the investigation, be more productive than others, because they
stimulate the interviewee to articulate specific forms and stores of know-
ledge in which the interviewer is interested. Nevertheless, one cannot for-
mulate any general ideal (whatever it might look like in detail) of the “right”
way to conduct an interview, one that would have overall validity regardless
of the particular case. It follows that we must always bear in mind, in gen-
eral and also when evaluating the data, that statements made in an inter-
view are also a function of the communication structure.
We now want to move on to provide a brief sketch of some selected inter-
action situations that are typical of expert interviews and which are shaped
by certain perceptions, ascriptions and suspicions about competence in
relation to the interviewer. These situations are based on our own research
experience,11 and we would like to pose the following questions: How can
one recognize that certain role expectations are operating in an interview
(indications)? What preconditions provide the basis for these expectations?
What are the advantages and disadvantages of this kind of interaction situ-
ation? In what kinds of investigation, and in relation to what kinds of
substantive question, are such interaction situations productive (area of
application)? The typology of these interaction situations we have drawn
up accords with the three dimensions of: the specialist competences attrib-
uted to the interviewer; the suspected convergence or coincidence of the
58 Alexander Bogner and Wolfgang Menz

normative background against which the concrete interaction in the inter-

view takes place; and the perceived potential of the interviewer as someone
with power to act or to exert influence in the field in which the interviewee
is active (see Table 2.1).12

2.3.2 The expert’s perception of the interviewer:

six kinds of ascription
(1) If the interviewer is seen as a co-expert, she or he will be treated as a
colleague, a partner with equal status, someone with whom the expert can
exchange knowledge and information about the specialist field in question.
The interviewee assumes a shared store of knowledge, which she or he and
the interviewer can fall back on without needing to restate it in detail. The
store of knowledge that is assumed to exist is not restricted to technical
and process knowledge, but includes a significant proportion of the (largely
implicit) normative and practical premises of the interviewee’s own orienta-
tions (interpretative knowledge). The expert assumes that the interviewer is
familiar with the practical conditions of his or her actions and shares their
normative implications; there is therefore no need for the interviewee to
justify any of these assumptions in the interview. However, this does not
mean that the interviewee assumes complete congruence between his or
her knowledge and that of the interviewer. In many cases it is assumed that
the interviewer has general competences and knowledge in a generalized
and systematized form, whereas the interviewee has an advantage in terms
of specific knowledge in the concrete case and in the context of his or her
background. This rests on a recognition of the interviewer as a representa-
tive of a university department or institute which is concerned with ques-
tions and subjects similar to those in which the expert specializes.
The standard situation in interviews is an asymmetrical one in which the
interviewer consults the expert. In this case, there is a tendency for this to
fade and for a polarized division of roles into interviewer and interviewee to
be replaced by a more emphatically horizontal communication structure.13
The interviewee him or herself asks questions in order to obtain informa-
tion from the interviewer and to familiarize him or herself with the inter-
viewer’s own position and assessment, engages in substantive debates with
the interviewer, and so on. Not infrequently, the conversation ceases to be a
consultation and takes on the character of a discussion between specialists.
The expert responds to the interviewer’s questions with questions of his or
her own, which in terms of interaction theory have the function of testing
the role expectations and ascriptions of competence that the expert has
formulated in relation to the interviewer. Acceptance of the interviewer as a
co-expert can be seen as a preliminary contribution made by the expert at
the beginning of the interview. It is based initially on vague indications and
impressions, and these are implicitly checked as the conversation develops.
Only in the course of the discussion itself does it become clear whether
 The Theory-Generating Expert Interview 59

the interviewee’s expectations and ascriptions can be confirmed or if they

will have to be abandoned. An interaction structure shaped by granting the
interviewer “co-expert” status usually only becomes firmly established dur-
ing the course of the subsequent conversation. Such a structure therefore
requires a large number of preliminary assumptions, and – especially in the
initial phase of the conversation – needs to be permanently re-established
and confirmed by the interviewer’s behaviour and discussion style. As a
rule, though, this structure does not rest solely on the concrete communi-
cation behaviour of the researcher in the interview, but also requires infor-
mation and impressions that were available in advance such as knowledge
of the professional background and institutional connections of the inter-
viewer, and/or pre-existing personal acquaintance.
In the debate about methods, this kind of interaction structure is seen as
problematic if and when there is a danger of a reversal taking place in the
interview situation: the interviewee tries to bring this about by insisting on
posing his or her own questions in such a way that the researcher does not
manage to ask all of his or her own questions in the usually limited time
available for the interview.14 In addition, the interviewee may make a stra-
tegic decision to pose repeated questions of his or her own in order to give
away as little as possible. The recognition of the interviewer as a co-expert
can also be a disadvantage if the interviewee displays a tendency to retreat
to discussion of specialized, technical scientific aspects of the subject or to
limit his or her comments to specialist academic discourse. If this happens,
the interview will concern itself with technical details, formulae and stat-
istics rather than subjective evaluations, normative goals and the expert’s
own action orientations, and will be of less sociological interest.
In another respect, acceptance of the interviewer as a co-expert can turn
out to be a specific advantage. If the researcher demonstrates his or her spe-
cialist interest in the subject, makes use of his or her own knowledge and
engages in a lively discussion on this basis, the interviewee will be prepared
to do the same and to reveal information and knowledge she or he would
be unlikely to make available if the interviewer’s role were assessed differ-
ently and competence attributed differently.15 What Abels and Behrens
(1998, p. 87) rather disparagingly refer to as “trading information,” and see
as something that endangers the interview, thus acquires a dynamism of its
own that can be used productively. What is more, a conversation situation
that is more strongly oriented towards discussion rather than consultation
is frequently in the interest of the researcher if she or he is able to introduce
the provisional findings of his or her investigation as part of the discus-
sion, take note of critical comments on his or her position, and even use
the discussion on the quiet as a way of helping to validate that position
communicatively.
If the expert interview is seen primarily as something that will prod-
uce “useful information” and elucidation of “facts” (as is the case with
60 Alexander Bogner and Wolfgang Menz

exploratory and systematizing interviews), the high level of specialist inter-

action between co-experts will have a productive effect and the interview
will be of value for the detailed analysis of the issue at stake. If, on the
other hand, the goal of the investigation is the reconstruction of interpret-
ative knowledge (as is the case with theory-generating expert interviews),
the “technicist element” becomes problematic, since the implicit normative
and practical premises of expert opinion will be presupposed as a shared
basis of the conversation between expert and co-expert, and it will be diffi-
cult to gain access to them for the purposes of analysis.
(2) Another attribution of competence we encounter frequently in the
practice of expert interviews is the idea of the interviewer as expert from a
different knowledge culture. What characterizes this attribution is that the
interviewee assumes that the interviewer possesses significant specialist
competence and knowledge (in this respect, it works in a similar way to the
expectation that the interviewer will behave as a co-expert), but takes into
account the fact that the interviewer comes from a different professional
background. The conversation is focused by a shared substantive interest
in the core issues, but it is clear that each participant in the conversation
is arguing against the background of his or her own stores of knowledge,
normative goals and practical obligations to act. The interviewer appears as
an expert, but as the representative of a different discipline rather than as a
specialist colleague.
If the interviewee orients his or her style of answering questions and dis-
cussion to role expectations of this kind, the answers she or he gives will as
a rule be formulated in more didactic terms than in circumstances where
the interviewer is perceived as a co-expert. Less frequent use will be made
of specialist terms from the interviewee’s own discipline, and she or he
will go to greater lengths to make clear the background to his or her own
action orientations. One quite frequently finds that the interviewee behaves
rather uncertain, because she or he is not really sure to what extent special-
ist knowledge about his or her own discipline can be presupposed. In this
kind of interaction structure, the interviewee will typically ask how much
explanation the interviewer requires of the specialist context of his or her
remarks, whether the concepts being used are familiar, and so on.
As a rule, the expert also attempts to do justice to (what she or he imagi-
nes to be) the epistemological interest of the researcher and to adapt his or
her response to the latter’s specialist and disciplinary background. In inter-
views of this type the answers frequently contain introductory passages in
which the interviewee says something like: “I’m sure you’ll find this inter-
esting from a social-scientific perspective...”, or “As a sociologist you may see
this differently, but...”. In this way the difference between the participants’
perspectives is openly acknowledged, which contrasts with the assumption
of shared specialist interest in cases where the interviewer is treated as a
co-expert. One frequently finds explicit references in the argument to the
 The Theory-Generating Expert Interview 61

fact that the views being expressed by the interviewee can be traced back
to his or her own specialist and professional experience, that is the expert
states explicitly that she or he is not assuming any shared horizon of mean-
ing as the basis of the conversation. This also indicates an expectation that
there will be differences between interviewer and interviewee not only with
regard to (assumed) “technical” specialist knowledge, but also in relation to
normative premises.
The role expectations and competence ascriptions we have set out here as
types are, of course, primarily a matter of assumptions made by the inter-
viewee, but as a rule these assumptions are based on interviewees’ actual
experience with interviewers and on the (usually vague) knowledge and
expectations the former have of the latter. For the practice of interviews it
is not always important whether these assumptions are completely accur-
ate. Nevertheless, the “interviewer as representative of a different knowledge
culture” is not only a type of role assessment we encounter frequently in
practice, it also corresponds as a rule to the actual distribution of compe-
tences to a much greater degree than cases where the interviewer is treated
as a co-expert. This means that an interaction structure shaped by the
model of a conversation between experts belonging to different knowledge
cultures requires fewer preconditions. Unlike the “co-expert,” the “expert
from a different knowledge culture” does not permanently have to demon-
strate his or her specialist knowledge, take care to use the correct termin-
ology, and show that she or he is well informed about every detail of the
issue under discussion. Initially, presenting a visiting card from a university
or a respected institute or showing that one has an academic qualification
is usually enough to ensure that the interviewee will form an expectation
that she or he is dealing with an expert in sociology. However, in this case,
too, he or she must contribute actively to sustaining this expectation as the
interview progresses. Here too, signs of ignorance of the expert’s specialist
field of action, repeated recourse to everyday language, and a general lack
of professionalism will cause the interviewee to revise his or her attribution
of competence.
In cases where the interaction situation of an interview is shaped by the
competence assessment of the “interviewer as expert from a different know-
ledge culture,” a range of different dynamics can develop in the course of
the conversation. Ideally, the interviewee reacts in a sensitive, interested
way to the researcher’s social-scientific epistemological interest and orien-
tates his or her responses towards this interest, though without abandon-
ing his or her specialist context as the expert whose knowledge is relevant
to the investigation. Both of the participants in the conversation “accept”
the different models of interpretation, the divergent forms of background
knowledge, and the different normative implications of the two knowledge
cultures. However, there are two quite different ways in which the interview
can diverge from this course. The first possibility is that the interviewee
62 Alexander Bogner and Wolfgang Menz

may feel him or herself obliged or forced to defend his or her own “culture”
(knowledge culture or specialist culture) actively, and to demonstrate the
superiority of that culture. The expert insists on the legitimacy of his or her
own perspective in the face of the suspected epistemological interest (and
possible criticism) of an interviewer who comes from a “different” specialist
world. If things go particularly badly, the conversation develops in the dir-
ection of an allocation of roles in which the interviewer is seen in advance
as a “potential critic.”
If things go differently, the interviewee will go out of his or her way to
demonstrate willingness to respond to questions posed from a social-scien-
tific perspective, and will consciously put aside his or her own point of view
on the grounds that it is probably not very interesting “for a sociologist.”
This conversation situation can become problematic if it goes beyond a pre-
paredness to recognize the epistemological interest of the researcher, which
of course is as a rule productive for the interview, and turns into an exagger-
ated adaptation to suspected expectations or, as all too frequently happens,
if the imagined interest of the social scientist diverges too sharply from his
or her real interest. It is also possible for the interviewee’s preparedness to
accommodate the interviewer to go so far as to take on elements of a “pater-
nalistic” style of communication. In this case, the communication situation
will be hard to distinguish from one in which the interviewer appears as a
“layperson.”
The competence ascription of the “interviewer as expert from a different
knowledge culture” proves advantageous by virtue of the fact that as a rule,
the expert’s specialist patterns of argument and orientations become more
apparent here than when the interviewer is treated as a co-expert, since in
the latter case it is assumed without question that the interviewer shares
them. Theory-generating approaches, or those concerned with the ana-
lysis of interpretative knowledge, are therefore well placed to benefit from
this structuring of roles in interviews. The disadvantage, however, is that
detailed specialist knowledge is rarely made explicit in such interviews. This
interaction constellation is therefore less fruitful for expert interviews of the
exploratory or systematizing type where the primary purpose is the acquisi-
tion of “technical knowledge.”
(3) In the literature on methods, the kind of interaction in expert inter-
views that treats the interviewer as layperson is usually considered to be a
negative example alerting us to dangers, and the result of a failure to con-
duct the conversation in the proper way. Trinczek, for example, argues that
interviewers who want to conduct successful expert interviews with man-
agers must, as an indispensable precondition, have expert status themselves
or, as a minimum requirement, “at least appear reasonably compatible with
and ‘equal’ to” the interviewee in respect of their age and qualifications”
(Trinczek, 1995, p. 65). Vogel (1995, p. 80) complains about the “demon-
strative good nature” of some interviewees, their attempts to show how well
 The Theory-Generating Expert Interview 63

disposed they are, and the efforts of the interviewee to dictate the content
of the conversation to the (seemingly) inexperienced or inferior interviewer,
which “has a lasting effect” by making it difficult to “build up a conversa-
tional atmosphere of reciprocal productivity.” However, instead of reacting
to these “paternalistic effects” (ibid.) with a display of resentment at not
being perceived or taken seriously as an expert in the desired way, inter-
viewers would be better advised to “turn this discriminatory paternalism
to their strategic advantage, as a way of making the collection of data more
productive” (Abels and Behrens, 1998, p. 86).16 Vogel and Trinczek are too
eager to try to live up to what they imagine to be the expectations of their
interview partners; the idea that only older men with doctorates are able to
conduct expert interviews successfully is not very convincing.
If we want to characterize what distinguishes this kind of competence
ascription, and if we want to go on to weigh up its specific advantages and
disadvantages, we need to reflect on the fact that this distribution of roles
is fundamentally ambivalent. This is because the expert can perceive the
interviewer as either a “welcome” or an “unwelcome” layperson. If the inter-
viewer is welcome, the expert acts as someone with the didactic task of
transmitting his or her experience, views, and stores of knowledge. The
interviewer is offered a painstaking introduction to the specialist founda-
tions of the field under investigation and the factual preconditions of spe-
cific action orientations. Because the interviewer is not expected to ask any
further constructive questions in the course of this exposition or, to put it
another way, because everything the expert says is assumed to be equally
relevant, the conversation can easily become a monologue. As this happens,
a situatively generated “pressure to narrate” can be constituted which leads
to deeper levels of the expert’s knowledge. If the lay interviewer is seen as
unwelcome, however, any further questions she or he may ask will be seen
by the interviewee as unwanted interruptions. As a result of verbal and non-
verbal refusal17 of a dialogic form of conversation, the interviewer is forced
to accept the role of a silent, attentive receiver of knowledge. Displaying no
interest in the researcher’s specific requirements, the expert does not say a
great deal and says it quickly. This competence ascription leads to a strictly
hierarchical communication situation.18
There are a number of strengths and problems which are features of an
interaction situation structured in this way and which correspond to the
fundamental ambivalence of this distribution of roles. One the one hand,
naïve questions stand a good chance of producing the most interesting and
productive answers – especially in the framework of a research design that
seeks to generate theory. Once the image of the interviewer as a layperson
has become stabilized in the course of the interview, and if it is also accepted
by the interviewer, this removes a considerable burden from the interviewer
because she or he no longer has to demonstrate his or her own expert sta-
tus or take care to avoid annoying the interviewee. The interviewer has
64 Alexander Bogner and Wolfgang Menz

the freedom to do whatever she or he wants, and can ask questions that
under other circumstances would have endangered the stabilized schema
of expectation. This can make it possible to gain access to information that
might not otherwise be revealed, particularly because a naïve interviewer
is seen as especially trustworthy (see Abels and Behrens, in this volume).
In this situation, experts see hardly any danger that the interviewer will
be able to make any strategic use of the information obtained, ask supple-
mentary, excessively critical questions, or put the interviewee in a position
where she or he will have to justify his or her position; experts therefore
act more freely and informally. If the interviewee thinks she or he needs to
explain the most basic elements of his or her ways of thinking and acting,
this can be of great interest for analyses of interpretative knowledge because
even simple patterns of argument that are not usually made explicit by the
expert will be set out in detail. Things are quite different when the inter-
viewer is perceived as a “co-expert,” and they do not go as far as this when
the interviewee considers him or herself to be dealing with an “expert from
a different knowledge culture.”
The disadvantages of the competence assumption to the effect that the
interviewer is a layperson are obvious: interviewees sometimes bore research-
ers with interminable monologues about trivia or things they already know,
they plod through the contents of textbooks, or they retreat to common-
places.19 There is hardly any likelihood that difficult specialist issues can
be clarified, it is easier to ignore supplementary questions, and there is a
tendency for the interviewee to take over the structuring of the course the
interview takes (see Gillham, 2000, p. 82). We have already mentioned, in
connection with the perception of the interviewer as an “unwelcome lay-
person,” the danger that the interview will be a short one if the interviewer
is not able to demonstrate his or her own competence. In addition, it is very
difficult to challenge and correct this kind of competence ascription. And,
last but not least, being treated as a layperson is an unpleasant experience
for the interviewer, who feels the expert has not understood him/her and
has underestimated him/her.
(4) The antithesis of the type just described is the perception of the inter-
viewer as an authority. There are two variants of this. In the first variant, the
interviewer is seen as a superior specialist in the field, who seems to be trying
to find out whether the interviewee possesses appropriate knowledge in his
or her field of action or is acting “correctly.” The interviewer is seen as an
expert who has come down from a higher sphere, equipped with the insig-
nia and academic competences of someone who belongs to a university or
other institute; this expert descends to the lowlands of practice in order to
pass judgment on a colleague with inferior qualifications, or at best to find
out how those involved in practical affairs see things. In this first variant,
the assumption of superiority relates in the first instance to specialist com-
petences. In the second variant, where the interviewer is seen as an evaluator,
 The Theory-Generating Expert Interview 65

it is the power dimension where the interviewer is considered to be superior.

In this case the researcher is seen as an envoy dispatched by a higher author-
ity, someone who wants to examine the interviewee’s organization and on
whose judgment the future of the organization or the interviewee depends.
The interviewee’s assessment is that the researcher has been sent to carry
out tasks such as monitoring the efficiency of a measure, checking on the
use of funds that have been allocated, or asking critical questions about
the interviewee’s qualifications. The interviewee feels placed in a position
where she or he must justify him or herself, wants to present him or herself
in the best possible light, and seeks to legitimize his or her actions and views
in the most perfect way possible in the eyes of the (supposed) authority.20
The interaction situation in which the interviewer acquires an authorita-
tive function in this way is characterized by mistrust about the confidential
use of the information provided, and by what seems to be exaggeratedly
positive self-promotion or, on the other hand, deliberate obsequiousness
towards the researcher. This expectation (or fear) of the researcher’s role can
frequently be observed in the fields of action of experts who are under par-
ticular political pressure to legitimize their work, for example in connection
with pilot schemes where success does not seem assured, with controversial
methods of treatment in social-pedagogical practice, and so on. A percep-
tion of the interviewer as evaluator means that one of the most important
criteria for a successful interview cannot be met – the assumption that it will
have no social repercussions. At least, this criterion is not met in the mind
of the interviewee, and that is what counts. There is an absence of trust as a
basis for spontaneous and honest answers. Problematic or “critical” facts are
not mentioned, and there is no way of getting at them by asking skilfully
phrased questions in the interview, because as a rule there seems to be so
much at stake from the point of view of the interviewee – future financial
support, admitting that one’s own concepts may have failed, or simply the
honour of one’s profession. But although there is hardly any point trying to
set up such interaction situations, interviews conducted on these lines are
not altogether without value. For example, one can learn a great deal about
the interviewees’ legitimation and self-justification strategies, the patterns
of argument they use in pursuit of their interests, and strategies employed
to immunize one’s own position against internal and external criticism. The
interviewer is no longer in a neutral position, no longer an external obser-
ver. She or he becomes a target and object of strategic ways of acting by the
actors in the field under investigation.
(5) If the interviewer is seen as a potential critic, there is an assumption that
she or he is not capable of judging the facts and issues at stake in the inter-
view in an objective way that does justice to the subject. The interviewer is
not considered to be a conscientious researcher with a value-neutral epis-
temological interest, but is seen as the ideologically prejudiced representative
of a certain undesirable worldview. The assumption is that she or he cannot
66 Alexander Bogner and Wolfgang Menz

be trusted because she or he wants to use the findings of the interview stra-
tegically for political or personal purposes rather than to place them at the
disposal of “neutral” science. The interviewee assumes that there is a fun-
damental divergence between the normative assumptions underlying the
investigation and his or her own principles. Even though the “interviewer
as expert from a different knowledge culture” may not necessarily share the
normative assumptions that prevail in the expert’s specialist field either, she
or he can still be reasonably sure that the interviewee will treat the “different
culture” with interest and respect. In the case of the “interviewer as potential
critic,” though, the result is rejection of the interviewer, which can even go
as far as concealed or open hostility. The expert feels that the interview ques-
tions are a form of criticism and believes that the integrity of his or her func-
tion or even of his or her person is being questioned. This becomes obvious if
the interviewee says things like: “You can think what you like, but we as rep-
resentatives of the management have to think of the economic health of the
company,” or: “Let me tell you, if you are not a practitioner with responsibil-
ity for dealing with such questions, it’s easy to adopt a moralising tone.” It is
also possible for the first indications of this attitude to emerge in the shape
of insistent critical questions about the influence of funding bodies on the
investigation, if this amounts to a suggestion that what is going on is not a
scientific investigation but something where the findings have already been
decided in advance. There is no clearly defined boundary between this kind
of response and paternalistic ways of treating the interviewer as a layperson.
If the interviewee suspects that the difference between his or her norma-
tive frame of reference and that of the interviewer is only caused by the lat-
ter’s ignorance and incompetence, the interviewee often argues in a manner
somewhere between benevolence and condescension (in much the same way
as when the interviewer is seen as a layperson), on the basis of the conviction
that his or her words will be able to provide some improving knowledge for
“the sociologist.” If the interviewer is suspected of being a potential critic,
sociologists (as representatives of a profession that does not always enjoy a
reputation for ideological neutrality) will have a difficult time – especially
when investigating politically or ethically controversial fields of action (for a
good example of this from research in a normatively sensitive specialist field,
see Bogner, 2005a, p. 105ff.).
It hardly needs saying that if this is the interviewee’s role expectation
of the interviewer, it will almost always be disadvantageous for the inter-
view. This interaction situation is characterized by limited preparedness to
answer questions because of a lack of trust, the interviewee’s desire to get
the interview over with as quickly as possible, and a reluctance to do any-
thing to support the research project (for example by putting the researcher
in touch with further potential interviewees). There are some possible bene-
fits, but they are limited to the way in which the interviewee, if she or he
feels his or her status as an expert is being challenged, usually spends more
 The Theory-Generating Expert Interview 67

time seeking to legitimize his or her own action orientations, views and
interpretations, so that the normative premises of the argument which usu-
ally remain implicit become more clearly visible.
(6) The role assessment of the interviewer as a potential critic rests on the
assumption that there is a divergence between the normative background of
the investigation and the corresponding implications of the expert’s profes-
sion or field of action. The core of the judgment that the interviewer is an
accomplice lies in the assumption of identity between the normative orienta-
tions of the interviewee and those of the interviewer. In this case, the inter-
viewer is seen as a comrade-in-arms in a field of action where power struggles
are going on. This turns him/her into someone who is seen as particularly
worthy of trust, so that concealed strategies will be explained, and confiden-
tial information will be revealed. The atmosphere of complicity is as a rule
created via the definition of a shared adversary (see Hermanns, 2004). The
interviewee explains quite candidly what is directly normative about his or
her ways of acting and patterns of argumentation and what is factual or purely
strategic. The interviewer becomes an intimate accessory in these disputes, and
the implicit rules according to which they are conducted are disclosed. This
happens because the interviewee is convinced that the interviewer is com-
pletely honest and discreet. In this respect, this role assessment on the part
of the interviewee requires that a large number of preconditions be met. As
a rule, interviewer and interviewee need to have been personally acquainted
before the interview, so that the interviewee is already informed about the
researcher’s normative views – for example, as a result of their participation in
politically oriented organizations, or via knowledge of the researcher’s publica-
tions in the field, or even as a result of personal friendship. The fact that the
researcher belongs to a university, an institute, or a profession, which often
gives the interviewee enough guidance for him/her to be able to identify the
interviewer as a “co-expert” or “expert from a different knowledge culture,” is
not enough on its own as long as it does not say enough about the researcher’s
normative orientations. The existence of this common ground is expressed
symbolically in the interview, for example by informal und personal codes of
communication, reference to shared experiences, and so on.21
The assessment that the researcher is an “accomplice” is an incalculable
advantage for the interview. The interviewer gains access to confidential
information, she or he can build on the high level of openness and honesty
of the answers, and she or he is given insights into real strategies and action
orientations that go well beyond official aims and objectives or legitimation
patterns. However, there is also a problematic element: the normative prem-
ises must remain largely unstated. Tacit agreement on the common ground
uniting interviewer and interviewee means that one cannot question this
situation by anything one says in the interview; if one did this, the relation-
ship of trust that has been stabilized over a long period would be unilat-
erally brought to an end (see Table 2.1).
68 Alexander Bogner and Wolfgang Menz

Table 2.1 Typology of interaction situations and interview strategies

Dimension of Indications of the (Ascribed)

typification communication preconditions on the
situation interviewer’s side
Interviewer as specialist symmetrical mastery of the
(1) co-expert competence interaction specialist vocabulary
(2) expert from (of same type situation; numerous (esp. 1), specialist
a different [1], of equal counterquestions knowledge,
knowledge culture value [2]) asked by the institutional
interviewee background, academic
titles

(3) Interviewer as specialist asymmetrical status of interviewer

lay-person competence interaction is lower than that
(low level) situation in favour of interviewee;
of interviewee: interviewer is not
monologues delivered a specialist in this
by interviewee, particular field
demonstratively
good-natured
manner; paternalism
(4) Interviewer as “evaluator:” asymmetrical institutional
authority power, “superior interaction background:
expert:” (greater) situation in favour interviewer has
specialist of the interviewer; status of a specialist
competence interviewee’s authority or a
legitimation strategy significant position
in terms of political
power
(5) Interviewer as normative secret knowledge is personal acquaintance,
accomplice background revealed, background of shared
(shared) interviewee speaks experience (for
in a “personal” example membership
way (for example of political
by addressing organizations)
the interviewer
informally)

(6) Interviewer as normative rejection of the interviewer is publicly

potential critic background interviewer, brief known to be a
(divergent) replies, critical “critic;” institutional
counter questions, background in
expert anticipates organizations the
the interviewer’s interviewee does not
questions approve of
 The Theory-Generating Expert Interview 69

Interview and Possible Possible Main area of

question style advantages disadvantages application

oriented towards high level interview exploratory or

dialogue, repeated of specialist remains within systematizing expert
supplementary knowledge, interviewee’s interviews;
questions, rapid high density professional frame investigations mainly
exchange of of facts (1, 2), of reference; concerned with
questions and more explicit “technicist gathering facts and
answers, “trading setting out of element” data
information” grounds and
orientations (2)
interviewer is high level of difficult to guide theory-generating
primarily a receiver trust on the interview in expert interview;
of knowledge, part of the desired direction investigations
questions generate interviewee, oriented towards
narrative, pressure is taken interpretative
committed off interviewer knowledge
but naïve
supplementary
questions
authoritarian style of increased rule that the not to be
questioning, critical self-presentation interview will recommended
supplementary of the have no social as an interview
questions, interviewee repercussions situation; sometimes
interviewer is contravened; unavoidable in the
interrupts “critical” issues are case of evaluations
interviewee not addressed

interview is high level of normative premises exploratory,

conducted in a trust on the are not revealed systematizing and
“personal” style and part of the theory-generating
everyday language interviewee; expert interviews;
is used; repeated access to investigations
confirmation of confidential where the object is
common ground; information technical and process
range of different knowledge
kinds of question
possible
critical or normative danger that the not recommended as
tendentious premises are set conversation will an interview
interview questions; out at length be broken off situation; this
no verbal or situation can
non-verbal come about in
confirmation of the investigations of
interviewee’s status ethically or politically
controversial fields;
sometimes productive
in investigations
oriented towards
interpretative
knowledge
70 Alexander Bogner and Wolfgang Menz

2.3.3 Farewell to the ideal of neutrality:

on the methodological advantages of the interaction typology
Needless to say, the typology we have sketched of interaction situations in
the expert interview is by no means exhaustive. The types it constructs are
compressed, extreme cases, which implies that they rarely appear in their
pure forms in the practice of interviews. It is possible for combinations of
specific competence and power ascriptions to arise in accordance with the
three dimensions in which types are constructed (for example through a
perception of the interviewer as layperson and evaluator), and also for ascrip-
tions related to competence and to normative orientations to be combined
(for example the interviewer as co-expert and accomplice). Furthermore, the
interaction structure of the entire course of a concrete interview can rarely
be classified as belonging to one or more particular types. What is more
likely to happen is that in the conversation process, expectations and attri-
butions that were initially based on vague suspicions are either stabilized or
revised.22
We have constructed these types not just as a contribution to descriptive
analyses of interaction situations that occur empirically. When we inter-
pret data collected in interviews, we need to reflect that the data cannot
be understood as an expression of abstract, general “expert knowledge” but
also as a variable produced by the interaction, and that the interviewee’s
statements are responses to a person seen as having concrete competences
and interests – and thus as statements that would have been different had
the interviewee had a different conception of the interviewer. Secondly,
certain ascriptions of competence, and so on can be deliberately provoked
and used strategically as a way of pursuing the researcher’s own interest in
the investigation. Which role expectations and competence ascriptions are
dominant depends on a range of factors – the interviewer’s age and sex, spe-
cialist knowledge as evidenced by an ability to use specialist terms, linguis-
tic competence, institutional background, academic titles, and so on. The
most significant factor, though, is the way in which the researcher presents
him or herself and his or her research interest, both in the interview itself
and in preparatory contacts with the expert. Of course, it is only the last of
these elements that can be influenced by the researcher him or herself.23
In addition, it is hardly possible for a researcher to outline in advance, on
the drawing board, the “ideal” role expectation for a given project (and it
will often be the case that the different interaction situations described
will be desirable at different stages of the investigation). But it certainly
makes sense to “play” with the role expectations and to employ different
self-presentation and conversational strategies in different interviews con-
ducted during an investigation, as a way of finding out which of these ways
of proceeding is best suited to one’s own epistemological interest. This will
not always be the “co-expert” self-presentation, which is usually considered
the only productive one.
 The Theory-Generating Expert Interview 71

We frequently notice in interviews that interviewees attempt to use

vague indications to build up a picture of the interviewer and, on this
basis, decide what discussion style and arguments they will use (and often
miscalculate). When this happens, the interviewer serves as a screen onto
which imagined expectations and counter-expectations are projected.
Alternatively, the interviewee may react to an unspecific interview situ-
ation by giving a “rhetorical interview” (Meuser and Nagel, 1991, p. 451)
containing nothing more than the standard aims and objectives of the
expert’s field of work, which are already well known to the interviewer. It
is therefore a good idea to provide interviewees with some pointers so that
they can form an impression of the interviewer. Being explicit about one’s
own position, that is both about one’s epistemological interest and about
one’s view of the relevant issues, can be far more productive for a success-
ful conversation than playing hide-and-seek. This places a question mark
over the ideal of the interviewer whose attitude is somewhere between
neutrality and empathy, an ideal, which is held to be applicable to qualita-
tive interviews in general and to expert interviews in particular. 24 Where
expert interviews are concerned, we are not usually dealing with people
who need to be encouraged to express their own opinions and views and
offered support so that they can seize the opportunity to speak freely. In
expert interviews, “getting the interviewee to speak” is not normally a
problem – and if it is, this is not usually because the expert cannot articu-
late his or her own position or because she or he is excessively respectful
towards the interviewer, who “comes from research.” Because of the prac-
tical work they do every day, experts are used to adopting controversial
positions and defending these positions against critical objections. The
complexity of the interaction strategies used to anticipate this specific
understanding of the expert’s own role and the particular competences
of experts, and the extent to which we are able to differentiate these strat-
egies, combine to distinguish the expert interview from other procedures
that aim, in a similar way, to reconstruct subjective action orientations
and implicit decision-making maxims. And there is an additional objec-
tion, which carries even more weight: neutrality in interviews is, in the
end, not credible. In expert interviews the interviewee knows very well
that the interviewer has already made a detailed study of the subject of the
investigation and has formed an opinion of his or her own (see Kaufmann,
1996). Insisting on a claim to be neutral looks more like an attempt to con-
ceal one’s own position in a situation where there is no serious possibility
of being “genuinely” neutral. Moreover, the interviewer can show his or
her own “commitment” in respect of the substantive issues in such a way
that the interviewee is encouraged to respond by expounding his or her
own stores of knowledge and information.25
We have not constructed this typology of interaction constellations as
a way of identifying a communication ideal that will be binding for “the”
72 Alexander Bogner and Wolfgang Menz

expert interview. The decisive consideration should be that the rules accord-
ing to which the conversation is conducted must be drawn up in relation to
the actual or desired role expectations and ascriptions of competence. What
one expects from a “co-expert” is an interview style oriented towards spe-
cialized discussion; in order to sustain the role expectation of “accomplice”,
the repeated confirmation of common ground is needed; the “critic” is per-
mitted to ask waspish counterquestions now and then; the “layperson”,
unlike the co-expert, need have no qualms about making general appeals to
the expert to say something.

2.4 Conclusion: the methodological pluralism

of the expert interview

As we argued in the first section of this article, it only makes sense to debate
whether or not the expert interview can justifiably be considered a distinctive
and autonomous method for the collection of data when we are concerned
with the theory-generating expert interview. The methodological specificity
of this kind of interview, though, does not lie in “the expert” as an object of
research, but rather in the researcher’s interest in a specific configuration
of knowledge. This configuration is characterized on the cognitive level as
a conglomeration of subjective, not necessarily consistent orientations and
patterns of explanation (“interpretative knowledge”), and at the social level
as determinants of action for others (the efficaciousness of the expert know-
ledge). This concept of the expert interview (in terms of the sociology of
knowledge, as a specific kind of reconstructive interview) signifies a rejection
of a concept of the expert that either treats experts as no more than products
of the researcher’s interest or defines their function solely with the help of a
special form of knowledge. In the second section of the article, we therefore
proposed a reformulation of the constructivist concept of the expert as used
in the sociology of knowledge. The expert should be seen as a person who
disposes of, or is believed to dispose of, particular competences, and who con-
sequently has a social status, or exercises a function, which places him/her in
a position where she or he may be able to gain general acceptance for his or
her action orientations and situation definitions. At the same time, there are
grounds for criticism of the widely accepted model of expert knowledge with
regard to what it says about the specific epistemological interest that oper-
ates in theory-generating expert interviews. From the perspective of a recon-
struction of “interpretative knowledge,” expert knowledge should be seen as
an “analytic construction” rather than, as has been the case up until now,
“special knowledge.” This redefinition, which removes the close connection
between expert status and exclusive stores of knowledge, makes it possible
to go beyond conventionally accepted rules for conducting interviews which
are theoretically unsatisfactory and methodologically inadequate. Giving up
the idea that expert knowledge is a homogeneous entity is connected with
 The Theory-Generating Expert Interview 73

criticism of an unsatisfactory conceptualization of the interaction situation

which assumes that the expert knowledge the researcher is interested in can
be dug up like a buried treasure, in as intact a condition as possible. The third
section of the article focused on a critique of this kind of “archaeological
model,” which is closely related to the dogmatic view that there is a single,
correct way of conducting interviews, and went on to present an alternative
outline of an “interaction model.”
We have argued against this ideal of how to conduct an interview, which
is implicitly fixated on the view that there is “one best way” of doing so,
in favour of a range of different interview strategies, which are all seen as
equally valuable. Which of these strategies is appropriate in different specific
circumstances depends on the competence of the interviewer and the inter-
est being pursued in any given investigation. However, the idea that each of
these strategies is equally valuable does not mean that anything goes: one
cannot use just any interaction strategy to open up the field of analysis in
which one is interested. What “equal value” means here is simply that the
complexity of expert knowledge can be opened up with the help of a variety
of (competing) approaches. Each interaction strategy, though, is connected
with a certain type of analytic goal. The (re)construction of interpretative
knowledge in expert interviews depends on the interaction strategies, and it
never identifies more than one segment of the expert’s knowledge.
It is time to abandon the idea that we can move step by step to identify a
perfect communication situation in which distortion is reduced to an abso-
lute minimum. In the end, this notion supposes that we can produce data
in laboratory-like conditions. We need to appreciate that the interaction
effects, which are usually treated as distortions, are constitutive and even
productive components of the course of every interview. The specific pre-
suppositions, expectations, and reactions that pre-structure and accompany
every conversation should be interpreted as situation-specific and person-
specific ascriptions which can be used strategically in anticipation and can
be used reflexively when the data are evaluated.
Our alternative outline of an “interaction model” is thus an attempt to
rethink the debate about the methodology of the expert interview. There
can be no single, unified set of methodological rules for qualitative inter-
views; we can only have different rules for different kinds of qualitative
interview. Nor is it possible to draw up a unified canon of methods for use
in expert interviews. A methodology of the expert interview must bear in
mind that interaction structures in such interviews are bound to diverge,
and we should adopt the principle of “pluralist methodology.”

Notes
1. See Bourdieu 1998 on the theoretical foundations of an observer position that is
not dependent on the cult of the subject or the mysticism of structure.
74 Alexander Bogner and Wolfgang Menz

2. On the distinction in research logic between “contextual knowledge” and “oper-

ational knowledge,” see Meuser and Nagel, in this volume.
3. We are unable to deal in more detail here with the theoretical and practical aspects
of this topic guide. For more on this, see Meuser and Nagel, 1991, p. 448f., 1997,
p. 486f., Deeke, 1995, p. 18f., Gläser and Laudel, 2004, p. 59ff., Helfferich, 2005,
p. 158ff. For an early warning about the danger of using the topic guide in a sche-
matic, inflexible way, see Hopf, 1978.
4. This includes cases where the analytic purpose of an investigation of the subject-
ive dimension of meaning is the reconstruction of collective orientations and pat-
terns of interpretation.
5. Meuser and Nagel (1997) claim that the expert interview can be distinguished
from “problem-centred” and “focused interviews” in respect of the way the con-
versation is conducted and analysed, but they provide no systematic account of
these differences. It seems to be relatively easy to distinguish the expert inter-
view from the focused interview (Merton and Kendall, 1946) on the basis of the
latter’s closeness to quantitative methodology, which follows from its deductive
orientation. The epistemological logic of the problem-centred interview (Witzel,
2000), on the other hand, is a combination of induction and deduction, which is
comparable to the logic of the expert interview in that both are guided by a topic
guide. In the end, though, the difference between the procedures lies in the role
of the interviewee in the conversation, which is determined by specific epistemo-
logical interests and so is more a matter of practical research requirements than of
criteria related to method.
6. Meuser and Nagel (1991, p. 450) treat “private” comments made by the expert as
an indication that something is going wrong with the interview discourse, but in
more recent work they have revised this position (Meuser and Nagel, 2009).
7. Our argument that expert knowledge should be seen as an “analytic construction”
is put forward as a way of sharpening the methodological profile of the theory-
generating expert interview. No problem arises in applying the concept of “special
knowledge” to the exploratory or systematizing expert interview. Indeed, it would
make no sense to conduct these “informative” expert interviews if one did not
assume that the expert involved had some kind of specific store of knowledge or
advantage in this respect.
8. This aspect emerges occasionally in the debate about method in the shape of the
creation of converging concepts, which stress the interpretative role of the expert.
Hitzer, Honer and Maeder (1994) speak of “the institutionalized competence of
being able to construct reality”, and Sprondel (1979) speaks of “the constitution of
relevant social relations”; from the perspective we are developing systematically
here, these concepts refer to the power of expert knowledge to shape outcomes.
9. The theory-generating expert interview, unlike for example Witzel’s problem-
centred interview, need not always be combined with a specific investigative
design. It is more likely to be the case that the position of the expert interview
changes during the course of the investigation. If the reconstruction of experts’
orientations is an independent goal of the investigation, expert interviews may be
the only empirical material needed (and the selection of people to be interviewed
can be made according to formal criteria). If, on the other hand, the expert inter-
views are used in the framework of case studies where the goal is an analysis of
organizational orders, it is self-evident that they can only constitute part of the
study. After all, no concrete field of action is structured exclusively by the orienta-
tions and knowledge of experts.
 The Theory-Generating Expert Interview 75

10. This model, which distinguishes between a desirable (as defined by traditional
quality criteria) but never achievable ideal and the way in which this ideal is
never realized because of situative effects, predominates even in cases where the
concrete structures of action and communication are examined more closely
in the (quantitative) interview, for example in Lueger, 1989. For a critique of
the “archaeological model” in relation to standardized surveys based on an con-
versation analysis, see Houtkoop-Steenstra, 2000. See also Kvale’s metaphorical
description of conventional conceptualizations of the knowledge sought after in
qualitative interviews as “buried metal....[which] is waiting in the subjects’ inter-
ior to be uncovered, uncontaminated by the miner” (Kvale, 1996, p. 3). However,
Kvale does not go on to draw any significant conclusions for the interview strat-
egy or the analysis of the interaction situation.
11. Our own empirical work, which provides the basis for the development of this
typology, has been largely in the fields of labour and industrial sociology and
research on expertise and science (for example Bogner, 2005a, 2005b, Bogner
and Menz, 2005, 2006, Bogner and others, 2008, Menz and others, 2003, Menz,
2009).
12. See Martens and Brüggemann 2006 for a distinction between kinds of expert in
the interaction constellations of the interview, which is developed according to
the dimensions of the expert’s communication style and intentions.
13. The symmetrical relationship should be seen as a special case which character-
izes this type of expert interview. Horizontal interaction structures are by no
means typical of the expert interview as such, as is misleadingly suggested by
Köhler’s analysis (1992).
14. Abels and Behrens (2009) and Vogel (1995, p. 80) describe this as an unwanted
“feedback effect.”
15. Kaufmann, for example, argues that the interviewer’s own “commitment”
(rather than the neutrality and restraint we are usually told is necessary) is an
important criterion for the success of the “understanding interview:” Only to the
extent that the interviewer gets involved will the interviewee do the same and be
prepared to reveal everything he knows (Kaufmann, 1996). See Trinczek’s contri-
bution to this volume.
16. Of course, this does not mean avoiding any analysis of discriminatory pater-
nalism in the interview situation, and it certainly does not mean one should
approve of it.
17. Disapproval expressed through shaking of the head or dismissive hand gestures
as reactions to the interviewer’s objections or supplementary questions are cer-
tainly extreme forms of refusal. One frequently finds, though, that the expert
expresses his or her “internal emigration,” his or her retreat to sacrosanct regu-
latory knowledge that is not accessible via dialogue, by means of a setting that
excludes the interviewer as a partner in the conversation: the expert presents him
or herself in profile, speaks while looking out of the office window, and so on.
18. For an empirical illustration of this kind of interaction, see Bogner, 2005a,
p. 109ff.
19. Martens and Brüggemann (2006, p. 10) also indicate that there is a danger of
experts who think they need to instruct the interviewer using over-simplified
examples, so that it frequently becomes unclear whether these are real cases or
imaginary ones constructed for pedagogical purposes.
20. Even though this assessment of roles occurs in practice, as a rule, predomin-
antly in cases of evaluation (see Leitner and Wroblewski, 2009, on this point),
76 Alexander Bogner and Wolfgang Menz

individual elements of this type are encountered fairly frequently in practice.

For example, we have in our own interviews encountered fears that the inter-
viewer is there as an evaluator on the part of members of factory committees
who suspected they were being checked on by the trade union, and on the part
of interviewees in state-run hospitals who were afraid of rationalization meas-
ures and perceived the interviewer in his function as a representative of an insti-
tution close to the government.
21. Constructions of common ground in interviews can, however, have other
causes, which have nothing to do with shared normative convictions. They
may be caused by such things as a shared regional background or the fact that
two people speak the same dialect. The reverse also applies: diverging language
conventions can prove to be invisible social barriers in an expert interview.
22. Typical examples of this are interviews where the interaction structure is initially
shaped by the role expectation of “the interviewer as critic,” after which the
interviewee revises this view and starts to see the interviewer as a representative
of a different knowledge culture. Another possibility: the interviewee finds that
his or her initial expectation that the researcher is a co-expert is not confirmed,
and shifts for the rest of the conversation to the model of the interviewer as lay-
person.
23. Interviewees will not normally be deceived for long by attempts on the part
of researchers to feign possession of specialist knowledge simply in order to be
perceived as a co-expert. However, this does not mean that this type is neces-
sarily the most difficult one to construct, since a 60-year old professor is equally
unlikely to be able to play convincingly the part of a clueless layperson. Needless
to say, there are considerations of research ethics that set limits to the role
expectations that can be “selected.” It would obviously be quite unethical for the
researcher to pretend to be a reliable “accomplice” when this would amount to
misleading the interviewee.
24. See Scheuch, 1967 and Koolwijk, 1974 for the initial and influential conceptions
of interview styles to be used in qualitative investigations in Germany; for more
specific treatments relating to the methodology of qualitative social research,
see Hoffmann-Riem, 1980 and Lamnek, 1995, p. 21ff. For approaches, which
explicitly challenge the postulate of neutrality, see Douglas, 1985, Fontana and
Frey, 1998 and Holstein and Gubrium, 1999.
25. Reference is frequently made to the “iceberg effect” (Vogel, 1995, p. 79, Abels and
Behrens, 2009), which means that the interviewee seems to be unenthusiastic
and uninterested, and is reluctant to give much away. In practice, the reason for
this is frequently to be found in ways of conducting the interview that lack com-
mitment and any self-positioning on the part of the interviewer. On the other
hand, one should not forget that the expert status of the interviewee also means
there are limits beyond which she or he cannot go in the forms of argument
used. The expert knows that she or he is a quasi-public representative of his or
her profession, field of action, organization, and so on, and is not in a position
where she or he can break their rules of discussion and self-presentation. It can
sometimes be more difficult for experts than for other interviewees to admit to
inconsistencies, give accounts of mistakes that have been made, or express “devi-
ant opinions.” In addition, an expert often feels obliged to reaffirm his or her
expert status, and tries to avoid saying anything that sounds “banal” or every-
day. In this respect even the guarantee of anonymity, which the researcher must
of course provide, does not usually help.
 The Theory-Generating Expert Interview 77

Further readings
Flick, Uwe (2006) An Introduction to Qualitative Research, 3rd edn (London and others:
Sage Publications), esp. pp. 147ff.
Fontana, A. and Frey, J. H. (1998) “Interviewing. The Art of Science” in Denzin, N. K.
and Lincoln, Y. S. (eds) Collecting and Interpreting Qualitative Materials (Thousand
Oaks and others: Sage Publications), pp. 47–78.
Kvale, S. and Brinkmann, S. (2009) InterViews: Learning the Craft of Qualitative Research
Interviewing, 2nd edn (Thousand Oaks and others: Sage Publications).

References
Abels, G. and Behrens, M. (1998) “ExpertInnen-Interviews in der Politikwissenschaft.
Das Beispiel der Biotechnologie” in Österreichische Zeitschrift für Politikwissenschaft,
vol. 27 , pp. 79–92.
Abels, G. and Behrens, M. (2009) “Interviewing Experts in Political Science. A Reflexion
on Gender and Policy Effects Based on Secondary Analysis”, in this volume.
Aichholzer, G. (2009) “The Delphi Method: Eliciting Expert’s Knowledge in
Technology Foresight,” in this volume.
Beck, U. (1992) Risk Society: Towards a New Modernity (London: Sage Publications).
Berger, P. L. and Luckmann, T. (1966) The Social Construction of Reality: A Treatise in
the Sociology of Knowledge (Garden City, NY: Anchor Books).
Blumer, H. (1969) Symbolic Interaction: Perspective and Method (Englewood Cliffs:
Prentice Hall).
Bogner, A. (2005a) Grenzpolitik der Experten. Vom Umgang mit Ungewissheit und
Nichtwissen in pränataler Diagnostik und Beratung (Weilerswist: Velbrück).
Bogner, A. (2005b): “Moralische Expertise? Zur Produktionsweise von
Kommissionsethik” in Bogner, A. and Torgersen, H. (eds) Wozu Experten?
Ambivalenzen der Beziehung von Wissenschaft und Politik (Wiesbaden: VS-Verlag),
pp. 172–93.
Bogner, A., Littig, B. and Menz, W. (eds) (2005) Das Experteninterview. Theorie, Methode,
Anwendung, 2nd edn (Wiesbaden: VS-Verlag).
Bogner, A. and Menz, W. (2005) “Bioethical Controversies and Policy Advice: The
Production of Ethical Expertise und its Role in the Substantiation of Political
Decision-Making” in Maasen, S. and Weingart, P. (eds) Democratization of Expertise?
Exploring Novel Forms of Scientific Advice in Political Decision-Making. Sociology of the
Sciences, Vol. 24 (Dordrecht: Springer), pp. 21–40.
Bogner, A. and Menz, W. (2006) “Wissen und Werte als Verhandlungsform.
Ethikexpertise in der Regulation der Stammzellforschung” in Wink, R. (ed.)
Deutsche Stammzellpolitik im Zeitalter der Transnationalisierung (Baden Baden:
Nomos), pp. 141–63.
Bogner, A., Littig, B. and Menz, W. (2009) “Expert Interviews. An Introduction to a
New Methodological Debate,” in this volume.
Bogner, A., Menz, W. and Schumm, W. (2008) “Ethikexpertise in Wertkonflikten. Zur
Produktion und politischen Verwendung von Kommissionsethik in Deutschland
und Österreich” in Mayntz, R., Neidhardt, F., Weingart, P. and Wengenroth, U. (eds)
Wissensproduktion und Wissenstransfer. Wissen im Spannungsfeld von Wissenschaft,
Politik und Öffentlichkeit (Bielefeld: Transcript), pp. 243–68.
Bourdieu, P. (1998) Practical Reason: On the Theory of Action (Palo Alto: Stanford
University Press).
78 Alexander Bogner and Wolfgang Menz

Brinkmann, C., Deeke, A. and Völkel, B. (1995) “Experteninterviews in der

Arbeitsmarktforschung. Diskussionsbeiträge zu methodischen Fragen und
praktischen Erfahrungen” in Beiträge zur Arbeitsmarkt- und Berufsforschung 191
(Nürnberg: IAB).
Deeke, A. (1995) “Experteninterviews – ein methodologisches und forschung-
spraktisches Problem. Einleitende Bemerkungen und Fragen zum Workshop”
in Brinkmann, C., Deeke, A. and Völkel, B. (eds) Experteninterviews in der
Arbeitsmarktforschung. Diskussionsbeiträge zu methodischen Fragen und praktischen
Erfahrungen. Beiträge zur Arbeitsmarkt- und Berufsforschung 191 (Nürnberg: IAB),
pp. 7–22.
Douglas, J. D. (1985) Creative Interviewing (Bevely Hills: Sage Publications).
Flick, U. (2004) “Constructivism” in Flick, U., Kardorff, E. von and Steinke, I. (eds)
A Companion to Qualitative Research (Thousand Oaks, London: Sage Publications),
pp. 88–94.
Fontana, A. and Frey, J. H. (1998) Interviewing. “The Art of Science” in Denzin, N. K.
and Lincoln, Y. S. (eds) Collecting and Interpreting Qualitative Materials (Thousand
Oaks and others: Sage Publications), pp. 47–78.
Froschauer, U. and Lueger, M. (2009) “Expert Interviews in Interpretative
Organisational Research,” in this volume.
Giddens, A. (1984) The Constitution of Society. Outline of the Theory of Structuration
(Cambridge: Polity).
Giddens, A. (1990) The Consequences of Modernity (Cambridge: Polity).
Gillham, B. (2000) The Research Interview (London und New York: Continuum).
Glaser, B. G. and Strauss, A. L. (1967) Discovery of Grounded Theory. Strategies for
Qualitative Research (Mill valley: Sociology Press).
Hägele, H. (1995) “Experteninterviews in der öffentlichen Verwaltung: Ausgewählte
praktische Probleme” in Brinkmann, C., Deeke, A. and Völkel, B. (eds)
Experteninterviews in der Arbeitsmarktforschung. Diskussionsbeiträge zu methodischen
Fragen und praktischen Erfahrungen. Beiträge zur Arbeitsmarkt- und Berufsforschung 191
(Nürnberg: IAB), pp. 69–72.
Helfferich, C. (2004) Die Qualität qualitativer Daten (Wiesbaden: VS-Verlag).
Hermanns, H. (2004) “Interviewing as an Activity” in Flick, U., Kardorff, E. von and
Steinke, I. (eds) A Companion to Qualitative Research (Thousand Oaks, London:
Sage Publications), pp. 209–113.
Hitzler, R., Honer, A. and Maeder, C. (1994) Expertenwissen. Die institutionalisierte
Kompetenz zur Konstruktion von Wirklichkeit (Opladen: Westdeutscher Verlag).
Hoffmann-Riem, C. (1980) “Die Sozialforschung einer interpretativen Soziologie.
Der Datengewinn” in Kölner Zeitschrift für Soziologie und Sozialpsychologie, Jg. 32,
pp. 339–72.
Holstein, J. A. and Gubrium, J. F. (1999) “Active Interviewing” in Bryman, A. and
Burgess, R. G. (eds) Qualitative Research, vol. II (London and others: Sage Publications),
pp. 105–21.
Hopf, C. (1978) “Die Pseudo-Exploration – Überlegungen zur Technik qualitativer
Interviews in der Sozialforschung” in Zeitschrift für Soziologie 7 , pp. 97–115.
Houtkoop-Steenstra, H. (2000) Interaction and the Standardized Interview. The Living
Questionnaire (Cambridge: Cambridge University Press).
Kassner, K. and Wassermann, P. (2005) “Nicht überall, wo Methode draufsteht, ist
Methode drin. Zur Problematik der Fundierung von ExpertInneninterviews” in
Bogner, A., Littig, B. and Menz, W. (eds) Das Experteninterview. Theorie, Methode,
Anwendung (Opladen: Leske and Budrich), pp. 95–111.
 The Theory-Generating Expert Interview 79

Kaufmann, J. C. (1996) L’entretien compréhensif (Paris: Nathan).

Kelle, U. and Erzberger, C. (1999) “Integration Qualitativer und Quantitativer
Methoden. Methodologische Modelle und ihre Bedeutung für die Forschungspraxis”
in Kölner Zeitschrift für Soziologie und Sozialpsychologie 51, pp. 509–31.
Knorr-Cetina, K. (1989) “Spielarten des Konstruktivismus” in Soziale Welt 40 ,
pp. 86–96.
Köhler, G. (1992) “Methodik und Problematik einer mehrstufigen Expertenbefragung”
in Hoffmeyer-Zlotnik, J. H. P. (ed.) Analyse verbaler Daten. Über den Umgang mit
qualitativen Daten (Opladen: Westdeutscher Verlag), pp. 318–32.
Koolwijk, J. van (1974) “Die Befragungsmethode” in Koolwijk, J. van. and Wieken-
Mayser, M. (eds) Techniken der empirischen Sozialforschung, 4. Band: Erhebungsmethoden:
Die Befragung (München, Wien: Oldenbourg), pp. 9–23.
Krafft, A. and Ulrich, G. (1995) “Akteure in der Sozialforschung” in Brinkmann, C.,
Deeke, A. and Völkel, B. (eds) Experteninterviews in der Arbeitsmarktforschung.
Diskussionsbeiträge zu methodischen Fragen und praktischen Erfahrungen. Beiträge zur
Arbeitsmarkt- und Berufsforschung 191 (Nürnberg: IAB), pp. 23–33.
Kvale, S. (1996) InterViews. An Introduction to Qualitative Research Interviewing
(Thousand Oaks and others: Sage Publications).
Lamnek, S. (1995) Qualitative Sozialforschung, Bd. 1: Methodologie, 3rd edn
(Weinheim: Beltz).
Leitner, A. and Wroblewski, A. (2009) “Between Scientific Standards and Claims of
Efficiency. Expert Interviews in Programme Evaluation,” in this volume.
Lueger, M. (1989) “Die soziale Situation im Interview” in Österreichische Zeitschrift für
Soziologie 14, pp. 22–36.
Martens, K. and Brüggemann, M. (2006) Kein Experte ist wie der andere. Vom Umgang
mit Missionaren und Geschichtenerzählern (TranState Working Papers No. 39,
Bremen).
Mayring, P. (1996) Einführung in die qualitative Sozialforschung, 3rd edn (Weinheim: Beltz).
Menz, W. (2009) Die Legitimität des Marktregimes. Leistungs- und
Gerechtigkeitsorientierungen in neuen Formen betrieblicher Leistungspolitik (Wiesbaden:
VS-Verlag).
Menz, W., Siegel, T. and Vogel, M. (2003) Leistungs- und Interessenpolitik aus der
Perspektive von Beschäftigten. Abschlussbericht an die Hans-Böckler-Stiftung (Frankfurt
am Main: JWG-Universität).
Merton, R. K. and Kendall, P. L. (1946) “The focused interview” in American Journal of
Sociology, Jg. 51, Heft, pp. 541–57.
Meuser, M. and Nagel, U. (1991) “ExpertInneninterviews – vielfach erprobt, wenig
bedacht. Ein Beitrag zur qualitativen Methodendiskussion” in Garz, D. and
Kraimer, K. (eds) Qualitativ-empirische Sozialforschung. Konzepte, Methoden, Analysen
(Opladen: Westdeutscher Verlag), pp. 441–71.
Meuser, M. and Nagel, U. (1994) “Expertenwissen und Experteninterview” in
Hitzler, R., Honer, A. and Maeder, C. (eds) Expertenwissen. Die institutionalisierte
Kompetenz zur Konstruktion von Wirklichkeit (Opladen: Westdeutscher Verlag),
pp. 180–92.
Meuser, M. and Nagel, U. (1997) “Das ExpertInneninterview – Wissenssoziologische
Voraussetzungen und methodische Durchführung” in Friebertshäuser, B.
and Prengel, A. (eds) Handbuch Qualitative Forschungsmethoden in der
Erziehungswissenschaft (Weinheim/München: Juventa), pp. 481–91.
Meuser, M. and Nagel, U. (2009) “The Expert Interview and Changes in Knowledge
Production,” in this volume.
80 Alexander Bogner and Wolfgang Menz

Plath, H. E. (1995) “Zum ‘Experteninterview’ – ein Diskussionsbeitrag” in

Brinkmann, C., Deeke, A. and Völkel, B. (eds) Experteninterviews in der
Arbeitsmarktforschung. Diskussionsbeiträge zu methodischen Fragen und praktischen
Erfahrungen. Beiträge zur Arbeitsmarkt- und Berufsforschung 191 (Nürnberg: IAB),
pp. 85–9.
Scheuch, E. (1967) “Das Interview in der Sozialforschung” in König, R. (ed.) Handbuch
der Empirischen Sozialforschung, I. Band, 2nd edn(Stuttgart: Enke), pp. 136–96.
Schmid, J. (1995) “Expertenbefragung und Informationsgespräch in der
Parteienforschung: Wie föderalistisch ist die CDU?” in Alemann, U. von (ed.)
Politikwissenschaftliche Methoden. Grundriss für Studium und Forschung (Opladen:
Leske und Budrich), pp. 293–326.
Schütz, A. (1962) “Common-sense and Scientific Interpretation of Human Action” in
Collected Papers, Vol. 1 (The Hague: Nijhoff), pp. 3–47.
Schütz, A. (1964) “The Well-Informed Citizen: An Essay on the Social Distribution of
Knowl-edge” in Collected Papers, Vol. 2 (The Hague: Nijhoff), pp. 120–34.
Sprondel, W. M. (1979) “ ‘Experte’ und ‘Laie’. Zur Entwicklung von Typenbegriffen in
der Wissenssoziologie” in Sprondel, W. M. and Grathoff, R. (eds) Alfred Schütz und
die Idee des Alltags in den Sozialwissenschaften (Stuttgart: Enke), pp. 140–54.
Trinczek, R. (1995) “Experteninterviews mit Managern: Methodische und
methodologische Hintergründe” in Brinkmann, C., Deeke, A. and Völkel, B.
(eds) Experteninterviews in der Arbeitsmarktforschung. BeitrAB191 (Nürnberg: IAB),
pp. 59–67.
Trinczek, R. (2009) “How to Interview Managers? Methodical and Methodological
Aspects of Expert Interviews as a Qualitative Method in Empirical Social Research,”
in this volume.
Voelzkow, H. (1995) “ ‘Iterative Experteninterviews’: Forschungspraktische
Erfahrungen mit einem Erhebungsinstrument” in Brinkmann, C., Deeke, A.
and Völkel, B. (eds) Experteninterviews in der Arbeitsmarktforschung. BeitrAB191
(Nürnberg: IAB),, pp. 51–7.
Vogel, B. (1995) “ ‘Wenn der Eisberg zu schmelzen beginnt...’ – Einige Reflexionen
über den Stellenwert und die Probleme des Experteninterviews in der Praxis
der empirischen Sozialforschung” in Brinkmann, C., Deeke, A. and Völkel, B.
(eds) Experteninterviews in der Arbeitsmarktforschung. BeitrAB191 (Nürnberg: IAB),
pp. 73–83.
Weber, M. (1980) Wirtschaft und Gesellschaft. Grundriss einer verstehenden
Sozialwissenschaft, 5. Auflage (Tübingen: J. C. B. Mohr).
Thomas P. W. (1970) “Normative and Interpretive Paradigms in Sociology” in
Douglas, J. D. (ed.) Understanding everyday life. Toward the reconstruction of sociological
knowledge (Chicago: Aldine), pp. 57–79.
Witzel, A. (2000) “The problem-centered interview” in Forum Qualitative Sozialforschung/
Forum: Qualitative Social Research [Online Journal], 1(1), available at http://www.
qualitative-research.net/fqs-texte/1-00/1-00witzel-e.htm.
3
At Eye Level: The Expert
Interview – a Talk between
Expert and Quasi-expert
Michaela Pfadenhauer

Even though this is generally assumed and persistently conveyed in research

methodology: The questioning of persons who are regarded as experts –
according to which criteria soever – does not constitute the specific char-
acteristic of the expert interview.1 Due to its underlying epistemological
interest (1) for one thing and due to the special kind of interview tech-
nique (2) for another thing, the expert interview is an independent pro-
cedure within the canon of what is known as the “qualitative interview”
(Hopf, 2000). In recent years, it has often been practised and has also been
reflected upon methodically and methodologically. The specific charac-
ter of the interlocutory form makes it necessary, (at least) for the interview
of certain (types of) experts, to embed this procedure in an ethnographic
research design (3).

3.1 The epistomological interest of the expert interview

The whole purpose of expert interviews seems to be obvious: They are

geared to the reconstruction of specific knowledge stocks, or of particularly
exclusive, detailed or comprehensive knowledge about particular know-
ledge stocks and practices, in short: to the reconstruction of expert know-
ledge. This must, however, be qualified as regards “habitual,” or “implicit”
elements of expert knowledge. For as all interviews, expert interviews too
lend themselves especially to the reconstruction of explainable knowledge
stocks.
One must therefore be sceptical about the suitability of the expert
interview as an instrument for the ascertainment and analysis of strat-
egies and relevances “that, in fact, apply in the decision behaviour, but
that are not necessarily reflectively available to the experts” (Meuser and
Nagel, 1997, p. 485). With this, it shall by no means be disputed that
experts too – just as incidentally as a matter of course – apply “implicit

81
82 Michaela Pfadenhauer

knowledge” that is not “clear and precise” (Schütz, 1972a, p. 87), but in
fact profoundly diffuse and that can at best be verbalized in a piecemeal
fashion by them. 2 Such “routinised expert knowledge” (Schröer, 1994)
can, however, hardly be reconstructed by means of an expert interview.
As regards the reconstruction of situated skills, of performance routines
and of quasi-automatic behaviour patterns, all kinds of interviews typ-
ically produce deficient or misleading results (cf. Hitzler 2000, p. 22). 3
However, for the reconstruction of stocks of knowledge that can be the-
matically distinguished and explained, that is for the reconstruction of
knowledge that is memorable as having been learned and is hence as a
rule known as knowledge (cf. Honer, 1993, p. 88), interviews prove to be
suitable instruments.

3.1.1 The knowledge of experts

The epistemological object of the expert interview proves to be focused on
a particular knowledge asset within the social knowledge stock: on special-
ist knowledge, which in the course of the progressive division of labour
is increasing proportionally to general knowledge as regards extent and
importance (cf. Schütz and Luckmann, 1979, p. 363). The societal specialist
knowledge assets are increasingly being differentiated and often have to be
acquired in protracted “secondary” socialization processes, from which that
type of knower emerges, which is referred to as “specialist.” He disposes of
a task-related, relatively well-defined partial knowledge within a specialist
knowledge field, which he needs to fulfil his specialist function. As opposed
to the specialist, the “expert” is that type of knower who has a good over-
view of the overall known knowledge in one field, that is an overview of a
specialist knowledge field, in other words who “knows what the (respective)
specialists know in their fields of knowledge – and how what they know
relates to each other” (Hitzler, 1994, p. 25).
The differentiation between special(ist) knowledge on the one hand and
expert knowledge as a kind of “overview knowledge” about specialist know-
ledge fields on the other hand makes it clear that by no means all kinds of
special knowledge can be equated with expert knowledge (cf. also Sprondel,
1979). Expert knowledge is not “only” a specific, not generally available
problem-solving knowledge. At a more fundamental level, it signifies the
knowledge one needs to probe into the causes of problems and the prin-
ciples of problem-solving strategies. As opposed to specialists, the expert
thus has a more comprehensive knowledge that enables him not only to
solve problems, but moreover to identify and to account for problem causes
as well as for solution principles. In short: The expert “typically knows the
knowledge stock that is “characteristic” or “relevant” for a certain field, he
has, so to speak, an overview of a specialist knowledge field and can offer
fundamental problem solutions or can apply these to individual problems
within this area” (Hitzler, 1994, p. 26).
 At Eye Level: An Expert Interview 83

With his knowledge about the principles of the issue respectively the fac-
tual logic, the expert – in comparison to other persons concerned with the
problem area, that is in comparison to non-experts (including specialists) –
has a relatively exclusive knowledge asset that is on principle not available
to everyone. His knowledge about the factual logic enables the expert to
clarify the logical consistency of the issue at hand. And as Soeffner (1989,
p. 222) makes clear using the example of interrogators, “the question, how
something should have been [to be] logically consistent is more important
[to experts] than what ‘actually’ happened.”

3.1.2 The competence of experts

According to Michael Meuser and Ulrike Nagel (1991, p. 443), experts have
“privileged accesses to information.” However, not only the information
at the exclusive disposal of the expert is of decisive importance for his
expertise, but furthermore also the (attributable) responsibility for problem-
solving related decisions. This refers to competence in the broader sense:
“Competence is obviously somehow connected to responsibility and to
skills and to willingness and to the fact that responsibility, skills and will-
ingness coincide” (Marquardt, 1981, p. 24). In the sense that the responsi-
bility for finding problem solutions is incumbent upon the expert as quasi
ultimate authority, expert competence transcends (exclusive) abilities and
(special) skills.4
Thus not only his exclusive knowledge stock – in comparison to other
persons dealing with the problem – is characteristic for the expert, but
moreover his responsibility (resulting from his knowledge of causes) for the
provision of possible problem solutions, that is of expertises. Responsibility
also means that irrespective of who else (apart from him) was involved in
whichever function or to whichever extent in its development process, the
ultimate responsibility for the expertise lies with the expert. This (ultimate)
responsibility is quasi the other side of the coin of the relative autonomy of
the expert, which follows from the fact that his knowledge is superior to the
“knowledge that is recallable or claimable from others, that is [that he has]
a (hardly ore even completely uncontrollable) advice and help competence”
(Hitzler, 1994, p. 26). For according to the “objective meaning” of respon-
sibility, it is ultimately he, who can be called to account by others for the
things that are done or not done (by him or by others) regarding the solu-
tion of problems.5
Bringing the definition proposal of Meuser and Nagel (1991) to a head, we6
thus regard those persons as “experts,” who have privileged access to infor-
mation and – moreover – who can be made responsible for the planning
and provision of problem solutions. From an elite theoretical perspective,
“experts” according to this definition appear less as members of a func-
tional elite (cf. Meuser and Nagel, 1994, p. 181), than as members of what
can, following Jaeggi (1960), be referred to as “relative,” that is local elites
84 Michaela Pfadenhauer

as opposed to global elites and, following Dreitzel (1962), as “performance

elites,” that is as providers of socially desired or demanded services, who as a
result of these services are granted (significant) privileges, options, resources
and/or esteem.7
The expert interview thus primarily lends itself as a data generating
instrument in those cases in which the research focuses on the exclusive
knowledge assets of experts in the context of their (ultimate) responsibil-
ity for problem solutions. In this broad sense, the epistemological interest
of the expert interview is aimed at the reconstruction of (explicit) expert
knowledge.

3.2 Interview techniques

As is generally known, a basic requirement of non-standardized social

research as opposed to standardized interview techniques consists in for-
bearing from imposing an external relevance system upon the interviewed
actor, letting him develop and formulate his own relevances instead.
However, one can by no means assume without further query that the
“interviewees can best develop their subjective attributions of meaning and
relevance structures in an interview situation characterised by far-reaching
non-intervention on the part of the interviewer.”8
Seeing that “the interviewed persons – just as the conventionally working
social scientists – envisage an interview as an onesided question-and-answer
relation,” Anne Honer (1994, p. 629) makes a case – especially for the first
interview phase of the explorative interview developed by her – for a “nor-
malization” or “everydaying” of the relatively extraordinary communica-
tion situation of the interview to the effect that it should comply as much
as possible with the cultural habits prevailing in the respective contexts
of speaking to each other. In principle, the aim of the expert interview is
neither to subject the interview partner to an interrogation-like nor to an
artificial “non-directive” interview situation, but rather to create a com-
munication situation with which he is familiar, that is to conduct a quasi-
normal conversation (for further details see Honer, 1993, p. 74).

3.2.1 The interview setting

The question of the ideal interview setting for the undisturbed develop-
ment of subjective relevance structures is discussed by Trinczek (1995 and
in this volume) using the example of interviews with managers. According
to him, communication in the companies is seldomly characterized by a
narrative basic structure. The structure of everyday communication in the
company-managerial context that as a rule features team-like work struc-
tures, in fact most closely corresponds to a “discursive-argumentative pro-
fessional discussion” within the scope of a (more or less) relaxed discussion
situation.
 At Eye Level: An Expert Interview 85

In general it can be stated that the communication of experts (of the same
provenance) amongst each other is characterized by the following features:
thematic focusing, use of professional terminology, deployment of index-
ical language, in short by the fact that experts (of the same provenance)
share a “communicative universe” (Schütz, 1972a, p. 97). This is not least
due to the fact that in discussions with his “peers,” an expert can assume
that he can take the knowledge of basic facts or interrelations for granted
and that he need not be afraid of being misunderstood – neither literally
nor figuratively – because his counterpart is not familiar with the tech-
nical terms and most notably with the relevances of his field of activity that
structure his thinking and acting. The individual knowledge stock of the
expert is structured by a given system of “imposed relevances” of the spe-
cific field of knowledge that is, it is not connected with his spontaneously
chosen goals. This relevance system is imposed upon him by the assumed
problems of his field of activity; with his decision to become an “expert,”
he, at the same time, accepts it as the only significant relevance system for
his thinking and acting (see Schütz, 1972a, p. 96).
On the other hand, particularly the awareness of diverging relevance
systems leads experts in communication with non-experts (observable
for example in interviews with journalists) to enrich their speech with
metaphors and analogies, to play down or to dramatize or to be inclined
to adopt a paternalistic or self-legitimizing conversational behaviour.9 The
respective semantics – for example lingo towards other “experts,” “trans-
lations” towards “well-informed citizens,” simplistic statements towards
“laypersons” – reveals which knowledge type the interviewee believes his
counterpart to be.10
In contrast, discussions among experts (of the same provenance) either
serve to augment their privileged knowledge access – in terms of two-way
briefing – or the reciprocal explanation of their actions with regard to their
competence and responsibility for the development and provision of solu-
tions to problems. What takes place in the process is not instruction11 or
(placating) justification,12 which one can typically observe towards a non-
expert (public), but a description and a discursive explanation of what he is
doing and why he is doing what he is doing in the way he is doing it.13 Thus
experts typically debate the significance and practical management of their
competence and responsibility for the development, implementation and/
or control of solutions to problems amongst each other.14 In general terms,
the basic conditions and implications of expert competence are the typical
subject matter of communication among experts.
A different information content from that normally meted out to a lay
(public) is being exchanged. For: “The expert knows [on the other hand]
that only another expert will understand all technical details and implica-
tions of a problem in his field, and he will never accept a layperson or an
amateur as a competent judge of his attainment” (Schütz, 1972a, p. 88).
86 Michaela Pfadenhauer

Insofar as the epistemological interest of the expert interview is directed at

issues that are regarded as relevant or are being debated among experts, the
accompanying basic concern of the expert interview is to create an inter-
view setting that approaches the conversation situation among experts as
closely as possible. And a fundamental requirement and condition for this
is precisely not only the status of the interviewed actor as an expert but also
a similar status of the interviewer.15

3.2.2 The status of the interviewer

In view of the two core components of expert competence mentioned
above, the – appropriately as well as privilegedly informed – interviewer typ-
ically at best achieves the status of a quasi-expert, inasmuch as he interacts
unfettered by responsibility for the development and provision of problem
solutions – and thus with the essential difference of being free from the
burden of action.
Consequently, the demeanour of the interviewed expert vis-à-vis the
interviewing quasi-expert does not only not correspond to that, which, for
instance, expresses itself in a paternalistic “mannerism” or justification pres-
sure towards a lay public, but also not (“really”) to the demeanour towards
other experts. For the meeting between experts (of the same provenance)
it is typically marked by a – though not necessarily concrete, but at least
fundamental – competitive pressure. Therefore it is characterized by a (at
any rate in principle) inherent reserve regarding the disclosure of “trade
secrets.” As the epistemological interest of the expert interview is directed
precisely towards this “operational knowledge” of experts, the “freedom
from the burden of competition” inherent in the exchange between an
expert and a quasi-expert represents a particular advantage. Thus Trinczek
(1995, p. 63, own translation) observes that the fact that the interview is
free from the burden of action and from social consequences “sometimes
[allows] the managers a degree of openness and candid self-reflection that
they normally do not thus concede themselves in their everyday working
environment with its predominantly strategically oriented communication
and interaction style.”16
Trinczek (1995, p. 65, own translation) too emphasises that the thematic
competence of the interviewer is a necessary condition for a successful
expert interview with managers: “The more one is able to slip in qualified
assessments, reasons and counterarguments every now and then in the
course of the interview, the more the managers on their part are prepared to
disclose their knowledge and their positions – and to reveal their subjective
relevance structures and orientation patterns in non-strategic intent.”
(At least) amongst the representatives of what is known as “qualitative
social research” there is a wide consensus to the effect that an orientation
on the situational-subjective agenda setting and relevance structuring of
the interview partner is significantly facilitated by means of the “flexible
 At Eye Level: An Expert Interview 87

interview-situational” use of a “guideline,” which is necessary to “guaran-

tee the openness of the interview course” (Meuser and Nagel, 1991, p. 449;
in this spirit also already Dexter, 1970). Opinions differ, however, as regards
the extent to which the researcher should actually let himself be guided by
his guideline during the interview.17 As a matter of principle, it must be
pointed out that only the detached stance of the interpreter when analys-
ing the whole interview text permits a decision regarding the “importance”
or “triviality” of a statement. Of much greater significance in this context,
however, is that the conceptual design of this – ideally only “mentally”
present – guideline already presupposes an as comprehensive and appropri-
ate as possible knowledge on the part of its “designer” (cf. Honer, 1994 as
well as Hitzler, 2000).

3.3 The ethnographic embedding of the expert interview

The here developed conception of the expert interview that essentially fol-
lows the pertinent note of Anne Honer (1994, p. 633), is based on the prem-
ise that people speak differently to other people – both as regards the “how”
and the “what” of communication – depending on whether they regard
their conversation partners as competent or incompetent (and thus in a
way also as relevant or irrelevant) concerning the debated issue. Contrary
to the well-nigh inflationary labelling of all kinds of interviews as “expert
interviews,” we advocate that only those forms of interviews be referred to
as “expert interviews” that are characterized by a discussion on an equal
footing.
The special conversational form which characterizes the expert interview –
the researcher as quasi-expert under discussion with an expert – makes it
seem essential to us on at least two counts to embed the method in an
ethnographic research design: for one thing with regard to the identifica-
tion of experts and, for another things, with regard to the qualification of
the interviewer.18

3.3.1 The identification of experts

The picture that the relevant literature draws of experts largely corresponds
to the figure of the professional, who obtains his skills by complying with
formal training requirements and who can also provide formal evidence for
his expert status by means of certificates issued by professional organiza-
tions: “Today, holding a professionally organised expert role in all cases pre-
supposes the completion of general as well as special training, in which the
special knowledge that is considered relevant is obtained. The acquisition
of this knowledge is sanctioned in corresponding certificates with societal
validity” (Sprondel, 1979, p. 151).
The expert status of the professional can be made plausible using doctors
as an example, who were for a long time regarded as the “prototype” of
88 Michaela Pfadenhauer

the professional in the sociology of professions: The doctor has privileged

access to information and he is (ultimately) responsible for the develop-
ment and provision of medical problem solutions. The professional claim
to competence of the medical fraternity essentially aims at permanently
and exclusively binding not only the entitlement to alleviate or cure ail-
ments but also that of medical expertises to that group of persons, which
verifiably meet the qualification standards defined by the profession: to the
duly qualified medical practioners (for further information see Hitzler and
Pfadenhauer, 1999).
Unlike for other types of experts, it is characteristic for the professional
that he acquires a canonized special knowledge asset via an institutionally
specialized and – as regards extent and duration – formalized training in
typically “public” institutions that the acquisition of this professional spe-
cial knowledge is tested (often in cooperation between state and professional
organisations) and that he is issued a certificate that “certifies” his profes-
sional competence – not only in terms of qualification but also in terms of
authorisation. In particular due to these characteristics, the professional is a
typically modern – and thus historically relatively “young” – manifestation
of the expert (see Hitzler, 1994, as well as Pfadenhauer, 1997).
Peter L. Berger and Thomas Luckmann (1969, p. 133) specify the “edu-
cated person” and the “intellectual” as further types of experts. They char-
acterize the latter as a (counter) expert “whose expertise is not desired
by society,” because he provides a counter design for the determination
of reality. According to Zygmunt Bauman (1987), the role of the intel-
lectual is changing from “legislator” to “interpreter” under postmod-
ern conditions. Under the generic term “Man of Knowledge,” Znaniecki
(1975) presented a so far largely unheeded typology of socially sanctioned
“knowers,” which differentiates between users of knowledge, administra-
tors of knowledge and discoverers of (new) knowledge. The following sim-
ple contrasting of this modern expert with the premodern expert must
suffice for now to make our argument clear that experts are not congruent
with professionals.
The first empirical manifestation of the expert is probably the shaman.
According to Mircea Eliade (1975), his social function is that of an expert for
the extraordinary, “who gains strength and authority to come to terms with
specific collective and individual borderline situations by dint of his ‘voy-
ages to the otherworld’, which only he is capable of” (Hitzler, 1982, p. 55).
As he is able to get in touch with spirits, he not only has privileged access to
information but is moreover responsible for the provision of problem solu-
tions that essentially consist in “fending off evil or wrong” or in “effecting
good or right.” His competences and special skills are the result or “outflow”
of the ecstatic spirit-edness of his body, which is preceded by a biographical
disruption or an identity crisis and which leads to the formation of a new
identity (cf. in more detail Hitzler, 1982).
 At Eye Level: An Expert Interview 89

This short excursion into shamanism is by no means aimed at mystifying

or putting the expert on a pedestal. The experts of today only seldom have
competences they gained by extraordinary means (such as divine inspir-
ation or gifted birth). But unlike the professional, they are by all means
endued with competences, whose acquirement demand other, to wit mainly
non-formalized, in fact diffuse, not clearly designated gateways. In this
connection, the focus “on the (...) expertise crystallised primarily in an
occupational role” (Sprondel, 1979, p. 141) construes the concept of expert
knowledge too narrowly. The expert is neither – as I wanted to show by
means of the example of the shaman – identical with the professional, nor
can expert knowledge “per se” be equated with canonised specialised know-
ledge that is formally conveyed and certified in educational institutions.
And least of all does an university degree represent a necessary or even suf-
ficient precondition for the expert status.19
In societies like ours too, the expert status is in no way necessarily linked
to schooling and training in – typically “public” – educational institutions.
In dynamic growth sectors, currently for instance in the IT and multimedia
sectors, the “provenance” or the proof of relevant skills by means of state-
approved certificates issued by professional organizations play a secondary
role. In particular the acquisition of hard and software related knowledge
and skills is supplied to a large extent autodidactically or in a net-based
exchange amongst each other. The same goes for event management prac-
tice (cf. Pfadenhauer, 2000/2007). Here the emphasis is persistently placed
on “learning by doing” and “training on the job.” A further ongoing trend
in this field is that of lateral entry, that is the entry from a course of studies
in another subject or even without a diploma or degree. Formal certificates
of competence are only assigned a higher priority as selection criteria – in
particular for executive positions or expert functions – in the course of con-
solidation processes and decreasing manpower requirements.
Generally one has to take into account that the question who is an
expert – here or elsewhere – depends on the approach, on the “framing” –
in the sense of Goffman (1977) – determined by the research interest or the
research object (cf. Meuser and Nagel, 1991). And if one assumes that for
each grouping, also within a society, other forms of knowledge and most
notably other hierarchies of knowledge types are relevant or can be rele-
vant (cf. Hitzler, 1999), already the possibility of being able to identify the
expert as such presupposes a relatively detailed ethnographic20 inventari-
zation of the research field in many spheres of activity and culture, inas-
much as the ethnographic epistemological interest is generally directed at
the reconstruction of the culturally typical (subjective and objective) stock
of knowledge.21
For a start, this requires that the researcher suspend his pre-conceptions
and ex ante certainties concerning the research field he is about to exam-
ine and instead enquires into how the actors themselves see their world
90 Michaela Pfadenhauer

(and what they see) (see for example Pfadenhauer, 2001b). As regards the
question of expert knowledge, he has to screen the specific cultural know-
ledge to the effect which “constituents” are in principle known by everyone,
that is belong to the specific general education and which socially relevant
stocks of knowledge, that is the (exclusive) knowledge regarded as essential
for solving the impending problems, stand out against the general know-
ledge. As to the identification of experts, he has to find out which (types of)
persons have a privileged access to information concerning these matters
and are being made responsible for the provision of solutions to problems.

3.3.2 The qualification of the interviewer

Apart from the sometimes quite problematic identification of experts, the
acquisition of a high degree of thematic competence on the part of the
interviewer before he conducts the interview is constitutive for an expert
interview. This implies that the interviewer must do his utmost to acquire
as much as possible of the – relatively exclusive – special knowledge that the
expert has normally acquired in the course of a long (secondary) socializa-
tion process.
To the extent that the knowledge corpus, which the experts must typic-
ally command in their fields, is relatively well – defined, viz. is put down
in study and examination regulations, in task and job descriptions and so
on the social researcher has numerous ways and means of acquiring this
knowledge. First and foremost, he will be inclined to obtain and to study
different types of “canonical documents” that impart the relevant special
and specialized knowledge, thus, for instance, textbooks, course books, sci-
entific journals and documentations as well as work reports, minutes of
meetings and discussions, professional codes of conduct and much more.22
Furthermore, there is the possibility of participating in (further) education
and training measures in public and private institutions in which the pro-
fessional expert himself acquires and extends his knowledge.
It is, however, obvious that this only concerns theoretical knowledge,
insofar as the prevalent work descriptions too (textbook explanations,
instruction sheets and so on) are always “theoretical” or at best impart
“how-to-do-recipes.” What they precisely mean only becomes discernible in
“practice,” in the practical implementation: “It is only in the course of prac-
tical work that the actor acquires the competence to carry out work proc-
esses ‘correctly’, to cope with imponderabilities and unpredicatable events
and to come to ‘sensible’ decisions situational” (Eberle, 1997, p. 267, own
translation).23 The researcher too typically only acquires a basic or back-
ground knowledge in this manner and thus at best that measure of insight
that enables him to competently assess expert competence – and thence to
conduct expert interviews in the sense advocated here.
The interviewers are, however, faced with particularly grave problems in
those action or research fields, in which the expert knowledge precisely
 At Eye Level: An Expert Interview 91

does not assume the shape of formalised and certified stocks of special
knowledge. While one can acquire the canonized, formally designated
stocks of knowledge of professionals – albeit partly only with great effort –
via well-known knowledge transfer paths that are more or less accessible to
“everyone,” the acquisition of non-certified, rather diffuse stocks of special
knowledge of “other” experts is invariably only successful if the researcher
follows those “tracks” through (initially) “alien” worlds that show him how
these actors acquire their skills, which make them into experts of certain
socio-cultural contexts.24
In principle, the basic non-standardized techniques of data generation –
which, as is generally known, consist of following the proceedings, of pro-
curing documents and of speaking to the people – lend themselves to this
purpose. All ethnographic variants have in common that the researchers go
into the field more or less intensively and at the same time act in the field in
such a way that they change it as little as possible. The importance ethnog-
raphers assign to “existential involvement” (Honer, 1993, p. 40), that is prac-
tical participation, last but not least results from a fundamental scepticism
towards the quality of data that have been generated by others, as these are
on principle data on how others situational present facts and circumstances
(a fact sometimes overlooked not only in the case of accounts “congealed”
in written texts) – and not data on the facts themselves.
For the qualification of the researcher as (quasi-) expert this entails that
he has to be present and participate as far as possible in all activities that
the experts identified by him undertake as experts. That is to say, the ideal
basis for the acquisition of a comprehensive as possible pertinent prior
knowledge – which is constitutive for conducting expert interviews – is the
“acquisition of a virtual membership in the events that are being researched,
and thus the advantage of an existential interior view” (Honer, 2000, p. 198,
own translation). Therewith the researcher gains a practical familiarity with
the research field that expresses itself as (at least potential) action compe-
tence and sufficiently qualifies the interviewer to conduct a conversation
“on equal footing.”

3.4 Conclusion

The form of interview we suggest be referred to as expert interview repre-

sents a data generating instrument that is ridden with prerequisites and is
thus very time-consuming. Therefore its use proves to be expedient only in
view of definite research interests. It is quite certainly not suited, as assumed
by Renate Mayntz and others (1972, p. 103), as a substitute for “procedures
of direct data collection” that are “more time-consuming, costly or failing
due to practical-technical problems.” That is to say, the expert interview in
the sense referred to above, is far less suited as an instrument for the swift
generation of data, compensating as it were the time-consuming travail of
92 Michaela Pfadenhauer

participation, than as a kind of surplus procedure, whose competent use

already more or less imperatively requires a high degree of field skills and
of field acceptance.

Notes
1. Cf. (representatively for many) Deeke (1995, p. 7), according to whom “the term
‘expert interview’ already denotes that its distinctive feature does not consist of a
specific form of interview but in the fact that ‘experts’ are interviewed.” According
to Deeke, an expert interview is therefore not a special procedure and does not
involve a particular method.
2. Routine knowledge is a structural element of each subjective stock of knowledge
and can – following the phenomenologically oriented sociology of knowledge
(cf. Schütz and Luckmann, 1979, p. 139) – be analytically subdivided into skills,
useful knowledge and knowledge of recipes.
3. Bergmann (1985, p. 307) in particular emphasizes the problem of using the inter-
view as an investigation instrument, as it “produces reconstructively transformed
data that is only analysable to a certain extent”; on this fundamental insight see
also already Oevermann, 1983 as well as Reichertz, 1988. An impressive portrayal
of the problems of prompting do-it-yourselfers to explain their routine knowledge
is provided by Honer, 1993. Like Honer Soeffner (1989, p. 211) too emphasizes the
usefullness of ethnographic procedures for the reconstruction of implicit routine
knowledge (just as Schröer 1994, using the police interrogarion of juvenile sus-
pects as an example).
4. From a stage management perspective, a fundamental question arises concerning
the recognizability of facts: Then the expert does not appear as “someone who
has special competencies, but as someone who is skilled in making it plausible to
society that he has special competencies” (Hitzler, 1994, p. 27, own translation,
cf. also Pfadenhauer, 1998 for this perspective on professionals).
5. Schütz (1972b, p. 256) distinguishes the subjective aspect of responsibility (in terms
of “being responsible towards”) from the “objective meaning” (in the sense of being
“responsible for”): “If I only feel subjectively responsible for that what I did or failed
to do, without being held responsible by others, the consequence of my wrongdoing
will not be reproach, criticism, censure or another form of punishment that some-
one imposes on me, but regret, pangs of conscience or remorse.”
6. As I make no claim to originality for many aspects of the view advanced in this
text, but see myself as moving within the discussion context of ethnographically-
oriented sociologists, I make use of we-form in various passages.
7. Experts no more than elites can simply be derived from the functional needs of
the social and policial system, but can rather be determined on the basis of their
personal performance. Whether or even that these perfomances are functional
for the existence of a “system” of whichever nature, is another matter (for further
details cf. Pfadenhauer, 2001a).
8. Cf. Trinczek (1995, p. 60), who harshly criticizes the “fetishising of” as weak as
possible interviewer intervention in the interpretative paradigm.
9. Unlike Vogel (1995, p. 80), who first and foremost puts the “paternalism effects,”
as he calls them, down to age and status differences, we primarily regard the (real
or supposed) competence gap as the factor “triggering” these effects. For more
information on the manifold communication styles of experts cf. also Martens,
Brüggemann, 2006.
 At Eye Level: An Expert Interview 93

10. For a sociology of knowledge perspective on the differentiation of these three

types regarding their individual knowledge assets see Schütz, 1972a.
11. A characteristic feature of the communicative genre “instruction” is statements
designated as explicit knowledge (cf. Keppler and Luckmann, 1991).
12. In the tradition of discourse analytical ethnomethodology, justifications (as
well as excuses) can be conceived as accounts, that is as linguistic devices
employed “whenever an action is subjected to evaluative inquiry” (Scott and
Lyman, 1976, p. 74).
13. I would like to point out again that the expert – even with the best of intentions –
can and will only disclose knowledge of which he is consciously aware.
14. Strictly speaking, this is also a form of accounting, which is here (in terms of
“theoretical” explanations) to all intents and purposes understood to include sci-
entific causal explanations or explanations with a similar epistemological claim.
In this context, it is noteworthy that parts of the respective essay of Scott and
Lyman (1976) were originally published under the title “responsibilities” in the
German translation by Heinz Steinert.
15. Cf. on another setting of priorities Maindok (1996), who in fact explicitly
speaks of “experts for conducting interviews,” but who in doing so in particu-
lar addresses the strategic-communicative meta(competence) of a “professional”
interviewer in discussions with laypersons.
16. “The appeal of this consequence – free interview situation also manifests itself
in the fact that the interviewees sometimes considerably exceed the time limit,
even if every quarter of an hour had been bargained over during the arrange-
ment of the meeting; frequently the researchers rather than the managers end
the interview on their own accord” (Trinczek, 1995, p. 63, own translation).
17. For a favourable opinion on a departure from the guideline in support of what
is known as “catharsis effects” cf. Kern, Kern and Schumann, 1988, Vogel, 1995;
critical as regards the “lapse” into personal and private matters Meuser and
Nagel, 1991.
18. As already indicated above, it seems essential to us to integrate expert interviews
into an ethnographic research design also with respect to an “overall ascertain-
ment” of the knowledge of experts, that is with respect to the reconstruction of
the implicit elements of expert knowledge too, as well as – without expatiating
on this at large – as condition for a context-related analysis of all the collected
data. On the different methods or techniques of data evaluation subsumed under
the label of “social scientific hermeneutics” within the scope of ethnographic
research work see Honer, 1993, pp. 89–110 as well as the contributions in Hitzler
and Honer, 1997, on the “rules of procedure” they have in common cf. Hitzler,
2000, pp. 25–28.
19. On this note cf. also Meuser and Nagel (1994, p.180) and their criticism of the
expert concept of Hartmann and Hartmann, 1982.
20. For a basic description cf. Hitzler, 2007, for an overview of different types of
ethnography see Hitzler, 2000 and for a contrastive comparison of conventional
and focused ethnography Knoblauch, 2001 and Knoblauch, 2003.
21. In the process of research, as many and as varied as possible data – with respect
to the research interest – are collected and analysed. In doing so, the ethno-
graphic “ideal” is the combination of as varied as possible methods while
following specific theoretical guidelines of data collection (cf., once more,
Honer, 1993).
22. The (partly highly specialized) educational series on television and radio as well
as the well-nigh unfathomable possibilities of the internet sometimes provide
94 Michaela Pfadenhauer

effective opportunities for a first introduction into the knowledge acquirement

and accumulation of professionals.
23. The interest of the ethnomethodologically-oriented “studies of work” is there-
fore directed at the concrete work execution in occupational practice. And the
programme of “workplace studies” that is interested in work execution in com-
plex technical surroundings, has already firmly integrated the ethnographic
approach (for an overview see Eberle, 2007).
24. To stick to the chosen and by no means only “archaic” example: The typically
subjective stock of knowledge of the shaman at best reveals itself to the researcher
if he tries to acquaint himself intensely with the world of the shamans. There
seems to us to be the only possibility for researchers to acquire the as compre-
hensive as possible appropriate prior knowledge that is constitutive for expert
interviews. The interviewer must, as it were, seek to become a (quasi-) shaman
himself, to be able to speak with shamans on an “equal footing. Therefore the
intense existential involvement of the ethnographer is crucial in this case.”

Further readings
Gillham, B. (2005) “Elite Interviewing” in Gillham, B. (ed.) Research Interviewing: The
Range of Techniques (London: Continuum).
Holstein, J. A. and Gubrium, J. F. (1997) “Active Interviewing” in Silverman, D. (ed.)
Qualitative Research: Theory, Method and Practice (London: Sage), pp. 113–29.
Knoblauch, H. (2005) “Focused Ethnography” in Forum Qualitative Sozialforschung
6 (3), http://www.qualitative-research.net/fqs-texte/3-05/05-3-44-e.htm.

References
Bauman, Z. (1987) Legislators and Interpreters. On Modernity, Postmodernity and
Intellectual (Cambridge: Polity Press).
Berger, P. L. and Luckmann, T. (1969) Die gesellschaftliche Konstruktion von Wirklichkeit
(Frankfurt am Main: Fischer).
Bergmann, J. (1985) “Flüchtigkeit und methodische Fixierung sozialer Wirklichkeit”
in Bonß, W. and Hartmann, H. (eds) Entzauberte Wissenschaft, Special volume 3 of
the Soziale Welt (Göttingen: Schwartz), pp. 299–320.
Deeke, A. (1995) “Experteninterviews – ein methodologisches und forschungsprak-
tisches Problem” in Brinkmann, C., Deeke, A. and Völkel, B. (eds) Experteninterviews
in der Arbeitsmarktforschung, BeitrAB191 (Nürnberg: IAB), pp. 7–22.
Dexter, L. A. (1970) Elite and specialized interviewing (Evanston: Northwestern
University Press).
Dreitzel, H. P. (1962) Elitebegriff und Sozialstruktur (Stuttgart: Enke).
Eberle, T. S. (1997) “Ethnomethodologische Konversationsanalyse” in Hitzler, R. and
Honer, A. (eds) Sozialwissenschaftliche Hermeneutik (Opladen: Leske and Budrich),
pp. 245–79.
Eberle, T. S. (2007) “Ethnomethodologie und Konversationsanalyse” in Schützeichel, R.
(ed.) Handbuch Wissenssoziologie und Wissensforschung (Konstanz: UVK), pp. 139 –60.
Eliade, M. (1975) Schamanismus und archaische Ekstasetechnik (Frankfurt am Main:
Suhrkamp).
Fontana, A. and Frey, J. H. (2000) “The Interview: from structured questions to nego-
tiated text” in Denzin, N. K. and Lincoln, Y. S. (eds) Handbook of qualitative research,
2nd edn (Thousand Oaks, CA: Sage), pp. 645–72.
 At Eye Level: An Expert Interview 95

Gillham, B. (2005) “Elite Interviewing” in Gillham, B. (ed.) Research Interviewing: The

Range of Techniques (London: Continuum).
Goffman, E. (1977) Rahmenanalyse (Frankfurt am Main: Suhrkamp).
Gorden, R. L. (1975) Interviewing: Strategy, Techniques and Tactics (Homewood, III:
Dorsey).
Gubrium, J. F. and Holstein, J. A. (2002) Handbook of Interview Research: Context and
Method (Thousand Oaks, CA: Sage Publications).
Hartmann, H. and Hartmann, M. (1982) “Vom Elend der Experten: Zwischen
Akademisierung und Deprofessionalisierung” in Kölner Zeitschrift für Soziologie und
Sozialpsychologie 34, 193–223.
Hitzler, R. (1982) “Der ‘begeisterte’ Körper. Zur persönlichen Identität von Schamanen”
in Gehlen, R. and Wolf, B. (eds) Unter dem Pflaster liegt der Stand, Vol. 11 (Berlin:
Kramer), pp. 53–73.
Hitzler, R. (1994) “Wissen und Wesen des Experten. Ein Annäherungsversuch – zur
Einleitung” in Hitzler, R., Honer, A. and Maeder, C. (eds) Expertenwissen. Die insti-
tutionalisierte Kompetenz zur Konstruktion von Wirklichkeit (Opladen: Westdeutscher
Verlag), pp. 13–30.
Hitzler, R. (1999) “Welten erkunden. Soziologie als (eine Art) Ethnologie der eigenen
Gesellschaft” in Soziale Welt 50, 473–83.
Hitzler, R. (2000) “Die Erkundung des Feldes und die Deutung der Daten.
Annäherungen an die (lebensweltliche) Ethnographie” in Lindner, W. (ed.)
Ethnographische Methoden in der Jugendarbeit (Opladen: Leske and Budrich),
pp. 17–31.
Hitzler, R. (2007) “Ethnographie” in Buber, R. and Holzmüller, H. (eds) Qualitative
Marktforschung. Konzepte – Methoden – Analysen (Wiesbaden: Gabler), pp. 207–18.
Hitzler, R. and Honer, A. (1997) Sozialwissenschaftliche Hermeneutik (Opladen: Leske
and Budrich).
Hitzler, R. and Pfadenhauer, M. (1999) “Reflexive Mediziner? Die Definition pro-
fessioneller Kompetenz als standespolitisches Problem am Übergang zu einer
‘anderen’ Moderne” in Maeder, C., Burton-Jeangros, C. and Haour-Knipe, C. (eds)
Gesundheit, Medizin und Gesellschaft. Beiträge zur Soziologie der Gesundheit (Zürich:
Seismo), pp. 94–111.
Hitzler, R. and Pfadenhauer, M. (2000) “Die Lage ist hoffnungslos, aber nicht ernst!
(Erwerbs-) Probleme junger Leute heute und die anderen Welten von Jugendlichen”
in Hettlage, R. and Vogt, L. (eds) Identitäten im Umbruch (Opladen: Westdeutscher
Verlag), pp. 361–80.
Holstein, J. A. and Gubrium, J. F. (1997) “Active Interviewing” in Silverman, D. (ed.)
Qualitative Research: Theory, Method and Practice (London: Sage), pp. 113–29.
Honer, A. (1993) Lebensweltliche Ethnographie (Wiesbaden: DUV).
Honer, A. (1994) “Das explorative Interview. Zur Rekonstruktion der Relevanzen von
Expertinnen und anderen Leuten” in Schweizerische Zeitung für Soziologie 20, 623–40.
Honer, A. (2000) “Lebensweltanalyse in der Ethnographie” in Flick, U., Kardoff, E. von
and Steinke, I. (eds) Qualitative Forschung (Reinbek: Rowohlt), pp. 194–204.
Hopf, C. (2000) “Qualitative Interviews – ein Überblick” in Flick, U., Kardoff, E. von
and Steinke, I. (eds) Qualitative Forschung (Reinbek: Rowohlt), pp. 349–60.
Hunt, H. W., Crane, W. W. and Wahlke, J. C. (1964/65) “Interviewing Political Elites in
Cross-Cultural Comparative Research” in American Journal of Sociology 70, 59–68.
Jaeggi, U. (1960) Die gesellschaftliche Elite (Bern: Haupt).
Keppler, A. and Luckmann, T. (1991) “ ‘Teaching’: Conversational Transmission of
Knowledge” in Markova, I. and Foppa, K. (eds) Asymmetries in Dialogue (Hempstead:
Harvester Wheatsheaf), pp. 143–65.
96 Michaela Pfadenhauer

Kern, B., Kern, H. and Schumann, M. (1988) “Industriesoziologie als Katharsis” in

Soziale Welt 39, 86–96.
Knoblauch, H. (2001) “Fokussierte Ethnographie” in Sozialer Sinn 1, 123–41.
Knoblauch, H. (2003) Qualitative Religionsforschung: Religionsethnographie in der eigenen
Gesellschaft (München/Zürich: Vandenhoeck and Ruprecht).
Maindok, H. (1996) Professionelle Interviewführung in der Sozialforschung (Pfaffenweiler:
Centaurus).
Marquardt, O. (1981) “Inkompetenzkompensationskompetenz? Über Kompetenz
und Inkompetenz der Philosophie” in Marquardt, O. Abschied vom Prinzipiellen.
Philosophische – Studien (Stuttgart: Reclam), pp. 23–38.
Martens, K. and Brüggemann, M. (2006) Kein Experte ist wie der andere. Vom Umgang
mit Missionaren und Geschichtenerzählern (Bremen: Arbeitspapier 39 des SFB 597
“Staatlichkeit im Wandel” an der Universität Bremen).
Mayntz, R., Holm, K. and Hübner, P. (1972) Einführung in die Methoden der empirischen
Soziologie (Opladen: Westdeutscher Verlag).
Meuser, M. and Nagel, U. (1991) “ExpertInneninterviews – vielfach erprobt, wenig
bedacht. Ein Beitrag zur qualitativen Methodendiskussion” in Garz, D. and
Kraimer, K. (eds) Qualitativ-empirische Sozialforschung. Konzepte, Methoden, Analysen
(Opladen: Westdeutscher Verlag), pp. 441–71.
Meuser, M. and Nagel, U. (1994) “Expertenwissen und Experteninterview” in Hitzler, R.,
Honer, A. and Maeder, C. (eds) Expertenwissen. Die institutionalisierte Kompetenz zur
Konstruktion von Wirklichkeit (Opladen: Westdeutscher Verlag), pp. 180–92.
Meuser, M. and Nagel, U. (1997) “Das ExpertInneninterview – Wissenssoziologische
Voraussetzungen und methodische Durchführung” in Friebertshäuser, B. and
Prengel, A. (eds) Handbuch Qualitative Forschungsmethoden in der Erziehungswissenschaft
(Weinheim/München: Juventa), pp. 481–91.
Oevermann, U. (1983) “Zur Sache. Die Bedeutung von Adornos methodolo-
gischem Selbstverständnis für die Begründung einer materialen soziologischen
Strukturanalyse” in Friedeburg, L. von and Habermas, J. (eds) Adorno-Konferenz
1983 (Frankfurt am Main: Suhrkamp), pp. 234–89.
Pfadenhauer, M. (1997) “Die (Re-)Konstruktion professionellen Handelns. Überlegungen
zur Annäherung an den Forschungsgegenstand” in M. Pfadenhauer (ed.) Explorationen
zum Begriff des professionellen Handelns. Dokumentation des 1. Workshops des
Arbeitskreises “Professionelles Handeln” am 28.02. und 1.03.1997 in München (München:
Eigendruck), pp. 3–6.
Pfadenhauer, M. (1998) “Problem zur Lösung. Inszenierung von Professionalität” in
Willems, H. and Jurga, M. (eds) Inszenierungsgesellschaft (Opladen: Westdeutscher
Verlag), pp. 209–304.
Pfadenhauer, M. (2000) “Spielerisches Unternehmertum. Zur Professionalität
von Event-Produzenten in der Techno-Szene” in Gebhardt, W., Hitzler, R. and
Pfadenhauer, M. (eds) Events. Zur Soziologie des Außergewöhnlichen (Opladen: Leske
and Budrich), pp. 95–114.
Pfadenhauer, M. (2001a) “Macht – Funktion – Leistung. Elitentheoretische
Überlegungen zur Profession” in Mieg, H. and Pfadenhauer, M. (eds) Professionelle
Leistung – Positionen zur Professionssoziologie (Reihe “Wissen und Studium” im
Universitätsverlag Konstanz: UVK), pp. 81–7.
Pfadenhauer, M. (2001b) “Was andere Augen sehen. Perspektiven der Rezeption
des Techno-Videoclips ‘Sonic Empire’ ” in Hitzler, R. and Pfadenhauer, M. (eds)
Techno-Soziologie. Erkundungen einer Jugendkultur (Opladen: Leske and Budrich),
pp. 235–52.
 At Eye Level: An Expert Interview 97

Pfadenhauer, M. (2007) “Das Experteninterview. Ein Gespräch auf gleicher

Augenhöhe” in Buber, R. and Holzmüller, H. (eds) Qualitative Marktforschung.
Konzepte – Methoden – Analysen (Wiesbaden: Gabler), pp. 449–61.
Reichertz, J. (1988) “Verstehende Soziologie ohne Subjekt?” in Kölner Zeitschrift für
Soziologie und Sozialpsychologie 40, 207–22.
Schröer, N. (1994) “Routinisiertes Expertenwissen. Zur Rekonstruktion des struk-
turalen Regelwissens von Vernehmungsbeamten” in Hitzler, R., Honer, A. and
Maeder, C. (eds) Expertenwissen. Die institutionalisierte Kompetenz zur Konstruktion
von Wirklichkeit (Opladen: Westdeutscher Verlag), pp. 214–31.
Schütz, A. (1972a) “Der gut informierte Bürger” in Schütz, A. Gesammelte Aufsätze,
Vol. 2 (Den Haag: Nijhoff), pp. 85–101.
Schütz, A. (1972b) “Einige Äquivationen im Begriff der Verantwortlichkeit” in
Schütz, A. Gesammelte Aufsätze, Vol. 2 (Den Haag: Nijhoff), pp. 256–58.
Schütz, A. and Luckmann, T. (1979) Strukturen der Lebenswelt, Vol. 1 (Frankfurt am
Main: Suhrkamp).
Scott, M. B. and Lyman, S. M. (1976) “Praktische Erklärungen” in Anwärter, M.,
Kirsch, E. and Schröter, M. (eds) Seminar: Kommunikation, Interaktion, Identität
(Frankfurt am Main: Suhrkamp), pp. 73–114.
Soeffner, H. G. (1989) Auslegung des Alltags – Der Alltag der Auslegung. Zur wissens-
soziologischen Konzeption einer sozialwissenschaftlichen Hermeneutik (Frankfurt am
Main: Suhrkamp).
Sprondel, W. M. (1979) “ ‘Experte’ und ‘Laie’. Zur Entwicklung von Typenbegriffen in
der Wissenssoziologie” in Sprondel, W. M. and Grathoff, R. (eds) Alfred Schütz und
die Idee des Alltags in den Sozialwissenschaften (Stuttgart: Enke), pp. 140–54.
Strauss, A. (1987) Qualitative Analysis for Social Scientists (Cambridge: University Press).
Trinczek, R. (1995) “Experteninterviews mit Managern: Methodische und meth-
odologische Hintergründe” in Brinkmann, C., Deeke, A. and Völkel, B. (eds)
Experteninterviews in der Arbeitsmarktforschung. BeitrAB191 (Nürnberg: IAB),
pp. 59–67.
Vogel, B. (1995) “ ‘Wenn der Eisberg zu schmelzen beginnt...’ – Einige Reflexionen
über den Stellenwert und die Probleme des Experteninterviews in der Praxis
der empirischen Sozialforschung” in Brinkmann, C., Deeke, A. and Völkel, B.
(eds) Experteninterviews in der Arbeitsmarktforschung. BeitrAB191 (Nürnberg: IAB),
pp. 73–83.
Znaniecki, F. (1975) The Social Role of the Man of Knowledge (New York: Octagon).
4
Interviewing the Elite – Interviewing
Experts: Is There a Difference?1
Beate Littig

4.1 Introduction

A comparison of the Anglo-American methodological debate in the social

sciences with its counterpart in the German-speaking world reveals that
scientists in the latter have now regarded expert interviews as a distinct
interview form for some years (Flick and others, 2003, Bogner and others,
2005, Gläser and Laudel, 2004, Mieg and Näf, 2006). With few exceptions,
for example (Brandl and Klinger, 2006), the notion of the elite interview is
rarely, if ever, encountered in German-speaking countries. Yet the reverse
is the case in the Anglo-American and (given the dominance of English as
the language of scientific publication) international social sciences debate
see, for example (Gubrium and Holstein, 2002 or Denzin and Lincoln,
2000), where the concept of the expert interview appears largely unknown.
However, closer study reveals that in many respects the content of publica-
tions on these two interview forms does not really differ fundamentally.
In fact, quite the opposite applies: central themes in both methodological
traditions include the problems of gaining access to the elite or to experts
(particularly at a high level) as well as the specifics of interaction and the
actual interview process itself. Although not identical, even the respective
target group definitions (experts and the elite) for such interviews overlap.
The focus of interest in both generally lies on the professional (functional)
elite and on professional experts. Indeed, it would seem that the differences
between interviews with the elite and interviews with experts lie primarily
in differing social and political sciences research traditions and interests.
The following article discusses the commonalities and differences in these
two methodological approaches, thereby bringing together current insights
on comparable techniques and contributing to establishing a more detailed
specification of the methodology of expert interviews. It concludes with
a sociology of knowledge based appeal that the (professional) functional
elite – given their positions of power – be considered as a specific group of
experts. From a methodological perspective and as a result of their specific

98
 Interviewing the Elite 99

interpretive knowledge (“know why”) and procedural knowledge (“know-

how”), experts (and thus also the elite) are of relevance to social and pol-
itical sciences research. Consequently, interviews with the elite aimed at
generating explicit, tacit, professional or occupational knowledge should be
seen as expert interviews.

4.2 Experts and the elite as interview partners

The terms “elite interview” or “expert interview” raise the assumption that
the methodological rationale behind them (as unique interview forms) is
linked to specific characteristics in their respective target groups. So what
actually makes a person a member of the elite or an expert? Similarly, what
makes these particular groups so interesting from a social or political sci-
ences perspective?
One aspect, which quickly becomes apparent in literature on interview-
ing the elite is that authors generally presume the notion of the elite to be
clear and see no need to explain it further. For the most part, reflections
begin and end with a more or less vague working definition of what consti-
tutes the elite. Consequently, the elite are often defined by their compara-
tively high social status and the associated privileges they enjoy: “However
the whole notion of an elite, implies a group of individuals, who hold, or
have held, a privileged position in society and, as such, as far as a political
scientist is concerned, are likely to have had more influence on political
outcomes than general members of the public.” (Richards, 1996, p. 199).
As Dexter notes in his now classic book on interviewing the elite,2 this
group’s members are “the influential, the prominent, and the well informed”
(Dexter 2006/1969, p. 19), a definition that, in essence, has remained con-
stant in methodological literature to this day.3 Indeed, any search for a clear
definition is usually in vain see, for example (Moyser and Wagstaffe, 1987,
Seldon, 1996, Odendahl and Shaw, 2002, Lilleker, 2003).
One exception is the 2002 article by Welch and others reflecting on the
methodology behind in-depth interviews with the corporate elite in which
the authors define the elite in relation to their field of research (international
business research) as: “an informant (usually male) who occupies a senior
or middle management position; has functional responsibility in an area
which enjoys high status in accordance with corporate values; has consider-
able industry experience and frequently also long tenure with the company;
possesses a broad network of personal relationships; and has considerable
international exposure” (Welch and others, 2002, p. 613).
This definition comes close to the definition of experts proposed by
Meuser and Nagel in their groundbreaking 1991 article:4 “The target
group for expert interviews is wide. Examples in literature include top-
level managers from politics, business, the judiciary, associations and the
sciences, along with teachers, social workers or staff representatives. Most
100 Beate Littig

are members of the functional elite, although categorizing some of them

in this way, for example staff representatives or social workers might be
misleading.” (Meuser and Nagel, 2005, p. 73, transl. from German B.L.)
Meuser and Nagel also relate expert status to the field of research. Selected
individuals are defined as experts, a status accorded to them by the research-
ers. Social scientific interest in experts is targeted at their specific context-
ual knowledge of a given research field or their internal knowledge of the
structures, procedures and events in a given organization. In other words,
experts serve as informants and possess knowledge otherwise not access-
ible to researchers. They are often also people in positions of power, that
is managers with greater decision-making responsibility. However, they do
not necessarily have to be the people who make the high-level decisions
at the top of an organization. Ultimately, anyone who is responsible for
and has privileged access to the knowledge of specific groups of people or
decision-making processes can be seen as an expert. Bogner and Menz (in
this volume, Chapter 2) formulate this as follows: “An expert has technical,
process and interpretative knowledge that refers to a specific field of action, by vir-
tue of the fact that the expert acts in a relevant way (for example, in a particular
organisational field or the expert’s own professional area). In this respect, expert
knowledge consists not only of systematised, reflexively accessible knowledge relat-
ing to a specialised subject or field, but also has to a considerable extent the char-
acter of practical or action knowledge, which incorporates a range of quite disparate
maxims for action, individual rules of decision, collective orientations, and pat-
terns of social interpretation. An expert’s knowledge, his/her action orientations
etc., also (and this is decisive) point to the fact that s/he may become hegemonic in
terms of practice in his/her field of action (for example, in a certain organisational-
functional context). In other words, the possibility exists that the expert may be
able to get his/her orientations enforced (at least in part). As the expert’s knowledge
has an effect on practice, it structures the conditions of action of other actors in the
expert’s field in a relevant way.” (italic in the original)
The explication and reconstruction of these different forms of knowledge
and their practical consequences form the focus of expert interviews and
their subsequent analysis (cf. the articles in Bogner and others, 2005). In
contrast, “the experts” own biographical experiences and personal opinions
take very much a back seat.
The scientific objective is rarely formulated in quite such an explicit man-
ner in literature in interviewing the elite. However, even at the end of the
1960s, Dexter already drew attention to the fact that the purpose of inter-
viewing the elite was not simply to gather objective facts and knowledge.
Indeed, he presented an initial concept for a “transactional theory of the
interview” based on an interaction approach to the interview situation:
“What this means is, of course, that whether investigators wish it or not,
interviewing is a social relationship and the interviewer is part of the rela-
tionship. The interviewee’s inarticulate and unexamined conception of the
 Interviewing the Elite 101

audience guides and determines what he says.” (Dexter, 2006/1969, p. 115

italics in original, B.L).
Furthermore, interviews with the elite in Dexter’s tradition are not seen
as a precise research tool. The sampling is not representative, the statements
made by interviewees can be distorted by gaps in their memories, different
interviewees can give different information on the same topic, and so on,
for example (Richards, 1996, p. 200f).
In comparison, the German-language debate on interviews with experts
is strongly influenced by the sociology of knowledge and takes a more dif-
ferentiated approach to the status of the data obtained (Bogner and others,
2005, Hitzler and others, 1994). However, it is also clear here that there is no
one such thing as the expert interview. Indeed, there is considerable plural-
ity regarding the underlying notion of what constitutes an expert and the
fields of research in the use of this instrument.5
According to Bogner and Menz (2005, p. 36ff, see also Bogner and Menz,
in this volume), there are three different types of expert interviews:

● exploratory expert interviews (used in a relatively unknown field of

research),
● systemizing expert interviews (used to reconstruct “objective” knowledge
in a specific field), and
● theory generating expert interviews (targeted not only at the expert’s
explicit specialist knowledge, but also at his or her tacit specific inter-
pretive knowledge (know-why) and procedural knowledge (know-how)
obtained through (professional) practice).

Based on this differentiation, the understanding of interviewing the elite in

Anglo-American methodological literature can be classified as exploratory
or systemizing in nature. Members of the elite serve as sources of informa-
tion on specific areas of knowledge that would otherwise be inaccessible.
Correspondingly, and depending on the field of research, the comparative
analysis of different interviews seeks to provide an objective reconstruc-
tion of the facts, problems, decision-making processes, networks and other
aspects described by the individual interviewees for example (Welch and
others, 2002, p. 613). Richards describes a similar situation in the polit-
ical sciences: “One of the most important functions of an elite interview
is to try to assist the political scientist in understanding the theoretical
position/s of the interviewee; his or her perceptions, belief and ideologies.
Such information can rarely be gleaned from examining books, documents
or records. By their very nature, elite interviews provide a subjective account
of an event or issue” (Richards, 1996, pp. 99–200).
Even if the effects of interaction cannot be ignored, they should none-
theless be kept to a minimum to enable maximum proximity to scien-
tific insight into a positivistically accepted, objective reality (cf. Dexter,
102 Beate Littig

2006/1969, p. 115ff.). This is the prevailing goal of the use of expert

interviews – using experts as informants – at least in applied social and
policy research in the German-speaking world, as well as in political sci-
ences and many other branches of sociology (Gläser and Laudel, 2004,
Mieg and Näf, 2006).
Consequently, the dividing line does not run between expert interviews
and elite interviews and their different target data sources. Instead, it runs
far more within such interviews through their different cognitive interests
and the epistemological view of the status of the data they provide. What
divides them is their view of science and social reality, namely a positivist
versus a social constructivist, that is always interpretive hermeneutic, view:

However...the hermeneutic argumentation is in essence anti-Cartesian. ... It

is based on historical-sociological constructions of reality (realities). It sees
individuals interacting with each other and their a priori intersubjectivity in
not in contrast to the interpreted world. It is directed not only at the observ-
ing, describing, understanding and explaining of social phenomena, but
with that as one also the social aspects of the typal, historically changing
perception and articulation patterns and purposes of observing, describing,
understanding and explaining... (Soeffner 1989, p. 56f; italics in original;
transl. from German B.L.)

Whereas exploratory and systemizing expert interviews are grounded in a

more positivist attitude and, consequently, an objectivist cognitive ideal of
both the generation and the analysis of the data, the so-called theory gen-
erating expert interview proposed by Bogner and Menz in line with Meuser
and Nagel (both in Bogner and others, 2005) emanates from the assump-
tion of a social-constructivist production of reality. Ergo, research is also
viewed as an active process of creating meaning and relevance. In addition
to obtaining information on a particular area of research, theory generating
expert interviews also follow sociology of knowledge based goals. In the pro-
cess, the generation, emergence, functioning, practice, content and effect of
explicit and tacit expert knowledge alike become the object of the research.
The expert knowledge is understood here as an analytical construct realized
depending on the interview strategy in an interactive interview situation.
In this respect, the expert knowledge is not simply neatly packaged up wait-
ing to be collected, it is an externalization of different levels of knowledge
that is only formed in the course of the interview. According to this social-
constructivist oriented view, an (expert) interview is not a “neutral” inter-
view situation from which interaction effects (through gender, age, status,
and so on) should be excluded wherever possible.6 Such influences are not
only unavoidable, they are also – just like expectations, emotions, sym-
pathy, antipathy, and so on – a constituent part of an (expert) interview
(Helfferich, 2004).7 As with all other interview forms, the interviewees (and
 Interviewing the Elite 103

the interviewers) in expert interviews should orient their behaviour (and

what they say) on those issues that are of specific relevance to the actual
situation in question.

4.3 Specifics of access and interviewing

Literature on both forms of interview, that is interviews with the elite and
interviews with experts often focuses on the issues of sampling, the spe-
cific access problems faced and the challenges of conducting the interviews
for example (Dexter, 2006/1969, Moyser and Wagstaffe, 1987, Vogel, 1995,
Odendahl and Shaw, 2002, Welch and others, 2002, Lilleker, 2003 and the
articles in Bogner and others, 2005).8

4.3.1 Sampling
The first issue usually discussed is sampling. This does not adhere to quanti-
tative conceptions of representativity, since there is no clearly defined pool
of experts and members of the elite from which a sample might be chosen in
line with specific guidelines. Indeed, the attributed expert or elite status is
more often set by the actual field of research and research goals. As Meuser
and Nagel (2005, p. 73) note, researchers to a certain extent attribute expert
status that is limited to a specific area of research. If the research focuses on
corporate human resources issues (for example hiring or redundancy prac-
tices), human resources managers and directors, managing directors, and
also even representatives of specific lobbying groups (such as associations
for the disabled) take on expert status. If it focuses on the drafting of a par-
ticular law and the related negotiation and decision-making processes, the
civil servants, party functionaries, assessors and – where applicable – repre-
sentatives of affected citizens’ action groups, and so on are the experts to
interview. Welch (and others, 2002, p. 613) also describe the attribution of
elite status to certain individuals in a similar manner, namely in relation to
the research question (in their case international business). Their interest
focused primarily on the long-term professional experiences of 90 interview-
ees in middle management positions in international companies. Notable in
their description is the explicit attribution of elite status to representatives
of middle not top management, a group typically viewed as a reservoir of
the elite. They base this extension of the notion of the elite on their topic
of research: their interest lay in operational matters, and they expected to
encounter more knowledge of such activities at middle management level
than at the top. The actual interviewees were identified and recruited in this
case via company profiles (now available on the internet), telephone direc-
tories, media reports, prior studies and so-called snowball sampling (that is
recommendations from other interviewees). To ensure no important people
were omitted from the sample, extensive experience in the particular field
was a prerequisite where participation in certain (for example legislative)
104 Beate Littig

processes was concerned. Recommendations from interviewees can prove

useful in assessing the importance of particular individuals.9 Theoretical
sampling (Strauss and Corbin, 1996, Glaser and Strauss, 1998) is an estab-
lished selection procedure in cases involving issues linked to specific pos-
itions (for example middle management in large companies).

4.3.2 Access problems

As one way of distinguishing themselves, members of the elite seek to dis-
tance themselves from the non-elite by setting up access barriers to their
private and working lives (Hertz and Imber, 1995, p. viii). The following
rule seemingly applies: the higher the social class, the more difficult access
becomes.10 This also applies to the high-level experts described in the pre-
vious section. These access barriers can be personified in the form of secre-
taries, personal assistants and, in the case of the economic and political top
elite, entire PR departments. The researcher has to quasi overcome this first
hurdle by convincing these guardians of the meaningfulness and necessity
of an interview with their respective superiors. Lack of time is another pos-
sible access barrier. High-level individuals are frequently obliged to keep to
a tight schedule and thus give strict order of precedence to important and
unimportant appointments and activities (Brandl and Klinger, 2006, p. 47).
Scientific research does not always feature highly on this list of priorities,
and since requests for interviews with the experts who are in demand can
quickly mount up, they have to be even more selective when choosing those
they actually want to accept (Imbusch, 2003, p. 11f).
In addition to distancing themselves, members of the elite also cultivate
network contacts to others of their ilk. Although such networks are still
characterized by a strong tendency towards social closure and homogeniza-
tion, access barriers to elite networks have, in the meantime, become more
penetrable (Hornbostel, 2004, p. 12, Hartmann, 2007). If the researcher suc-
ceeds in gaining initial access, elite networks can prove a rich source of
information for scientific research.
A number of strategic suggestions on how to overcome these access
problems can be found in literature on both expert interviews and elite
interviews alike. Such suggestions are often founded on exploiting the pos-
sible motives of potential interviewees for participation in such interviews.
Brandl and Klinger (2006), for example, refer to instrumental interests on
the part of the target group – such as the hope of gaining useful or utilizable
information in the course of the interview. Advantage can be taken of such
a motive by offering quick access to the research results in return for partici-
pation. A further instrumental motive is the possible public relations value
to be gained from cooperating with a well-known research institute. Other
interviewees may be interested in participating for reasons independent of
the actual research project. Psychosocial motives – such as a lack of compe-
tent people to talk to or loneliness – can come into play as well. Similarly,
 Interviewing the Elite 105

altruism can also be a motivation for agreeing to participate in such inter-

views, for example a desire to support young researchers in their scientific
endeavours or generally make a contribution to scientific advance.
To enhance their own status, making targeted references to the excellent
reputation of their research institute, the sponsoring body or their supervis-
ing tutors is particularly recommended for students or lower-level scientists.
Also recommended is the use of personal contacts or references to common-
alities with the high-level interviewee (for example having studied at the
same university, originating from the same city, and so on).

4.3.3 Interviewing
The goal of both expert and elite interviews alike is to generate knowledge for
scientific purposes. In other words, the interviewees should provide infor-
mation on a specific topic related to the research. The interviews themselves
are open-ended and do not follow a standardized format, thus providing the
interviewees with ample space to express their views. Dexter mapped out
his definition of an elite interview in 1969 as follows:

It is an interview with any interviewee – and stress should be placed on

the word “any” – who in terms of the current purposes of the interviewer
is given special, nonstandardized treatment. By special, nonstandard
treatment I mean:
1. stressing the interviewee’s definition of the situation,
2. encouraging the interviewee to structure an account of the situation,
3. letting the interviewee introduce to a considerable extent (an extent
which will of course vary from project to project and interviewer to
interviewer) his notion of what he regards as relevant, instead of relying
upon the investigator’s notions of relevance. (Dexter, 2006/1969, p. 18)

Interviewers usually work with a set of flexible guidelines, which also con-
tain a list of the relevant issues. As a rule, experts and members of the elite
are accustomed to talking about their field of expertise and explaining to
others what they know – frequently for strategic purposes. Managers are also
often loath to be told what to do and tend more towards testing others. To
handle such situations, interviewers need certain skills and abilities. Firstly,
they must be extremely flexible, allowing interviewees to lead the conver-
sation, yet not losing sight of the information they are actually interested
in. At the same time, they must show themselves to be competent partners.
Pfadenhauer (2005, see also Pfadenhauer, this volume) aptly describes this
as “talking at eye level.” In other words, interviewers may even be required
to present themselves in preliminary meetings as quasi-experts and compe-
tent partners who are familiar with the expert’s area of expertise. To pursue
an interaction model based on quasi-expertise, interviewers must prepare
extensively for such meetings to ensure they are familiar with the subject
106 Beate Littig

matter, speak the “right language” and have the necessary insider know-
ledge of the field. To a certain extent, this counteracts any differences in
status between the interviewer and the interviewee that might lessen the
latter’s interest in the interview.
Bogner and Menz (2005, p. 60ff., see also Bogner and Menz, in this
volume) do however; note that interviewees can perceive their interviewers
in different ways and that differences in status are not necessarily disad-
vantageous in an interview setting. Interviewees can categorize interviewers
into the following types:

● Co-experts or experts from another knowledge culture

● Lay people
● People in positions of authority
● Accomplices
● Potential critics.

As far as the interview is concerned, each of these types has its own advan-
tages and disadvantages, which can be strengthened or reduced using
appropriate strategic interventions.11 Although not described as such in the
methodological literature on the subject, the same applies to interviews
with members of the professional elite, who (as will be discussed in more
detail below) can also be viewed as high-level experts.

4.4 Summary and conclusions

Given the above, what conclusions can be drawn on the commonalities and
differences in expert interviews and interviews with the elite?
On a practical, methodological level, the same problems are discussed in
both sets of literature. No systematic differences can be determined as far
as access to the field and actual interaction in the interview are concerned.
Similarly, there are no fundamental differences between the target groups
for expert interviews or interviews with the elite. In fact, the notion of the
expert and the notion of the elite overlap in two key criteria: the knowledge
and the power at their disposal. Yet, at the same time, both criteria also play
a decisive role in distinguishing experts from the elite (see below for further
details).
Bogner and Menz (1995, p. 40f., see also Bogner and Menz, in this volume)
differentiate and discuss a voluntarist, constructivist, sociology of know-
ledge based concept of the expert (cf. also Hitzler, 1994). Following the line
of argumentation in Bogner and Menz (2005, see also Bogner and Menz, in
this volume), simply possessing specific knowledge does not alone suffice to
determine who constitutes an expert. From a sociology of knowledge per-
spective, professional or occupational experts have such special or specialized
knowledge. Sprondel (1979) refers to people with specialized professional or
 Interviewing the Elite 107

occupational knowledge as experts. He differentiates between such experts

and lay specialists, who also possess specialized knowledge, but not related
to their work or occupation for example (DIY hobbyists or amateur sports
people). For the purposes of linguistic differentiation, individuals with occu-
pational expertise should be referred to as specialists (Hitzler, 1994, p. 21f.).
Aside from their specific knowledge, what sets experts apart is the possibil-
ity that their knowledge – at least in principle – will also have a practical
effect and, thus, significantly influences the freedom of others to act. At the
same time, this concept of the expert distinguishes itself from the voluntar-
ist definition, whereby each individual is the expert in his or her own life
(cf. Gläser and Laudel, 2004). However, expanding the notion of the expert
in this way blurs existing differences in knowledge and evens out differ-
ences in social status. Furthermore, such a broader concept of the expert
also raises methodological questions. If, as a consequence, every interview
were then considered an expert interview, what is the difference between
expert interviews and narrative, biographical, focused, topic- oriented or
other forms of interview?
These considerations lead to the following conclusions as far as differenti-
ating between expert and elite interviews is concerned: if the expertise (that
is specific interpretive knowledge (“know-why”) and procedural knowledge
(“know-how”) in a particular occupational or professional field) is central
to the area of research, then the interview can be regarded as an expert
interview (Bogner and Menz, 2005, p. 46, see also Bogner and Menz, in this
volume). People should only be referred to as experts if they also have a cer-
tain degree of power. This vague formulation not only permits a restriction
of the notion of the expert to top-level economic, political and governmen-
tal (that is the elite) decision-makers (Meuser and Nagel, 2005), it also, at
the same time, permits a differentiation between experts and specialists or
(specialized) lay people.
Further distinguishing between the elite and experts requires a differ-
entiation of the concept of power: experts can have both formative and/or
interpretive power (Bogner, 2005, p. 201f.). According to Bogner’s (ibid.) line
of argumentation, having formative power means having the authority to
establish socially binding definitions of problems and predetermine solu-
tions. This can apply both to individual decisions concerning the imme-
diate professional or occupational environment (for example in the case of
medical directives or work procedures) as well as to broader matters of social
control (for example the ordering of compulsory mass screenings or, more
generally, the passing of legal acts). In contrast, interpretive power describes
the opportunity open to the expert to provide and establish significant
terms and concepts for interpreting phenomena, for legitimizing decisions
and, thus, ultimately for the social confrontation with certain phenomena
(Bogner, 2005, p. 201f). By way of example, Bogner (ibid.) refers to prenatal
diagnostics, where medical expertise is of legitimate relevance in diagnosing
108 Beate Littig

possible disabilities. In addition (for example in genetic consultation), such

expertise serves as the central guideline and, thus, plays a decisive role in
any decisions taken. Ultimately, it is argued, this differentiation between
the different ways of exercising power is relevant for distinguishing between
experts and the elite. Hence, the elite have more formative power, because
they occupy the positions in which the higher decisions legitimized by this
form of power are taken. However, they do not necessarily have significant
interpretive power because the experts – with their notions, concepts and
relevance – have established themselves the opportunity of conferring or
starting to confer meaning to decision and negotiation processes (Bogner,
2005, p. 201f.). So-called knowledge workers exert a far higher degree of
interpretive power, although they may not necessarily have far-reaching for-
mative power (Willke, 1998).
Figure 4.1 illustrates the differences between specialized lay people, spe-
cialists, experts and the elite with regard to knowledge and power.
A summary of the ideas presented in this article leads to the conclusion
that the answer to the question posed in the title is yes: there are indeed
differences between interviews with the elite and interviews with experts,
but also commonalities.
Interviews with the elite can but do not necessarily also have to be expert
interviews. Experts, who are defined by their occupational or professional

The elite
Power

Experts

Specialized
lay people

Specialists

(Expert) knowledge

Figure 4.1 Differentiating between experts and the elite

 Interviewing the Elite 109

knowledge and their decision-making competences, can but do not have to

be members of an elite group. This depends on the extent of their power to
act (in the sense of formative power). The ultimate decision on whether an
interview is an expert interview or not depends on the field of research and
interest. When it comes to researching professional or occupational explicit
or tacit interpretive knowledge (know-why), procedural knowledge (know-
how) and process knowledge, experts (even elitist experts) are the obvious
interview candidates. When the focus lies on issues relating to elite research,
questions regarding membership, the (re)production, career paths, social
power, lifestyles and thought patterns, and so on of the elite (cf. Wasner,
2004), preference will be given (along with other data) to focused, biograph-
ical, subject-matter oriented interviews with representatives of the elite (Mey
and others, 2004). In many cases, research into the elite is confronted with
similar methodological problems to those encountered in the interviewing
of experts. If the actual experts in question are also members of the elite,
the access problems will be considerable.12
The interview also requires careful preparation on the part of the inter-
viewer: a set of guidelines must be drawn up and extensive knowledge
obtained on the actual area of expertise to indicate the interviewer’s compe-
tence to the interviewee. The latter is presumably of particular importance
in the case of experts with a high degree of interpretive power.

Acknowledgements

I would like to thank all the participants in my “Interviewing Experts”

course at the ECPR (European Consortium for Political Research) meth-
odology summer school organized in cooperation with the University of
Ljubljana in 2006 and 2007 for the inspiring debates on the commonal-
ities and differences between expert interviews and interviews with the
elite.

Notes
1. A German Version of this article has also been published in the Open Access
Journal Forum Qualitative Research as Littig, B. (2008). Interviews mit Eliten –
Interviews mit ExpertInnen: Gibt es Unterschiede? [37 Absätze]. Forum Qualitative
Sozialforschung/Forum: Qualitative Social Research, 9(3), Art. 16, http://nbn-
resolving.de/urn:nbn:de:0114-fqs0803161.
2. The new edition of Dexter’s 1969 book “Elite and Specialized Interviewing” was
published in 2006 by the European Consortium for Political Research (ECPR) in
the ECPR Classics series. The political scientist’s essay-like book significantly influ-
enced the English-language methodological debate on interviewing the elite and,
despite the complexity that has since entered this debate, many of his arguments
are still valid to this day. In this respect, the book can rightfully be described as a
classic (cf. Littig, 2008).
110 Beate Littig

3. One popular definition refers to the elite as a “group in society considered to

be superior because of the power, talent, privileges and so on of its members”
(Hornby, Cowie and Gimson, 1983, p. 280). However, sociological studies of the
elite, link this group not only to social, economic and political influence but
also to position (Hartmann, 2007, p. 18, Wasner, 2004). The linking of elite sta-
tus to a specific office or position (and not an individual) provides continuity
to the power held by top politicians, managers or high-level economic officials.
Contrary to popular belief, this excludes people like artists, scientists, sportsmen
and women, pop stars, actors or other celebrities from the elite. In contrast to
Hartmann, Bude maintains that the notion of the elite needs to be taken further
if it is to encompass a required performance condition, noting that the Chairman
of the Board of Daimler Chrysler is just as much a member of the German elite as
the Chairman of the Federal Government Finance Committee. The same applies
to the publisher of the Feuilleton in the Frankfurter Allgemeine Zeitung and the
Head of Entertainment at the SAT1 television station or Günter Grass and Franz
Beckenbauer (Bude, 2004, p. 300f).
4. The 1991 article entitled “Expert interviews – often tested, less considered”
(in German: “ExpertInneninterviews – vielfach erprobt, wenig bedacht” –
translation from German B.L.) is reprinted in Bogner and others (2005). But it
would take about ten years for Meuser and Nagel’s conclusions to gain momen-
tum (cf. the anthology on expert interviews by Bogner and others, 2005/2002 or
Gläser and Laudel, 2004).
5. The articles in the anthology by Hitzler, Honer and Maeder (1994) offer a detailed
discussion of what constitutes an expert and what differentiates experts from
specialists and lay people from a sociology of knowledge perspective (cf. also
Bogner and Menz 2005, p. 40f.). In addition, Mieg and Brunner (2004) discuss
various cognitive psychological conceptions of the expert.
6. For more on the meaning of gender in expert interviews see Littig, 2005 and
Abels and Behrens, 2005.
7. They can even be used as a resource in conducting the interview (cf. various art-
icles in Bogner and others, 2005). See also the following section.
8. In the case of interviews with the elite, this often takes the form of essay-like
experience reports from research practice with hints on how to address these dif-
ficulties (for example, Dexter, 2006/1969, also Lilleker, 2003 or Richards, 1996).
9. The fact that high-level experts and members of the elite frequently occupy
exposed positions often complicates the issue of anonymity in such interviews.
Nonetheless, anonymity and confidentiality in the treatment of interview mater-
ial should be assured, except with the express permission to the contrary of the
interviewees. Indeed, Dexter (2006/1969, Chapter 2) first cautioned against the
unauthorized use of interview material for teaching purposes over 30 years ago.
10. Almost the reverse can be applied to the non-elite: the lower the social status,
the easier the entry to the field, even if establishing trust is not always easy
where the non-elite are concerned either. This is perhaps one of the reasons
why the living conditions of the poor have been studied from a social sciences
perspective to such a greater extent than those of the rich.
11. Abels and Behrens (2005, see also Abels and Behrens in this volume) came to
comparable conclusions for gender differentiating interaction effects.
12. One aspect that should, in any case, be weighed up with regard to the research
question is whether the anticipated results justify the actual effort involved in
obtaining an interview. In some cases, an interview with representatives of the
lower hierarchical levels may prove more fruitful.
 Interviewing the Elite 111

Further readings
Dexter, L. A. (2006/1969) Elite and specialized interviewing, with a new introduction by
Ware, A. and Sánchez-Jankowski, M. (University of Essex, Colchester: ECPR Press –
ECPR classics, 1st edn (1969) Evanston: Northwestern University Press).
Welch, C., Marschan-Piekkari, R., Penttinen, H. and Tahvanainen, M. (2002)
“Corporate elites as informants in qualitative international business research” in
International Business Research Review 11, 611–28.
Weiss, R. S. (1995) Learning from Strangers. The Art and Method of Qualitative Interview
Studies (New York: The Free Press).

References
Abels, G. and Behrens, M. (2005) “ExpertInnen-Interviews in der Politikwissenschaft.
Geschlechtertheoretische und politikfeldanalytische Reflexion einer Methode” in
Bogner, A., Littig, B. and Menz, W. (eds) Das Experteninterview – Theorie, Methode,
Anwendung, 2nd edn (Wiesbaden: Verlag für Sozialwissenschaften), pp. 173–90.
Bogner, A. (2005) Grenzpolitik der Experten. Vom Umgang mit Ungewißheit und Nichtwissen
in pränataler Diagnostik und Beratung (Weilerswist: Velbrück Wissenschaft).
Bogner, A. and Menz, W. (2005) “Das theoriegenerierende Experteninterview.
Erkenntnisinteresse, Wissensformen, Interaktion” in Bogner, A., Littig, B. and
Menz, W. (eds) Das Experteninterview – Theorie, Methode, Anwendung, 2nd edn
(Wiesbaden: Verlag für Sozialwissenschaften), pp. 33–70.
Bogner, A., Littig, B. and Menz, W. (eds) (2005) Das Experteninterview – Theorie, Methode,
Anwendung, 2nd edn/1st edn (2002) (Wiesbaden: Verlag für Sozialwissenschaften).
Brandl, J. and Klinger, S. (2006) “Probleme des Feldzugangs zu Eliten” in Österreichische
Zeitschrift für Soziologie 31(1), 44–65.
Bude, H. (2004) “Elitewechsel. Deutsche Führungsgruppen zwischen ‘Bonner’ und
‘Berliner Republik’ ” in Hitzler, R., Hornbostel, S. and Mohr, C. (eds) Elitenmacht
(Wiesbaden: Verlag für Sozialwissenschaften), pp. 295–313.
Denzin, N. and Lincoln, Y. S. K. (eds) (2000) Handbook of Qualitative Research, 2nd edn
(Thousand Oaks: Sage Publications).
Dexter, L. A. (2006/1969) Elite and specialized interviewing, with a new introduction by
Ware, A. and Sánchez-Jankowski, M. (University of Essex, Colchester: ECPR Press –
ECPR classics, 1st edn (1969) Evanston: Northwestern University Press).
Flick, U., Kardorff, E. v. and Steinke, I. (eds) (2003) Qualitative Sozialforschung. Ein
Handbuch, 2nd edn (Reinbek bei Hamburg: Rowolth).
Glaser, B. G. and Strauss, A. M. (1998) Grounded Theory. Strategien qualitativer Forschung,
(Göttingen: H. Huber).
Gläser, J. and Laudel, G. (2004) Experteninterviews und qualitative Inhaltsanalyse
(Wiesbaden: Verlag für Sozialwissenschaften).
Goffman, E. (1969) Wir alle spielen Theater (München: Piper).
Gubrium, J. and Holstein, J. (eds) (2002) Handbook of Interview Research: Context and
Methodology (Thousand Oaks, CA: Sage Publications).
Hartmann, M. (2007) Eliten und Macht in Europa. Ein internationaler Vergleich (Frankfurt
am Main: Campus).
Helfferich, C. (2004) Die Qualität qualitativer Daten. Manual für die Durchführung
qualitativer Interviews (Wiesbaden: Verlag für Sozialwissenschaften).
Hertz, R. and Imber, J. B. (eds) (1995) Studying Elites Using Qualitative Methods
(Thousand Oaks, CA: Sage Publications).
112 Beate Littig

Hitzler, R. (2004) “Wissen und Wesen des Experten. Ein Annäherungsversuch – Zur
Einleitung” in Hitzler, R., Hornbostel, S. and Mohr, C. (eds) Elitenmacht (Wiesbaden:
Verlag für Sozialwissenschaften), pp. 13–30.
Hitzler, R., Hornbostel, S. and Mohr, C. (eds) (2004) Elitenmacht (Wiesbaden: Verlag
für Sozialwissenschaften).
Hitzler, R., Honer, A. and Maeder, C. (eds) (1994) Expertenwissen. Die institutionalisi-
erte Kompetenz zur Konstruktion von Wirklichkeit (Opladen: Leske und Budrich).
Hornbostel, S. (2004) “Denn viele sind berufen, aber wenige sind auserwählt” in
Hitzler, R., Hornbostel, S. and Mohr, C. (eds) Elitenmacht (Wiesbaden: Verlag für
Sozialwissenschaften), pp. 9–21.
Hornby, A. S., Cowie, A.P. and Gimson, A.C. (eds) (1983) Oxford Advanced Learner’s
Dictionary (Oxford: Oxford University Press).
Imbusch, P. (2003) “Konjunkturen, Probleme und Desiderata sozialwissenschaftli-
cher Eliteforschung” in Hradil, S. and Imbusch, P. (eds) Oberschichten – Eliten –
Herrschende Klassen (Opladen: Leske und Budrich), pp. 11–34.
Lilleker, D. G. (2003) “Interviewing the political elite: Navigating a potential mine-
field” in Politics 2003, 23(3), pp. 207–14.
Littig, B. (2005) “Interviews mit Experten und Expertinnen. Überlegungen aus
geschlechtertheoretischer Sicht” in Bogner, A., Littig, B. and Menz, W. (eds) Das
Experteninterview – Theorie, Methode, Anwendung, 2nd edn (Wiesbaden: Verlag für
Sozialwissenschaften), pp. 191–206.
Littig, B. (2008) Rezension zu: Lewis A. Dexter (2006) “Elite and specialized
interviewing” in With a new introduction by A. Ware and M. Sánchez-Jankowski
[16 articles] (Forum Qualitative Sozialforschung/Forum: Qualitative Social Research),
9(1), Art. 5., http://www.qualitative-research.net/fqs-texte/1-08/08-1-5-d.htm.
Meuser, M. and Nagel, U. (1991) “ExpertInneninterviews – vielfach erprobt, wenig
bedacht. Ein Beitrag zur qualitativen Methodendiskussion” in Garz, D. and
Kraimer, K. (eds) Qualitative empirische Sozialforschung: Konzepte, Methoden, Analysen
(Opladen: Westdeutscher Verlag), pp. 441–71.
Meuser, M. and Nagel, U. (2005) “ExpertInneninterviews – vielfach erprobt, wenig
bedacht” in Bogner, A., Littig, B. and Menz, W. (eds) Das Experteninterview – Theorie,
Methode, Anwendung, 2nd edn (Wiesbaden: Verlag für Sozialwissenschaften),
pp. 71–93.
Mey, G., Puebla, C. C. A. and Faux, R. (eds) (2004) “Interviews” in Forum Qualitative
Sozialforschung 5(3), Special Issue, http://www.qualitative-research.net/fqs/fqs-d/
inhalt3-04-d.htm, date accessed in June 2008.
Mieg, H. A. and Brunner, B. (2004) “Experteninterviews. Reflexionen zur Methodologie
und Erhebungstechnik” in Schweizerische Zeitschrift für Soziologie 30, 1999–2022.
Mieg, H. A. and Näf, M. (2006) Experteninterviews in den Umwelt- und
Planungswissenschaften. Eine Einführung und Anleitug (Lengerich: Papst).
Moyser, G. and Wagstaffe, M. (eds) (1987) Research Methods for Elite Studies (London:
Allen and Unwin).
Odendahl, T. and Shaw, A. M. (2002) “Interviewing elites” in Gubrium, J. and
Holstein, J. (eds) Handbook of Interview Research: Context and Methodology (Thousand
Oaks, CA: Sage Publications), pp. 299–316.
Richards, D. (1996) “Elite interviewing: Approaches and pitfalls” Politics 16(3),
199–204.
Pfadenhauer, M. (2005) “Das Experteninterview – ein Gespräch zwischen Experte und
Quasi-Experte” in Bogner, A., Littig, B. and Menz, W. (eds) Das Experteninterview –
Theorie, Methode, Anwendung, 2nd edn (Wiesbaden: Verlag für Sozialwissenschaften),
pp. 113–30.
 Interviewing the Elite 113

Seldon, A. (1996) “Elite Interviews” in Brivati, B. and others (eds) The Contemporary
History Handbook (Manchester: Manchester University Press), pp. 353–65.
Soeffner, H-G. (1989) “Anmerkungen zu gemeinsamen Standards standardisierter
und nicht-standardisierter Verfahren in der Sozialforschung” in Soeffner, H-G.
Auslegung des Alltags – Der Alltag der Auslegung. Zur wissenssoziologischen Konzeption
einer sozialwissenschaftlichen Hermeneutik (Frankfurt am Main: Suhrkamp),
pp. 51–65.
Sprondel, W. M. (1979) “ ‘Experte’ und ‘Laie’: Zur Entwicklung von Typenbegriffen in
der Wissenssoziologie” in Sprondel, W. and Grathoff, R. (eds) Alfred Schütz und die
Idee des Alltags in den Sozialwissenschaften (Stuttgart: Enke), pp. 140–54.
Strauss, A. and Corbin, L. and J. (1996) Grounded Theory: Grundlagen qualitativer
Sozialforschung (Weinheim: Beltz/PVU).
Vogel, B. (1995) “ ‘Wenn der Eisberg zu schmelzen beginnt ...’ – Einige Reflexionen
über den Stellenwert und die Probleme des Experteninterviews in der Praxis der
empirischen Sozialforschung” in Brinkmann, C., Deeke, A. and Völkel, B. (eds)
Experteninterviews in der Arbeitsmarktforschung. Diskussionsbeiträge zu methodischen
Fragen und praktischen Erfahrungen (Nürnberg: Landesarbeitsamt Nordbayern,
Geschäftsstelle für Veröffentlichungen), pp. 73–83.
Wasner, B. (2004) Eliten in Europa. Einführung in Theorien, Konzepte und Befunde
(Wiesbaden: Verlag für Sozialwissenschaften).
Welch, C., Marschan-Piekkari, R., Penttinen, H. and Tahvanainen, M. (2002)
“Corporate elites as informants in qualitative international business research” in
International Business Research Review 11, 611–28.
Willke, H. (1998) “Organisierte Wissensarbeit” in Zeitschrift für Soziologie 27(3),
161–77.
This page intentionally left blank
Part II
Methodological Practice:
Generating Data
This page intentionally left blank
5
On Interviewing “Good”
and “Bad” Experts
Jochen Gläser and Grit Laudel

5.1 The “quality” of interviewees matters

The success of interview-based investigations considerably depends on the

“quality” of the interviewees, that is on the extent to which they meet our
expectations in the interview situation. We expect interviewees to under-
stand which information we need, to provide this information in exten-
sive, complete and detailed responses, and to adjust their communication to
our steering of the conversation. We also hope to meet respondents which
reflect on their own social situation and who are able to provide informa-
tion about their perceptions, social relations and motives.1
These aspects of “quality” are particularly important in interviews that
are intended to “mine” a respondent’s special knowledge about a social situ-
ation or a social process. We consider all interviews that have this function
to be “expert interviews.” Thus, we define “experts” as people who possess
special knowledge of a social phenomenon which the interviewer is inter-
ested in, and expert interviews as a specific method for collecting data about
this social phenomenon. This conceptualization of the “expert interview”
is based on the expert role of the interviewee in the interview. Thus understood,
expert interviews are a distinct method that is applied in investigations
of a specific type, namely investigations that reconstruct social situations
or processes and use interviewees as a source of information (Gläser and
Laudel, 2009, pp. 11–14). This understanding of “expert interviews” has
been introduced to the German methodological discussion by Hopf (1993).
In the Anglo-American literature, the distinction between the roles of inter-
viewees as sources of information respectively objects of study is reflected
by the concepts of “informants” and “respondents” (for an early reference to
this distinction see Zelditch, 1962).
Another approach to “expert interviews” is their conceptualization as
interviews with people who have an expert role in the investigated social set-
ting. In that perspective, which dominates the wider literature as well as the
contributions to this volume, experts are people who are set apart from other

117
118 Jochen Gläser and Grit Laudel

actors in the social setting under investigation by their specific knowledge

and skills. Such superior knowledge is usually produced by designated proc-
esses of learning and training (for example vocational training). Members
of professions such as physicians, lawyers, or architects are the best-known
examples of “trained” experts. However, expert roles in social settings are
not limited to the professions.
This understanding of experts ties the concept “expert interview” to a
specific kind of respondents or informants but does not limit the forms or
functions of interviews with them. It also focuses the methodological inter-
est on the particularities of an interview that result from the expert status
of the interviewee in the investigated field. Thus understood, the expert
interview is not a specific method of data collection but includes all forms
of qualitative interviews that are conducted with experts. In our opinion,
this is the reason why there is no specific methodological discussion of the
expert interview as a method of data collection, as has been noted by Bogner
and Menz (Bogner and Menz, 2005, pp. 11–16).
Since the roles of “interviewees as experts” and “experts as interviewees”
are constructed in different social situations (the interview respectively the
investigated social settings), the two definitions of expert interviews are not
disjunct (Figure 5.1). In many social science investigations only one of the
roles is relevant. The left (white) area in Figure 5.1 represents investigations
in which interviews are aimed at the reconstruction of social processes and
situations but are conducted with people who don’t have an expert role in the
field. For example, interviews about health care could be conducted not only
with physicians but also with patients. The latter don’t have an acknowledged
expert role in health care but possess special knowledge about the way health
organizations handle patients and thus would be interviewed as experts for
this aspect of the investigation. The right (black) area stands for interviews
with people who have an expert role in the investigated social field but who
are not interviewed as sources of information about social processes in which
they participate. An example of this would be a study of biographical self-
representations of physicians, where the reflection of social processes by
physicians rather than the processes themselves are the subject matter of the
investigation. In the middle (grey area) we find the area of overlap between
the two definitions. The overlap refers to interviews with “experts in the field”
who serve as sources of information about a part of the social reality.
For this specific group of interviews an additional aspect of the “quality”
of interviewees becomes important. If we use “experts in the field” as sources
of information, our interviews depend not only on their performance in
their role as interviewees but also on their “quality” in their expert role in
the field. We must ask ourselves whether we receive the same information
about medical treatments from good and bad physicians, the same infor-
mation about baking from good and bad bakers, and the same information
about competitive sports from world champions and average athletes.
 Interviewing ‘Good’ and ‘Bad’ Experts 119

Interviews for other aims of

Interviews for the reconstruction of social issues
investigation

Interviews with people who Interviews with holders of expert roles in the field
have no expert role in the field

‘Experts in the field’

as a source of
information for the
reconstruction of
social issues

Figure 5.1 Area of overlap of the two definitions of expert interviews

Source: Gläser and Laudel, this volume.

This problem has not yet explicitly been dealt with in the literature
even though it might be implicit to discussions about interview responses.
We address it in this chapter by using examples from our own empirical
investigations, in which we must compare and synthesize statements of
researchers. In doing so, we face the above-described quality problem: Do
“good” researchers describe their situations in other ways than “bad” ones?
What does it mean that certain issues are described in the same way by good
and by bad researchers? How can we find out how “good” our informants
are in their researcher roles?

5.2 “Good” and “bad” researchers as information

sources of science studies

There can be little doubt that researchers are “experts in the field.” It also is
no secret that the abilities of researchers differ and that the quality differ-
ences between very good (“excellent”) and medium or even bad researchers
are huge. Nevertheless, science studies hardly ever took the abilities of the
investigated researchers into account. The abilities of researchers who act
as informants in qualitative investigations have not been a topic of meth-
odological reflection at all. This may be due to the fact that outsiders are
hardly able to judge the abilities of a researcher. Indeed, there are only very
few researchers who are regarded as outstanding and are well-known out-
side science, such as Nobel price winners and – to a lesser extent – members
120 Jochen Gläser and Grit Laudel

of the learned academies. Below this threshold of visibility the quality of

researchers remains hidden in the implicit judgements of their scientific
communities.
Two kinds of empirical science studies can be distinguished according
to their handling of the quality problem. Most studies ignore the prob-
lem of quality differences between respondents. The first interview-based
investigations are classical examples of this neglect (for example Hagstrom,
1965, Crane, 1972). The subsequent ethnographic studies also ignored the
“quality” of the observed researchers and possible influences of this aspect
on the outcomes of observations (Knorr-Cetina, 1984, Lynch, 1985, Latour
and Woolgar, 1986 [1979]). This is surprising because we cannot a priori
assume that the observed processes of the social construction of knowledge
is performed the same way by “good” and “bad” researchers. The labora-
tory studies are implicitly based on this very assumption because they
do not include the quality as intervening factor – in spite of the obvious
importance of researchers’ abilities to the observed scientists. The quality
discourse among the observed scientists was reported (Latour and Woolgar,
1986 [1979], pp. 163–65) but the quality problem neither influenced the
empirical strategy of laboratory studies nor their interpretation. Since the
constructivist tradition – which constitutes the mainstream of the sociology
of science since the 1980s – still disregards the quality of the makers of sci-
entific knowledge, we still don’t know for sure whether good researchers
construct knowledge differently from bad ones.
A second group of studies chose research performance as its topic and
made the performance levels (productivity, creativity) of researchers the
dependent variable of their investigations. The aim of these studies was to
identify influences on the productivity and the quality of scientific work.
While investigating the influence of organizational research conditions
on the productivity of scientists, Pelz and Andrews included the quality –
measured by the number of publications – as a dependent variable (Pelz
und Andrews, 1966). These kinds of studies were taken up later in the con-
text of investigating scientific excellence and scientific creativity (Jackson
and Rushton, 1987, Zuckerman and Cole, 1994). In her study of Nobel Prize
winners, Zuckerman did not encounter the problem of quality differences
because the study was limited to a homogeneous group of a visible elite
(Zuckerman, 1977). Thus, she did not address quality differences in her
methodological reflections but instead wrote about the specifics of inter-
views with members of the elite (Zuckerman, 1972).
With the development of the Science Citation Index it became possible to
use a more sophisticated measure of quality, namely the frequency with
which publications are cited by other scientists. This measure was first
introduced by Cole and Cole who started to treat quality not only as a
dependent but also as an independent or intervening variable (Cole and
Cole, 1967, Cole, 1970, Cole and Cole, 1972). Their use of citations created
 Interviewing ‘Good’ and ‘Bad’ Experts 121

a new research tradition within bibliometrics for which the measurement

of the quality of researchers, organizations and national science systems is
an important topic. Shortly thereafter, research policy became interested
and began to generate an ever-increasing demand for performance meas-
urements. Today these measurements also utilise other than bibliometric
indicators, particularly the amount of external research funding. All these
indicator measures are somewhat problematic, only work in some research
fields, and are of questionable validity at the level that is of interest here,
namely that of the individual researcher (Laudel, 2005, Gläser and Laudel,
2007a).
This short overview demonstrates that science studies have either ignored
the varying performance levels of researchers or turned them into the sub-
ject matter of their investigations. The methodological question about dif-
ferent responses from “good” and “bad” researchers and the consequences
of such differences for studies that are based on interviews with researchers
has not yet been asked.

5.3 The dependence of data

on the performance levels of researchers

5.3.1 Different phenomena, different perceptions

In our own research we are interested in the influence of institutional and
institutionally produced conditions of research on the content of knowledge
production. We use qualitative interviews with researchers and research
managers as the main method of data collection. We ask researchers about
their research projects (the problems they are working on, the methods and
objects they use, their collaborations, and so on) and about the specific
resource requirements of these projects. Another set of questions focuses
on the conditions of research, particularly the available time for research,
access to resources and possible influence of the organizational environ-
ment on the content of research. From the responses to these questions
we reconstruct the changes in the content of research that result from the
adaptation of topics, objects, methods and collaborations to the conditions
of work.
We had to learn that in our investigations talking to “good” or “bad”
researchers makes a difference. The “quality” of the interviewees influences
our investigations at two levels. Firstly, the situations of “good” and “bad”
researchers may differ, for example with regard to the availability of external
funding, the workload, or the status of the interviewees in their organiza-
tion. In this case, the researchers experience different phenomena. Secondly,
it is possible that “good” and “bad” researchers provide different descriptions
of the experienced phenomena because their perceptions or communication
behaviour differ. If we combine these two variations we can construct four
variants of communication in the interview (Table 5.1).
122 Jochen Gläser and Grit Laudel

Table 5.1 Quality-dependent communication situations in interviews with researchers

Descriptions of the phenomena

Same Different

Same (False) implicit premise Appears as contradiction

of science studies in the responses. Main
problem: identifying the
Phenomena “real” situation
Different Not visible in the Most common situation.
responses. Main Main problem: Recognizing
problem: Recognising how the data are
that a difference exists influenced by quality
differences

The left upper cell of the table describes the implicit premise of science
studies, namely that all responses are equal in both dimensions.
Of course, nobody has ever said that this is actually the case. However,
disregarding the quality problem in interview-based and ethnographic
studies means that data collection and analysis is in fact based on this
presumption.
The common situation in science studies is the direct opposite of this
presumption. It is situated in the right lower cell of the table, indicating
that the “good” and the “bad” researchers experience different things and
provide descriptions of their experiences that are influenced by their “qual-
ity.” For example, a historian explained during the interview that he can
freely choose his topics because the publishers take everything that he
writes. In contrast, another historian described that his book project has
been rejected by several publishers, and therefore he will now change the
topic of his research. The publishers neither requested reviews about the
book project from peers nor did they tell him the reasons for the rejections.
He assumed that the topic was regarded as not profitable. In the course of
the ethnographic observation both historians were interviewed. We found
a clear quality difference between the two historians, which let us assume
that the insufficient quality of the book project was the reason for the
repeated rejections and that the reference to the topic as reason was merely
a rationalization of the respondent. However, the ethnographic observa-
tion also showed that publishers indeed reject books for thematic reasons.
Given that the two historians worked in two entirely different areas, a mix
of different experiences and quality differences occurred that was difficult
to untangle.
In the “standard case,” the complexity of an analysis is dramatically
increased because two dimensions must be added in the interpretation of
interview statements. We avoid this case in our following demonstration of
 Interviewing ‘Good’ and ‘Bad’ Experts 123

the problem because its discussion would require a large amount of back-
ground knowledge that cannot be provided in a book chapter. Instead, we
use the two “pure” cases that only vary in one of the two dimensions. In
our experience the pure cases are rare – it does not often happen that “good”
and “bad” researchers give identical descriptions of different phenomena
or different descriptions of the same phenomena. However, in our current
investigation we conducted enough interviews to obtain these kinds of
responses. We use them in order to demonstrate how difficult the standard
situation is. It is the right lower cell of the table that describes the real chal-
lenge because in most situations we are confronted by different descriptions
that are likely to point to different phenomena. It is up to us to determine
how the performance levels of interviewees influences their descriptions.

5.3.2 Identical descriptions of different phenomena

Our interviews contained several questions that targeted adaptations of the
research content to the resource and institutional conditions of research.2 We
intentionally phrased these questions in an indirect and open way in order
to avoid an orientating influence on the interviewee. One of the questions
was: “Are there research topics that you would like to work on but can’t?” The
answers to this question varied considerably. The most frequent answer was
yes, there are indeed research problems that would be nice to work on but
there is not sufficient time. Other interviewees mentioned their resource base
as limiting factor. A third group did not feel any restrictions. Among those
were two historians who worked at the same university. One of them is a
professor who has published numerous books that are read and cited beyond
Australia. He regularly acquires external research funding and is a member of
the Australian Academy of the Humanities. The other historian obtained his
doctoral degree six years ago, but has neither published his PhD thesis nor
any other substantial research result. He has never acquired external research
funding and was unable to describe research projects of his own even after
being repeatedly asked in different phases of the interview. Who is who?

Interviewer: And are there research topics that you are interested in but
can’t work on?
Historian 1: No. I’ve been very lucky.
Historian 2: ... that I’m interested in but I can’t work on? No. No, I think
there is still ample freedom to pursue that curiosity driven approach.
If I was to be seized with a particular idea, the resources are here, all
the resources are available [...], because I think with historical research,
the demands, the cost issues, are relatively modest. It’s really my own
time, photocopies, inter-library loans and travel. Put all those together
and that’s really all you need for historical research. So, for that reason,
I think most projects that I would be interested in pursuing, are always
124 Jochen Gläser and Grit Laudel

going to have fairly modest resource implications, and that’s definitely

no real reason why I couldn’t pursue them.

Both interviewees perceive a situation, which does not limit their research
financially, institutionally or in any other way. In both cases the aspiration
levels concerning the conditions of research correspond to the actual condi-
tions of research. However, the satisfaction of both interviewees is situated
at significantly different levels. One of the interviewees has sufficient time
and resources for the projects he wants to conduct (not the least because he
can use the external funding to reduce his teaching load), while the other
one doesn’t conduct projects and therefore also perceived a correspondence
between his interests and his conditions of work. The difference between
a situation with sufficient external funding and a situation in which even
the recurrent funding is only intermittent and based on internal grant pro-
posals is not visible in the two assessments of the respective situations.
Researchers from other disciplines also gave near-identical answers that
resulted from different aspiration levels. One of the two mathematicians
and one of the two geologists rarely publish, are hardly cited, don’t acquire
external grants and don’t have a clear research programme, while the other
two publish a lot, are cited above average, continuously get grants and pur-
sue long-term research programmes.

Interviewer: Now, are there any research topics that you are interested in
but can’t work on?
Mathematician 1: Not really, not really at all. I mean I’d like to have, I’m
hindered in the financial maths to some extent by not having someone
to talk to. Well, there is one other person here that does work in it but
he works on it from a different angle so, in some respects I would like to
do more in financial maths, but on the other hand I’d like to do more in
symmetries because that’s what I really like to do. But, no, I don’t think
that there’s an area that I can think of that I really want to do.
Mathematician 2: I don’t think so. No, I feel at the moment I have quite a
few research topics and I’m not looking for – I’m not actively looking for
more. I mean, often research projects just arise naturally. You just get into
them because you’re looking at some problem that leads into something
else. So I don’t sort of look to start a project from scratch. It always comes
from another source. No, I just follow what happens.
Geologist 1: I guess, at the moment it’s sufficient to keep me going at this
stage. I think I’ve got sufficient projects to keep me occupied at this stage.
So I’m not really thinking about other opportunities.
Geologist 2: No, not really, I mean everything I am interested in ... I can
maintain this balance between the applied and pure aspects enough to
keep me going now.
 Interviewing ‘Good’ and ‘Bad’ Experts 125

We found yet another example of identical descriptions of different

phenomena in the self-evaluations of research performance that was
sometimes given by the interviewees. The questions about the perform-
ance expectations of universities and evaluations of research performance
inspired two political scientists at the same university to reflect on how they
meet the expectations of their university. One of the two publishes irregu-
larly (with interruptions of up to four years). His only journal publications
are book reviews. The other publishes regularly, writes articles in refereed
journals, and is internationally visible.

Interviewer: So, my last question about the evaluations would be: What
are the consequences of these evaluations for you?
Political Scientist 1: In terms of, obviously, promotion. It’s whether you get
promoted or not. In terms of professional development – I mean, I am not
highly productive but I produce on a reasonably regular basis. I’ve never been
in a situation where someone said, you haven’t produced anything and
you’re going to. I can always say I’m in the middle of a book now and
it’s going to come out next year or the year after. So I’ve never been in
a position where anyone’s questioned me so I don’t know on what basis
they would.
Interviewer: Are there also specific special expectations in terms of the
quality of your research from the University or School?
Political Scientist 2: Well, no there is always some talk I guess again, [...] As
far as I know there is nothing on that type of functions. And it is clearly
that for most people in all universities never mind in [our university],
where it’s the number of publications that seems to matter more than
quality. And I don’t see any staff pressure counteracting that. I don’t pub-
lish very much. Even though I keep count of my rate, which is now slowing
down because of too much teaching. I don’t think I feel affected by any
pressures either way. I am aware of the pressures, but only a small num-
ber of us can do it.

The two passages in italics highlight the problem: without additional data
about the publication activities of respondents we would consider both
interviewees as not being particularly productive and as equally productive.
We would even consider the first (“I produce on a resonably regular basis”) as
the more productive and the second as less productive. But it is exactly the
other way around. Now let us imagine the many descriptions of this kind
that we cannot check independently of the interview ...

5.3.3 Different descriptions of identical phenomena

We encountered the other of our pure cases in description of evaluations,
that is of situations in which the university assessed the quality of its
126 Jochen Gläser and Grit Laudel

academics and based decisions on these assessments. Among these evalu-

ations, those that were conducted in the contexts of internal research
funding and promotions had the strongest influence on the situation of
researchers. The evaluations for promotions are based on a university-wide
standard procedure that is only slightly modified in order to accommodate
differences between disciplines. The following example includes two his-
torians from the same university who faced the same procedure and evalu-
ation criteria when it came to promotions. From the quotations it is clear
which interviewee is “good” and which is “bad.” It is far less clear that they
both describe the same procedure.

Interviewee: And is this the only occasion when your research is evalu-
ated? When you apply for internal grants?
Historian 1: No it’s evaluated if I was to submit an application for study
leave, which I have done, and to submit an application for promotion.
And I prepared an application under the previous Dean back in 2000 and
the then Dean made it quite clear even before I submitted the application
that she wasn’t going to support it because the first thing she said to me, I
had a meeting with her, was my research profile is not very good.
Interviewer: And since then you didn’t try again to ...?
Historian 1: I revised my application and again got as far as having a meet-
ing with the Dean last year but it was indicated that I would be ... And my
Head of School has indicated to me that I can be sure of being successful
in my application for promotion, this is only to Lecturer C, once I submit
the book manuscript. So in other words he’s telling me that I will only
be successful in promotion from a Lecturer B to a Lecturer C if I pub-
lish the book, which to my mind is ... We have people in the faculty and
elsewhere who’ve been promoted to Associate Professor and we’ve had
people who’ve been appointed as Professor who don’t even have a book
and yet I’ve been told as a condition of my application for promotion
being successful that I have to have submitted a book manuscript which
seems to me unfair.

Interviewer: And, how does the university evaluate your research?

Historian 2: Okay. Well, promotions would be the major mechanism. [...]
Yes, so on the individual level it’s promotion and career development
interviews.
Interviewer: What are the consequences of these evaluations for you?
Historian 2: Well, in my case, because I’m quite productive, it’s usually
good in the sense of getting promoted. But at present there haven’t been
negative consequences. But if, say, within the Faculty, people are not
being very productive, then they’re given mentoring, and I’ve done that
 Interviewing ‘Good’ and ‘Bad’ Experts 127

as well. You engage with people and you advise them on what they could
do to get together a great application or what kinds of publication strat-
egies they need to have.

Both descriptions refer to the same promotion procedure. Academics write

an application in which they describe their achievements in teaching,
research, and academic administration. They discuss this application with
their supervisors (first with the head of school and thereafter with the dean
of their faculty), who advise them on the content of the application and
on their chances of success. When the application is moving up the hier-
archy, each supervisor adds their comments. The final decision is made by
a university committee. If the supervisors believe the application would be
unsuccessful they advise academics to withdraw it. Such a withdrawal is in
the interest of both the applicant and their supervisor. Academics whose
application is rejected must wait several years before they can apply again,
while postponing an application that has little chance of success enables
another application with better chances in the following year. A weak appli-
cation that goes ahead also creates a dilemma for the supervisors. If they
support it, they lose their credibility at the higher levels of the hierarchy. If
they don’t support it they lose their credibility with the applicant.
The application procedure is strictly regulated. The rules are made pub-
licly available and are strictly followed. Both interviewees we quoted above
wrote their applications and had consultations in which they received
advice. However, their perceptions of the process are quite different. One
interviewee mentions his promotions only casually and gives the impres-
sion that they were not problematic. For him, the stronger and more present
experience is that of his involvement as a mentor, a role, which has been
introduced to support weak applicants. The other interviewee applied with
achievements that were considered as insufficient by two deans and the
head of school. The supervisors advised against an application. The appli-
cant regarded this as an unfair interpretation of performance criteria, a
rather unlikely interpretation given the strict rules and the consistent judg-
ment by the two deans and the head of school. However, it is interesting to
note that something that is interpreted by the “bad” applicant as an unfair
intervention that prevented the promotion is of no personal concern to the
“good” applicant, who hardly even mentions it. The different performance
levels of the two interviewees have a massive impact on their perceptions
and descriptions of what is essentially the same phenomenon, namely the
promotion procedure. This would not be a problem if only their perceptions
would matter for the investigation. However, we also need to know the pro-
motion procedure. While in this particular case the “true” procedure could
be established by triangulation from many different sources, it is again easy
to imagine situations where this would be impossible, and quality-dependent
descriptions were the only information about a phenomenon.
128 Jochen Gläser and Grit Laudel

5.3.4 Implications for the investigation

In our comments on the quotations from interviews we have already hinted
at their different informational yield for our project. Now we would like
to systematize these consequences of the quality problem. The differences
in the performance levels of our respondents shaped their communication
about two aspects of their work that were central to our investigation. The
first aspect is a respondent’s content of work. When we ask for an explan-
ation of the respondent’s research problem, reasons for the selection of spe-
cific research objects and methods, or other information about the content
of research, “bad” researchers provide information that is different from
that provided by “good” researchers. This information about the content of
work is crucial for our investigation because the content of work shapes the
resource demand. “Good” researchers formulate more challenging research
problems, and their assessment of strategic aspects of their research such as
the solvability of problems or the applicability of methods to their problems
differs from assessments of the same aspects by “bad” researchers. These
strategic choices and assessments produce the specific resource requirements
of the respondent’s research, which constituted one of the central variables
of our investigation. This is why the results of our study partly depend on
what the biochemists tell us about their biochemistry, geologists about their
geology and political scientists about their political science.3
In addition to the specific information about the respondents’ own research
the interview also yields general information about the characteristic epistemic
practices of their research field. Some characteristics of fields (for example
the typical duration of research processes, specific uncertainties or hetero-
geneity of the used knowledge claims) are important intervening variables in
our investigation because they explain the varying implications of uniform
resource allocation mechanisms in different research fields. This is why we
must extract such field-specific factors from the statements of all respondents
belonging to the same field. A description of field-specific practices must be
synthesized from statements by researchers of varying performance levels.
This “learning from the field,” that is the use of experts as a source of
information about characteristics of the content of work, occurs in many
science studies projects and probably in many other studies that are based
on expert interviews. It is always a means to an end because the knowledge
about work processes of experts is needed to answer sociological questions.
This is why a second aspect of our respondents’ work – the conditions of work
under which the research was conducted – was more important to us than
the content of research, and why most of our examples refer to these con-
ditions. The performance levels of respondents strongly affected the data
about conditions of work. Our examples illustrated that

● the statement “my research is not constrained by limitations to my access

to resources” may reflect two completely different situations, namely one
 Interviewing ‘Good’ and ‘Bad’ Experts 129

of a researcher whose aspiration level is so low that it can be met by any

resource base, or one of a researcher who is so “good” that they receive all
the resources they ask for; and that
● the perception of organizational routines largely depends on the perform-
ance levels of respondents, that is that largely identical routines may be
differently perceived by “good” and “bad” researchers.

For these reasons we could not simply extract general, university-spe-

cific or faculty-specific conditions or routines but also had to include
the performance levels of interviewees in our interpretation. We could
not avoid this problem by relying on documents describing the formal
bureaucratic routines because the actual use of these routines in every-
day university life (which mediates the influence of institutions on the
content of research) inevitably deviates from the formal prescriptions in
documents.
If we tentatively generalize our experiences we find that quality differences
between experts interfere with two essential tasks of expert interviews:

(1) If we need to learn something about the content of an interviewee’s

work, interviews with “good” experts provide information about this
work that is different from that provided by “bad” experts.
(2) If we need to learn something about the interviewee’s conditions of work
we obtain quality-affected information for two reasons. Firstly “good”
experts have higher aspiration levels and higher standards against
which conditions of work are assessed. Secondly, interventions by the
environment are dealt with in different ways by respondents depending
on their performance levels, and are thus different for “good” and “bad”
experts.

Investigations that rely on at least one of these kinds of information have to

face the quality problem.

5.4 Quality differences as a methodological

problem of expert interviews and their analysis

5.4.1 How do we assess experts?

Whenever the performance levels of the expert affects their responses, we
must know these performancel levels and take them into account when
interpreting statements. However, a quality assessment of experts seems to
be an impossible task. After all, we interview experts for the very reason
of accessing their special knowledge, that is knowledge we don’t possess.
How could we possibly assess how much of this knowledge they have and
how well they are able to use it? Apparently there are only few exceptions
to this dilemma. Competitive athletes, for example, are subject to a public
130 Jochen Gläser and Grit Laudel

performance assessment that is comprehensible outside their group because

they compete against each other.
On first glance our own task – to assess the quality of researchers – puts
us in a quite comfortable position, too. The demands of science policy and
management have pushed the development of evaluation methods, which by
now have widely diffused throughout public science systems. Unfortunately,
this does not make our task much easier. Science studies agree that ultim-
ately only researchers’ peers are able to judge the quality of their work, and
that even this peer review may get it wrong. “Objective” indicators that can
be applied by outside observers because they abstract from the content of
research only offer partial insights into the quality of research, and evaluate
past rather than current performance. Therefore, they can only be used as
additional information in a peer review. Moreover, we have already pointed
out that quantitative indicators cannot be regarded as reliable at the individ-
ual level that we are interested in.
The limited opportunities of performance measurement are highlighted
when we ask which aspect of quality we actually intend to measure. Is it the
current performance of the investigated researchers or their ability to perform?
In our investigation we came across two groups of interviewees for which
this difference affected assessments. The careers of some scientists were not
yet long enough to enable conclusions about extraordinary research abilities
although there were some indications of that. The opposite case occurred,
too. Researchers who had made outstanding contributions in the past had
ceased to do so at the time of the interview because they didn’t receive exter-
nal funding anymore, had moved away from the mainstream of their field,
or were close to retirement and didn’t want to start new projects. In both
cases it is likely that the scientific abilities of the interviewee exceed the
level suggested by the current performance. For our purposes, a categoriza-
tion according to abilities is more important than one according to the cur-
rent research performance. However, the opportunities to assess abilities are
even more limited when current performance is likely to offer a distorted
picture.
Given the numerous problems of performance measurement in science,
which we have dicscussed in our own publications, we had to be extremely
careful in our categorization of researchers. We used as many indicators
as possible, supplemented them with “independent” information from the
interviews, and compared the researchers of one discipline to each other.
Prior to the interview we downloaded information about research inter-
ests and projects, indicators of peer esteem such as prizes for research or
memberships of learned academies, publications, citations and external
grant acquisition from internet pages, the Web of Science, and other pub-
lication databases. All these indicators are discipline-specific (for example,
citation analyses are extremely problematic in the social sciences and the
humanities) and can only be interpreted by taking the context into account.
 Interviewing ‘Good’ and ‘Bad’ Experts 131

In fields with a high degree of collaboration it is not always easy to ascribe

achievements to a specific researcher. Researchers may profit from being a
member of a strong group and participate in publications and grant acquisi-
tions without making outstanding contributions themselves. We addressed
this problem by checking indicators for all co-authors of highly cited publi-
cations and of co-applicants for grants.
This information was used for preparing the interviews. The preparation
already included a preliminary assessment of performance levels and the
selection of an appropriate strategy for handling the quality problem in
the interview. We could not rule out that research was a sensitive topic for
those interviewees whose performance levels were considerably lower than
those of their colleagues. This is why we applied a specific strategy when
interviewing academics with apparently sparse research records. We didn’t
start with questions about research, which would have presumed that the
interviewee conducts research as everybody else, but with questions about
teaching. In this initial conversation, we established the extent to which the
interviewee actually conducts research, and steered the interview towards
research activities depending on that information.
Our interviews contributed to the assessment of quality because our
question about the research biography, conditions of research and future
research plans touched a trait of a researcher that is closely linked to quality.
We obtained clear indications concerning the aspiration level of the inter-
viewed researchers, that is their standards concerning their own work, their
collaborators, other colleagues, and their conditions of research. Thus, the
descriptions of the research and conditions of work provided in the inter-
view are important indicators of quality themselves. Another indicator we
could use is the consistency of the research topics. Do interviewees pursue a
consistent research programme, or are they merely responding to external
stimuli such as demands by collaborator or supervisors? Does the inter-
viewee independently conduct research at all?
After the interview we refined the information about our indicators and
collected additional information from the internet and publication data-
bases. Using the indicators and information from the interviews, we com-
pared the researchers of each discipline with each other and constructed
three groups: a high-level group (16 researchers across all disciplines), a
medium-level group (79) and a group of academics who hardly conduct any
research (23). In order to highlight the borderline cases whose classifica-
tion was most difficult, we specifically identified those academics in the
medium-level group who were close to either the top or the inactive groups.
We conducted the categorization together (and largely achieved agreement).
The advantage of this collaborative categorization exercise is the necessity to
verbalize impressions from the interview (which only one of us had prepared
and conducted). These impressions could thus be discussed in the light of
the other information that was gathered independently of the interview.
132 Jochen Gläser and Grit Laudel

5.4.2 How do we include the “quality” of our experts

into the data analysis?
After we got an idea which of our interviewees are “good,” “average” or
“inactive” researchers, we coded this categorization in the file names of the
interview transcripts, which means that this information appears in the
source tag of each interview statement that was used in the analysis.
Always knowing how “good” a source/expert is helps but doesn’t solve the
problem. Being well aware of the problem that the interpretation of the data
can never be fully described, we will nevertheless try to demonstrate how we
used the information about quality in our analysis. We already mentioned
the two possible ways of using the quality information in our short over-
view of the treatment of research quality by science studies. Firstly, know-
ledge about the quality of the source must be used for the interpretation of
the data. Secondly, in our investigation we also need to know whether the
adaptive behaviour of “good” researchers differs from that of “bad” ones.
Thus, we must also include the performance level of a researcher as a subject
of the investigation.
The way in which we take into account the performance levels of respond-
ents in the interpretation of the data can only partly be made explicit. We
are sure that we read interviews and interpret data “differently” because we
have an opinion about the performance levels of the sources. This is also a
danger for the interpretation if a prejudice against a group occurs. We coun-
teracted these influences by conducting the first step of the qualitative con-
tent analysis – the extraction of raw data – prior to determining and coding
performance levels. In the subsequent analysis, we used a “check list” and
determined whether

● information is shaped by the ‘aspiration level’ of the interviewee (in

our case: for example if above or below average expectations concern-
ing research performance and conditions of work may have influenced a
statement);
● a specific content of work that is due to specific performance levels might
have shaped the information (in our case: for example if above or below
average research performance manifests itself for example in the num-
ber of parallel projects a researcher works on or in the continuity of the
respondent’s publication record),
● differences between performance levels are accompanied by different
self-images and valuations (in our case: for example whether different
research performance is accompanied by varying assessments of tasks in
teaching); and
● whether an interviewee’s performance level affects reported causalities (in
our case: for example whether unfavourable research conditions are used
in a specific way to explain lower research performance).
 Interviewing ‘Good’ and ‘Bad’ Experts 133

These questions are particularly important when we find contradictions in

the data or when information that is central to the investigation only stems
from few sources. An identification of possible influences of performance
levels does of course not mean that we discard the data as “unreliable.” We
are just in a better position to decide the role we assign a certain piece of
information in the construction of our explanation. The extent to which
critical aspects of our typologies, generalizations and explanations could be
influenced by performance levels of the sources becomes visible.
Information about performance levels of respondents was not only used
for the interpretation of data but also in a second, more direct way. In our
investigation, the quality of the researchers is also an intervening variable.
If researchers adapt the content of their research to institutional conditions
then we must find out whether the adaptive behaviour of “good” researchers
is different from that of “bad” ones. We achieve this by constructing categor-
ies of performance levels and separately investigating the adaptative behav-
iour for each category. An example of that strategy is Grit Laudel’s comparative
study of conditions for external fund acquisition and their impact on the
content of research (Laudel, 2006a, 2006b). Studies of hindering conditions
of research frequently face the objection that the complaints of researchers
are just the “rationalizations of the loosers,” that is that “bad” researchers use
unfavourable research conditions as an excuse for performing badly rather
than admitting their insufficient abilities. The comparative study of external
fund acquisition of experimental physicists in Germany and Australia took
this argument into account by distinguishing two performance levels (“top”
and “other”) and two categories of the scientists’ resource bases (“rich” and
“other”). A cross-comparison of the four groups showed that all four fields
were populated (Table 5.2). There are the expected concentrations in the lower
right cells, which indicate that external fund acquisition indeed is related to
the performance level of a researcher. The existence of “top other” and “other
rich” scientists shows, however, that quality is neither necessary nor suffi-
cient for a good external fund acquisition. This already puts the “rationaliza-
tion argument” on doubt. Additionally, an analysis of adaptation strategies

Table 5.2 Differentiation of experimental physicists in Australia and Germany

according to their research performance and acquisition of external funds

Amount of external funding

Germany Australia

“Rich” “Other” “Rich” “Other”

“Top” scientists 8 7 3 4
“Other” scientists 7 23 4 10
Soure: Based on Laudel, 2006b, p. 382.
134 Jochen Gläser and Grit Laudel

showed their occurence across all performance levels as a reaction to insuffi-

cient resource bases. Thus, the “rationalization argument” could be refuted.
The second example is taken from our current investigation that provided
the material for this article. During a preliminary analysis of the interviews
we discovered that the adaptative behaviour of “good” and “bad” research-
ers to the conditions of research funding differed. There is a small group of
“top” researchers who

● work on a comparatively broad range of research topics and thus could

adapt to political demands by selecting politically desired (which means
in Australia application-oriented) topics but
● don’t have to surrender to this pressure because they receive the external
funding for their favourite projects even if these projects do not corres-
pond to the political orientations of the research landscape (their high
performance being the substitute for “thematic correctness”).

In the case of these researchers two things come together: a high thematic
flexibility and a low pressure to adapt. At the other end of the spectrum are
researchers who experience a high adaptation pressure because their current
research projects are not successful. However, these researchers are not able
to change their research in the required way because their low performance
levels also mean that they cannot apply their accumulated knowledge to a
new topic.
This observation made us formulate a hypothesis about the relation-
ship between performance levels and adaptative behaviour (Figure 5.2).

Pressure to adapt Ability to adapt

Observable adaptation

‘Quality’ of academics

Figure 5.2 Relationship between performance levels and the adaptation of the
research content
Source: Gläser and Laudel 2007b.
 Interviewing ‘Good’ and ‘Bad’ Experts 135

Adaptations to institutionalized expectations concerning the content of

research occur neither at the top level (because these academics perceive lit-
tle pressure), nor at the weakest performance level (because these academics
are not able to make the necessary changes of direction even under strong
pressure). The institutionalized pressure to adapt has its main impact at the
medium level of academics who are “not good enough” to protect them-
selves from pressure through “excellence” but good enough to respond to
this pressure with changes to the content of their research.

5.5 Concluding remarks

To what extent can our considerations about the role of the quality of
experts be generalized? We believe that the problem of “quality-dependent”
data in expert interviews affects many investigations because the defining
property of an expert – specific knowledge and abilities – varies between
experts. We therefore regard it as necessary to consider possible influences
of quality differences between experts at the beginning of an investigation.
If data may be affected, appropriate strategies for the selection of experts,
the conduct of interviews and the data analysis must be developed.
Depending on the aim of a study, the performance levels of experts can
also play a role beyond the “area of overlap” identified in the introduc-
tion. We might for example be interested to know whether biographical
self-represenations of “good” physicians are different from those of “bad”
ones. In this case, the quality is explicitly part of the subject matter of the
investigation and must be dealt with. Quality differences occur as a “hid-
den” methodological problem only in studies where “experts in the field” are
sources of information about social situations and processes they observed.
Although the techniques of collecting information vary with the research
problems and the experts involved, the steps we outlined can presumably
be generalized. The main point is to decide early whether the quality of
experts could be a problem and whether the strategy of interviewing must
be adapted. This includes the decision about using the interviews for collect-
ing data on indicators of quality such as the respondent’s aspiration level
concerning their own work and conditions of work. With regard to the data
analysis a decision must be made whether the quality of experts could have
influenced the information they provided. If this is the case, the strategy of
data analysis must be adapted accordingly. Our questions about the aspir-
ation levels, work load, self-image, assessments, and rationalizations seem to
be applicable beyond interviews with researchers.
Researchers are an interesting “show-and-tell piece” because quality dif-
ferences between them are a generally recognized and in part publicly
negotiated fact, and because quality-relevant information is publicly avail-
able. In other investigations that are based on expert interviews it may be
far more difficult or even impossible to get “objective” information (that is
136 Jochen Gläser and Grit Laudel

information that is independent from the interviewer’s judgement) about

the “quality” of experts. As is the case with all methodological problems in
social research, the first necessary step is to emphasize the problem and to
accept its relevance for the investigation. It was our aim to initiate this step.

Notes
1. See for those aspects of the interview situation for example Dean and Whyte
(1958), Becker and Geer (1970), Bernard and others (1984), Richards (1996),
Bernard (2002, pp. 187–91), as well as Gläser and Laudel (2009, pp. 178–82).
2. A description of our project and the empirical methods can be found in Gläser
and Laudel (2007b).
3. Unfortunately, it is nearly impossible to find out how information on the content
of research is influenced by the performance levels of the interviewees because
each research process is unique. Differently from bakers, physicians or managers
whose work is repetitive to a large extent, researchers aim at producing new know-
ledge, which makes each research process unique. Consequently, there is no solid
base for a comparison of research processes. Even in the case of collaborative
research processes whose participants could all be interviewed, the comparison
is hindered by the unique disciplinary perspective of each collaborator. PhD stu-
dents are the only exception to this “problem of uniqueness” because their super-
visor is able to assess the research process as well as the researcher.

Further readings
Bernard, H. R., Killworth, P., Kronenfeld, D. and Sailer, L. (1984) “The Problem
of Informant Accuracy: The Validity of Retrospective Data” in Annual Review of
Anthropology 13, 495–517.
Dean, J. P. and Foote Whyte, W. (1958) “How Do You Know If the Informant is Telling
the Truth” in Human Organization 17, 34–9.
Laudel, G. (2006) “The ‘quality myth’: Promoting and hindering conditions for
acquiring research funds” in Higher Education 52, 375–403.

References
Becker, H. S. and Geer, B. (1970) “Participant Observation and Interviewing: A
Comparison” in Filstead, W. J. (ed.) Qualitative Methodology: Firsthand Involvement
with the Social World (Chicago: Markham Publishing), pp. 133–42.
Bernard, H. R. (2002) Research Methods in Anthropology: Qualitative and Quantitative
Approaches (Walnut Creek: Altamira Press).
Bernard, H. R., Killworth, P., Kronenfeld, D. and Sailer, L. (1984) “The Problem
of Informant Accuracy: The Validity of Retrospective Data” in Annual Review of
Anthropology, Vol. 13, 495–517.
Bogner, A. and Menz, W. (2005) “Expertenwissen und Forschungspraxis: Die mod-
ernisierungstheoretische und die methodische Debatte um die Experten. Zur
Einführung in ein unübersichtliches Problemfeld” in Bogner, A., Littig, B. and
Menz, W. (eds) Das Experteninterview – Theorie, Methode, Anwendung, 2nd edn
(Wiesbaden: Verlag für Sozialwissenschaften), pp. 7–30.
Cole, J. R. and Cole, S. (1972) The Ortega Hypothesis, Science 178, pp. 368–75.
 Interviewing ‘Good’ and ‘Bad’ Experts 137

Cole, J. R. and Cole, S. (1967) “Scientific Output and Recognition, a Study in the
Operation of the Reward System in Science” in American Sociological Review 32,
377–90.
Cole, S. (1970) “Professional Standing and the Reception of Scientific Discoveries” in
American Journal of Sociology 76, 286–306.
D. Crane (1972) Invisible Colleges: Diffusion of Knowledge in Scientific Communities
(Chicago: University of Chicago Press).
Dean, J. P. and Whyte, W. F. (1958) “How Do You Know If the Informant is Telling the
Truth” in Human Organization 17, 34–9.
Gläser, J. and Laudel, G. (2009) Experteninterviews und qualitative Inhaltsanalyse als
Instrumente rekonstruierender Untersuchungen, 3rd edn (Wiesbaden: Verlag für
Sozialwissenschaften).
Gläser, J. and Laudel, G. (2007a) “The social construction of bibliometric methods” in
Barker, K., Gläser, J. and Whitley, R. (eds.) The Changing Governance of the Sciences:
The Advent of Research Evaluation Systems (Dordrecht: Springer), pp. 101–23.
Gläser, J. and Laudel, G. (2007b) “Evaluation without Evaluators: The impact of fund-
ing formulae on Australian University Research” in Gläser, J. and Whitley, R. (eds.)
The Changing Governance of the Sciences: The Advent of Research Evaluation Systems
(Dordrecht: Springer), pp. 127–51.
Hagstrom, W. O. (1965) The Scientific Community (Carbondale, IL: Southern Illinois
University Press).
Hopf, C. (1993 [1979]) “Soziologie und qualitative Sozialforschung” in Hopf, C and
Weingarten, E. (eds.) Qualitative Sozialforschung (Stuttgart: Klett-Kotta).
Jackson, D. N. and Rushton, J. P. (eds.) (1987) Scientific Excellence. Origins and
Assessment. (Newbury Park: SAGE).
Knorr-Cetina, K. (1984) Die Fabrikation von Erkenntnis. Zur Anthropologie der
Naturwissenschaft (Frankfurt am Main: Suhrkamp).
Latour, B. and Woolgar, S. (1986 [1979]) Laboratory Life: The Construction of Scientific
Facts (Princeton: University Press).
Laudel, G. (2005) “Is external funding a valid indicator for research performance?”
in Research Evaluation 14 , 27–34.
Laudel, G. (2006a) “The art of getting funded: How Scientists adapt to their funding
conditions” in Science and Public Policy 33, 489–504.
Laudel, G. (2006b) “The ‘quality myth’: Promoting and hindering conditions for
acquiring research funds” in Higher Education 52, 375–403.
Lynch, M. (1985) Art and Artifact in Laboratory Science: A Study of Shop Work and Shop
Talk in a Research Laboratory (London: Routledge and Kegan Paul).
Pelz, D. C. and Andrews, F. M. (1966) Scientists in organizations. Productive Climates for
Research and Development (New York: Wiley).
Richards, D. (1996) “Doing Politics: Elite Interviewing: Approaches and Pitfalls” in
Politics 16, 199–204.
Zelditch Jr., M. (1961) “Some Methodological Problems of Field Studies” in American
Journal of Sociology 67, 566–76.
Zuckerman, H. and Cole, J. R. (1994) “Research Strategies in Science: A Preliminary
Inquiry” in Creativity Research Journal 7, 391–405.
Zuckerman, H. (1977) Scientific Elite: Nobel Laureates in the United States (New York:
Free Press).
Zuckerman, H. (1972) “Interviewing an Ultra-Elite” in Public Opinion Quarterly 36,
159–75.
6
Interviewing Experts in Political
Science: A Reflection on Gender
and Policy Effects Based on
Secondary Analysis
Gabriele Abels and Maria Behrens

6.1 Introduction

Unquestionably, German political science draws from a plurality of

methods. In comparison to Anglo-Saxon political science, it is for the most
part coined less by the “precedence-position of quantitative methodologies
and methods hailing from the natural sciences” (Dackweiler, 2004, p. 53;
our translation) – simultaneously, it is also characterized by a lack of meth-
odological reflexion. In his 1991 appraisal of qualitative processes Patzelt
points out that the practice of empirical political research was once grasped
by the “popularity wave known as qualitative analysis” (Patzelt, 1991, p. 53;
our translation) and considered herein a faulty approach. Thus, collective
statements “ostensible inclinations in regard to categorizing reflection about
research methods as irrelevant or to attribute these to Sociology” (ibid.)
should be voiced. Interestingly, gender studies in (German) political science
are no exemption. Looking at introductory-, text- and handbooks one has
to realize how rarely chapters on methods may be found (Dackweiler, 2004,
Ebbecke-Nohlen/Nohlen, 1994).
Since the mid-1990s reflection on methods has indubitably increased
and been documented (for examples see von Alemann 1995, Kriz and
others 1994, Behrens 2003, Behnke and others 2006a, Behnke and others
2006b, Behrens and Henning 2008). It is notably anchored in professional
associations’ publications, too.1 In any case, a debate on basic methods and
on the development of sufficient processes that would be fully compatible
with the discipline’s intricacies is yet to occur.
The debate should, however, not focus on creating or promoting a stand-
ard set of disciplinary methods, as it genuinely distinguishes itself from the
neighbouring social science disciplines. Simultaneously, German political

138
 Interviewing Experts in Political Science 139

science lacks certain methodical approaches and processes for data collec-
tion and analysis – particularly “de-personalized” approaches to research
subjects – preferably strategies and processes requiring methodological and
methodical reflection. The expert-interview falls into this category, as its
purpose is to explore political action (Meuser and Nagel 1991) as well as
to query and evaluate the mundane side of politics and its every day prac-
tices (Patzelt, 1994, p. 398). Expert-interviews are vitally important to policy
research, in this case especially in implementation studies (Voelzkow, 1995,
p. 55) and evaluative analysis.

6.2 What is an expert-interview?

In the methodological classic published in 1970, Dexter (2006) cites an “elite

specialized in interviewing,” even if his understanding of elite includes a
variation of groups. Meuser and Nagel define expert-interviews as a specific
type of qualitative interview, differing from other questioning practices in
the fashion conversations are conducted, as well as in the evaluation of
statements, but also concerning cognitive interests (“Erkenntnisinteresse”)
and respondent groups (“Befragtengruppen”) (Meuser and Nagel, 1997,
p. 482, cf. Helfferich, 2004).
This partially connects to Dexter’s stipulation; he mentions the follow-
ing characteristics: (1) the respondent defines the interview’s context, (2) he
or she structures the framework of relevance/meaning and (3) he or she
decides within the interview’s course what is and is not meaningful and
relevant. However, these criteria may apply for narrative and biographical
interviews, too.
The term “informational conversations” (“Informationsgespräche”) is used
by von Alemann and Forndran (1990, p. 169). More precisely, it defines the
scientists’ cognitive interests in the process: “gaining”2 information, leaving
the addressee unnamed. The term “expert-interviews,” however, puts the
respondent into attention’s focus and accentuates that the interviewer faces
people who are endowed with a specific form of professional wisdom, which
differs from the knowledge of a layperson. With this information at hand,
the term “expert-interviews” seems more precise than any other alternative
definitions or words known to us.
Who is considered to be of expert status is always a social and method-
ical construction (Deeke, 1995, p. 9). In particular this question can only
be answered in relation to the subject researched and cognitive interests
(Meuser and Nagel, 1997). Overall experts are “agents bearing specific func-
tions within an organisational or institutional context,” who “(re)present
solutions to problems and decision-making-processes” (Meuser and Nagel,
2005, p. 74, our translation). Experts do not – as in the investigation by
Meuser and Nagel – belong to an organization’s management elite,3 but very
often hail from the mid- and lower ranks of organizations. Commonly, they
140 Gabriele Abels and Maria Behrens

are highly educated individuals, who are very conscious of their status and
accustomed to present themselves favourably, to handle inquisitive situ-
ations and to elaborate on complex contexts.
While in elite-interviews the respondent’s personal inclinations might be
of interest (cf. van Schendelen 1984, Aberbach and others 1975, Semmel
1975), this fact does not apply to expert-interviews. The expert is not inter-
viewed as an individual; the interview context “is an organizational or insti-
tutional, which is not identical to the individuals acting therein; instead
these individuals are merely a “factor” within this framework” (Meuser and
Nagel, 2005, pp. 72f., our translation).
In an expert-interview in political science, respondents are present three-
fold: as individuals, as representatives and as strategists: an individual inter-
viewed also corresponds to an institution or organization, which he or she
represents, and finally he or she is present as a collective, strategic actor (for
example a political party, an association, a department) operating in the
political sphere. From these findings we derive specific problems for the con-
versational situation, requiring a systematic reflection on methodological
criteria such as validity, reliability and generalization, as the information
shared with a researcher could reflect the private opinion of the interviewee,
depict the official position of the organization or even be spiked with an
instrumental purpose, by deliberately giving out information to coin public
discourse, without revealing the actor’s strategic interests. At this point the
scientist is in danger of being instrumentalized by a collective actor for their
official organizational politics (Behrens, 2003, p. 229).
An interview is a “relational space” (Tietel, 2000; our translation); accord-
ing to Heinzel (1997, p. 100; our translation) “it is fundamental to each dia-
logical interview-situation that expectations of relations are effectual and
patterns of interaction are staged.” Hence, the expert is also always pre-
sent as an individual in the conversation. Moreover: when a respondent’s
subject-status is negated, this is perceived as a narcissistic offence. This could
hurt the temporary relationship of trust required in the fragile situation
that is the interview and hence put the interview’s success in jeopardy.4
Dexter already stated that the interview itself is an effectual factor “making
the elite feel more special and making them more willing to share informa-
tion they would not give out under different circumstances”; because the
expert’s identity is obligatorily bound in his or her belief in their status as
expert (Ware and Sánchez-Jankowski, 2006, p. 5). This thinking is therefore
part of a collective identity and demands confirmation within the inter-
view’s context.
Similarly, the researcher is not a neuter, but all interview participants are
also always present not only as subjects, and not only as representatives of
their respective organization or “science.” Interests, trust, power, control and
hierarchy also influence this specific form of social interaction. This implies
furthermore – from a gender studies perspective – that the participants in
 Interviewing Experts in Political Science 141

the interview situation will always act as gendered subjects. This in turn
calls attention to the question concerning the interviewer’s behaviour in
the research situation.
In the interpretative paradigm, communication and interaction are con-
sidered a main constituent of the research process. The key principle is that
“the scientist usually only receives access to relevant and structured data
when he [sic!] enters a communicative relation with the researched sub-
ject, and consequently instates the conversational rule system of the subject
researched” (Hoffmann-Riem, 1980, p. 346f.; our translation). Accordingly,
it is recommended to the interviewer to maintain a gesture of interested
aloofness, even within the framework of an open and semi-structured
expert-interview. The interviewer is to be an “empathic” and “stimulating-
passive” individual who merely encourages the respondent to speak, so
that the interview protocol may be read and interpreted as a monologue
(cf. Lamnek, 1989, pp. 67, 69, 179). Along these lines the interview situ-
ation should “strive to adhere to the communicative rules of and be simi-
lar to mundane action” (Lamnek, 1988, p. 24; our translation). For expert
interviews in political science we can assume that the more professional
respondent represents a collective, strategic actor, subjective and mundane
codifications become less important.
The problems consequently resulting for political scientific research prac-
tice are obvious. On the one hand, the interview’s context requires open-
ness, exploration and flexibility. On the other hand, the respondent may
retort with counter-questions, strategic comments, or ask the interviewer’s
take on a specific problem. The scientist must then react. Concerning her
or his interest to gain information, an attitude including negation to the
respondents’ reactions will most likely be unfavourable to the researcher.
Schmid (1995, p. 285) recommends – while making political statements for
example – to show a “benevolent interest”; however, a “detective’s flair” – the
use of suggestive questions as inescapable while dealing with critical topics.
Moreover “comments, gestures and actions” by the interviewer should be
controlled (ibid., p. 311; our translation).
It is clear that the scientist must act as neutral as possible, but does or can-
not always act neutral in an interview, while he or she must attempt “luring”
information out of the respondent. Many socially constructed and situational
factors are out of the researcher’s reach; these are communicated verbally
and nonverbally and could at best be partially controlled for, and partially
applied deliberately in one’s favour. Key factors are the following: gender, age,
professional status/title, experience and background, idiosyncrasies/attitudes
and organizational affiliation. They operate in the specific interview situ-
ation not independently, but in sync with each other, effectively enforcing
or lessening each other. This makes it simply impossible to single out indi-
vidual effects and to weigh their precise influence (cf. Bryant and Hoon,
2006). We do not consider these factors as “distortions” and “disturbances,”
142 Gabriele Abels and Maria Behrens

which could and should be minimized. On the contrary: we claim that these
factors are always inescapably present in research as a social interaction. For
us, the gender factor is a structural social category with a prominent role in
the interview situation. Alluding to the “doing-gender” approach it should
be noted, “that gender differences in the interview are created by interaction
and are present throughout the interview” (Littig 2005).
Considering this, let it be pointed out that the respective literature widely
ignores the gender issue. Meuser and Nagel (2005, p. 79; our translation)
briefly state that with “certainty ... gender differences between the partici-
pants of an interview [impact] the situation’s definition.” Warren (1988,
p. 44) remarks it is almost a truism in interview science that women receive
more feedback than men most of the time, because they are perceived as less
threatening and have higher communicative skills. Van Schendelen (1984,
p. 307) mentions that access to parliamentarians in different structures was
easier for women, and therefore gender was part of recruitment characteris-
tics when casting an interview team – but without drawing any further con-
clusions for data collection and data analysis. Padfield and Procter (1996) in
contrast plead that gender effects should be minimized to the best of one’s
capability. These general statements on gender effects are then to be limited
so that – taking aside the danger of reification of gender stereotypes – the
subject researched and also other changes in social gender relations are to
be respected (for example, the slow rise of women into leading positions,
paving the way into what are traditionally considered male domains), espe-
cially then when the gender hierarchy shifts.
For gender studies, as remarked above, it is certainly not a new insight
that gender “accompanies” the whole research context (cf. the discussion in
Sarantakos 2004). First methodological reflections on gender-specific effects
already exist regarding narrative interviews (McKee and O’Brian, 1983,
Padfield and Procter, 1996, Williams and Heikes, 1993, Littig 2005). However,
in most of these cases the researched topic is deeply gendered (abortion, sin-
gle fathers, nursing); therefore, gender relations, the topic and biographical
interview technique have most likely influenced the interview’s setting.
Gurney (1985) in contrast, reflects her research experience in a male domain
as greatly coined by problems of access to the field due to sexist behaviour.
Until now these types of reflections are more of a “side product”; there is a
remarkable lack “of fundamental (meta-)studies on gender category’s mean-
ing for an expert-interview” (Littig 2005, p. 203, our translation).

6.3 Effects in conversational interaction

and their methodological tracing

In the following sections we will draw on our experience with semi-structured

expert-interviews that we collected while being involved in two empiric
research projects in the field of biotechnology policy in the mid-1990s.5 For
 Interviewing Experts in Political Science 143

the interview setting it was relevant for a gigantic difference between the
scientist and the respondent to exist. Next to gender, these differences con-
cerned age and qualification, as well as the formal occupational position.
Furthermore, the interviews were also influenced by the fact that biotechnol-
ogy policy is a greatly controversial political and technical topic in society.6
A central result of our methodical reflection on the basis of conversational
analytic category-building was the identification of specific interaction-
effects or patterns within the interview situation, which were in particular
embossed by gender, which we take as a fundamental social category. Based
on Vogel’s suggestions (1995) we identified a general paternalism, catharsis,
iceberg and reactionary feedback effect, and added to these the profile- effect.
The results, especially those contained in our gender-theorem-interpretation,
are traced by means of a secondary analysis.
Secondary analysis of qualitative data, defined as a process during which
primary data is made available to another research group, is a technique
barely used in political science until today (see in detail Dale and others,
1988, Corti 2000, Corti and Thompson, 2004, van den Berg, 2005). The
purpose is to verify validity and reliability of data. For qualitative expert-
interviews, interview protocols (that is usually a transcription of a conver-
sation) and memos are archived. Interview memos are recordings made by
the original research-team, including personal impressions, as well as infor-
mation on the non-verbal interview situation (for example where a talk took
place, seating plan, gestures made by the respondent).
In a synchronic secondary analysis, primary data is exchanged between teams,
in order to re-examine and compare interpretations with one’s own results.
Supported by primary data from previous projects diachronic secondary analysis
allows reconstructing the subjective characteristics (class or ethnic background,
age, gender) of the interview participants and their relevance for the interpret-
ation of data. Based on interview memos taken throughout a research project
on the development of young people between 1962 and 1964, Goodwin and
O’Connor (2006, p. 390) observed: “Alongside the lens of class, there was also
some evidence that gender and gender-based prejudices influenced the inter-
view process [...] For example, the interview memos reveal some evidence that
male researchers made assumptions that the girls would give up paid employ-
ment for marriage and motherhood at the first opportunity. Likewise, assump-
tions were also made about the need for male respondents to secure higher paid
jobs in order that they could “ ‘provide’ ” for their future families.”
We consider secondary analysis a very adequate technique for the further
development of qualitative methods. We have carried out a diachronic ana-
lysis of the memos recorded during and/or after each of our interview. These
memos reveal our own subjective impressions; to some extent they resemble
a research diary common in ethno-methodological studies. They can conse-
quently be analysed as materials themselves, in order to disclose the research-
er’s feelings within the interview situation and their projections on the studied
144 Gabriele Abels and Maria Behrens

objects. Alternatively, systematic supervision accompanying the research pro-

cess is possible. In both dissertation projects neither data analysis within a
team nor independent supervision (due to financial reasons) was an option.
This is a quite common situation in social sciences. It is simply assumed that
the scientist him- or herself has an adequate professional distance and enough
reflexive competences. This assumption can be tested by secondary analysis.

6.4 Interaction-effects in expert-interviews

Interaction-effects in expert interviews are assessed differently throughout

the literature. Often they are categorized as failure in the interview’s run
of communication, and as something the scientist cannot change as the
framework of the interview does not permit it (cf. Meuser and Nagel 2005).
The informational content of such conversations is believed to be small.
Such a development of the conversation can however, like the following
examples depict, include very positive elements. We also assume that gen-
der is a constitutive factor in the interview. Gender specific behaviour and
resulting interaction-effects or typical patterns were not only “disturbing”
or “distorting,” but could be used positively for gaining information.7
In what follows we will allude to four effect-types described by Vogel
(1995): paternalism-, catharsis-, iceberg- und feedback-effect, and describe
these from a gender perspective. To these we will add the profile-effect.
However, individual conversations can only seldom be attributed to one
type of effect, as expert-interviews are dynamic social interactions.
Similarly, the effectual type might dominate throughout most of the
interview, but each interview phase may change the basic pattern of a
conversation due to the participants’ adaption or reaction. For example,
the iceberg- effect may be succeeded by a catharsis-effect (cf. ibid., p. 81).
Furthermore, the scientists’ own learning process during the course of the
research project and also within interviews is relevant. In a first step we
will present the primary analysis results, and then re-examine them by
secondary analysis (see Table 6.1).

Table 6.1 Typical effects in conversational interaction

Effect Basic feature

Paternalism effect Manifest goodwill by the interviewee towards interviewer

Catharsis effect Interviewee uses interview as compensation for professional
dissatisfaction
Iceberg effect Interviewee’s disinterest and inert willingness to give out
information
Feedback effect Interviewee tries to reverse the question-answer-context
Profile-effect Interviewee seeks to ‘show off’ in front of the interviewer
Source: Our compilation.
 Interviewing Experts in Political Science 145

6.4.1 Paternalism-effect
The paternalism-effect is typically “a manifest goodwill by the interviewee
towards the interviewer’s research subject” (Vogel, 1995, p. 80; our transla-
tion). We often encountered this effect with men, who approached us women
in quite a fatherly way as we moved within a male-dominated domain. Here
gender, age, professional status and knowledge-hierarchy mingled: the pater-
nalism we as “young” women faced was fed by an informational imbalance
inherent in any expert-interview. Behnke and Meuser (1999, p. 78; our trans-
lation) state that “young female scientists moving within a male-dominated
field, [...] are affected in particular, because in perception of those studied,
the scientists gender-status dominates over their professional position. The
female scientist is perceived as acceptably incompetent.”
A second gender-specific bias was also apparent because we were women
interested in male spheres: biotechnology and the politics behind it. We
could choose between two reactions to these assumptions about our lit-
tle knowledge on these political and scientific technical materials: For one,
we were partially pressured to present our capability in the matter, mean-
ing our own expert-status which we had acquired throughout time. Being
taken despite this seriously offended the female scientist. Then, we often
could choose to use the discriminatory paternalism in a productive way. In
many conversations with experts, we received especially important infor-
mation, because our male respondents felt a dire need to explain things to
us thoroughly. Or they felt a need to state facts more explicitly, because they
assumed that we were unable to correctly analyse these answers.8 These
gendered projections onto us and the therefore existing openness on the
research topic, were (optionally) enforced by our questions, which suggested
to the expert that we must be naïve and simple-minded, appealing to his
willingness to enlighten us.
The secondary analysis confirms this paternalism-effect. One of the inter-
view memos reads “X staged himself as a fatherly authority.” It is interesting to
point to the learn-effect chronicled throughout the timeframe during which
interviews were held. Whereas during the first interview, we were inclined to
contradict and prove wrong our interview-partners in their thinking that we
were uninformed, in consecutive interviews we often accepted the frequent
“father-daughter-relation” in order to use it for our own purpose – gaining
information – as we realized it was easier to receive access to information
in this manner. At this point, the factors gender and age mingled. A hand-
written interview memo from an early conversation reads: “what a macho.”
Aside from linking the gender and age factors, the memo expresses anger to
being treated like a dumb girl. Another interview memo reads: “X was more
insulting than anything, arrogant as a scientist, [...] when he said that it is
difficult to explain certain things to people, who [meaning me] never had
done their own research and who had not written proposals for funding.”
The later diaries document no more entries like the one above or similar
146 Gabriele Abels and Maria Behrens

showing just how difficult it is to weigh the gender-factor and its interaction
with other social factors, like age and qualification.

6.4.2 Catharsis-effect
The catharsis-effect stands for respondents using the interview as compen-
sation in order to give free rein to their professional dissatisfaction (Vogel,
1995, p. 81). The female scientist may be faced with a problem, namely
that the interviewee strays off-topic and the “informational harvest” will
be little. An interview memo reads: “Y enjoyed speaking about ethics and
the German’s cultural problems with genomic science.” At the same time,
there is a chance of receiving detailed statements. We as women, thus our
interpretation, were attributed with a lower status, and therefore perceived
as less threatening than a male interviewer. Therefore, we encountered the
catharsis-effect quite often – but not only in the variation depicted above.
The catharsis-effect might be understood as a broadening of Vogel’s def-
inition of the respondent’s role change from expert to private individual
(for a comparison see Meuser and Nagel, 2005 and in this volume, McKee
and O’Brien, 1983). Respondents reported on family-events, the state their
garden was in or discussed the capabilities of a new vehicle. In these situ-
ations the female scientist can either end the interview and risk not receiv-
ing required information or be patient and redirect the interviewee back to
the topic of interest.
Female scientists may tend to opt for the latter. A role change by the
respondent to the private individual perhaps occurs because women are
attributed to the private sphere. For the female scientist it is hard to return
the conversation to the research topic, as he might perceive this as disin-
terest in his person, endangering the fragile situation of trust. A possible
consequence is that the respondent might chose to stonewall during the
interview’s continuation (iceberg-effect).
The secondary analysis confirms that the return from private person to
expert is particularly difficult. It seems that women are attributed to the
private sphere, and simultaneously her professionalism is ignored. When
returning to being an expert, we often find that the respondent becomes
“bored” while giving out information. While during the first interviews we
evaluated private stories positively as an indicator of an existing “relation-
ship of trust,” later interviews indicate an increasing lack of patience with
straying off-topic in form of a catharsis-effect, which led to tensions dur-
ing some of the conversations. We suppose that male researchers are not
confronted with this specific variation of the catharsis-effect in which the
respondent switches between roles of expert to private individual.

6.4.3 Iceberg-effect
The iceberg-effect describes disinterest and inert willingness to give out
information (Vogel, 1995, p. 79) caused by a variety of reasons. Perhaps the
 Interviewing Experts in Political Science 147

respondent is not a “real” expert on the topic studied, or he or she is not

(or no longer) knowledgeable in the subject area. It could also be due to
a widely-distributed scepticism among interviewed natural scientists and
company representatives towards hypotheses from political science on bio-
technology. An interview memo reads: “he is a full-blooded scientist and
has had nothing to do with political decision-making processes,” overall the
conversation was “difficult and tiring, and it was hard getting him to talk.”
Meuser and Nagel (2005) recommend cutting such conversations off. In
principle, we agree. One should not abort a conversation too early though.
Another reason for the effect may be that the respondent wants to quiz the
interviewer about her knowledge on the subject, about the (political) stance
she takes on the topic, if she is trustworthy, and so on. In case of “liking”
the blockade is abandoned and very gainful conversations might evolve. To
a certain degree this effect is normal at the beginning of every interview,
as a situation of trust should be built up first. This is depicted in the memo
from an interview with a company representative: “X was very distrustful
at first, what did I want to know anyway, and how would I then use the
information, X didn’t want the conversation to be recorded, but eventually
agreed to do so.”
We met the iceberg-effect, in its most rigid form, seldom and broke it in a
relatively short period of time. We assume male interviewers might be con-
fronted more forcefully by this effect, due to “quarrels” about status com-
petence perhaps playing a part in the interview. Tietel describes this very
vividly for a conversation conducted by a male/female research team. The
female interviewer was later completely “butted out” of the conversation
and “was excluded from the men’s rivalry” (Tietel, 2000, p. 8).
We attribute the fact that our interviews showed very little of this type
of the iceberg-effect, on the premise that we as women were received with
far less distrust, and that expectations of us were lower. When encounter-
ing a lack of knowledge on the subject studied by the respondent, these did
not retrieve to a blockade.9 On the contrary, respondents usually attempted
to conceal their lack of knowledge by providing different information or
making general statements in order to save face (for a comparison see the
profile-effect).
The secondary analysis shows that we took on classic female role
behaviour – whether consciously or not – in order to break the iceberg
effect. An interview memo states the iceberg was melted with “admir-
ing statements regarding his importance for the development of genetic
engineering in Germany.” This illustrates that “doing gender” allows for
the use of interview tactics to which male scientists cannot resort to in
the same manner. It also shows that not only the respondents’ behaviour
should be analysed, but also that of female scientists. It is possible we
might have ourselves in some cases provoked a change from iceberg effect
to profile-effect.
148 Gabriele Abels and Maria Behrens

6.4.4 Feedback-effect
Feedback-effect describes those conversational situations, in which respond-
ents try to reverse the question-answer-context (Vogel, 1995, p. 80). The
respondent tries to make the female scientist a co-expert. He or she wants to
ask the female scientist how his or her actions are perceived from a societal
environment, or would be perceived in case of acting. The feedback-effect
may put the female scientist into an awkward situation, when blocking
off questions she might seem impolite and when tolerating the contextual
change she might not receive information or in fact give suggestions on
what socially-accepted or desired answers might be.
In both cases the interview might fail. This effect is in part dependent
on the subject; in terms of a conflict-laden topic such as biotechnology it
might lead to potentially critical statements being challenged and because
social scientists are in general perceived as critical. On the other hand, the
feedback-effect is tied to the specifics of a respondent group. As Dexter
already remarked (cf. Ware and Sánchez-Jankowski, 2006, p. 10) scientists
often try to act as “gatekeepers to information.”
We encountered the feedback-effect only seldom and only after several
interviews had already taken place. Throughout the project’s course, the
female scientist becomes an expert herself. Due to her targeted questions,
the respondent soon understands the female scientist to be knowledgeable,
enticing counter-questions. Particularly in interviews with company repre-
sentatives, the respondent attempted to receive information from the inter-
viewer on their competitor’s positions and strategies or from environmental
or consumer-political associations. Aside from obtaining such information,
the reversal of the question-answer-direction is beneficial for the company
representative as he or she must not reveal much information him- or her-
self. The second project experienced the same: one tried receiving infor-
mation on other respondent’s positions, asking the interviewer’s personal
opinion on topics, asking who had been interviewed as well, or even asking
for an explanation for other actor’s actions. Occasionally, an interesting dis-
cussion can evolve; sometimes “trading information” can help. However,
it also implies a problem: answers might be distorted. We ascribe the few
encounters of this effect-type to the fact that women in political fields are
prima facie mostly not understood as experts, and that we weren’t pres-
sured much during conversations to identify ourselves as capable interview-
partners (or competitors).
The secondary analysis depicts that the feedback-effect primarily, but
not exclusively, was found with experts from enterprises and enterprise
associations. Differences resulting from varying scientific cultures could
be held accountable, too. Aside from gender, organization-specific fac-
tors might deliver an explanation for the seldom arising of the feedback-
effect. Furthermore, it is obvious that in “professionals” from a company,
the organization’s interest and strategic information were employed more
 Interviewing Experts in Political Science 149

often than personal inclination. Here, the gender-factor seemingly slides

into the background.

6.4.5 Profile-effect
We encountered the profile-effect quite often. The respondent seeks to prove
his (or her) high level of knowledge and is very willing to give out informa-
tion overall. One interview memo states briefly: “scientific careerist” who
“wants to breathe fresh wind into the worn out structures of the old men.”
We presume that the profile-effect pops up more often in interviews
between men and women, and that it could therefore be a gender-specific
effect.10 The female scientist might react provokingly, as in “but surely you
don’t have any more detailed knowledge on this, do you?” and thus obtain
information. Or affirm admiration concerning the respondent’s capabilities
in order to “retrieve” more information. In some interviews respondents
went as far in strengthening their profile, and succeeded in patting them-
selves on the back for committing prosecutable actions.
The conversational situation is one of classic gender-specific role distribu-
tion. The man is capable and worldly-wise, the woman admires him. To a
certain degree, the female scientist can be a part of these gender roles. But
there is a great danger of the male respondent dominating the interview as
a private individual.
Aside from this, we encountered a case in which the respondents did not
want to strengthen their profile through showing how knowledgeable they
were on the studied subject, but chose “side show scenery”: one respondent
scheduled a phone appointment with a TV-reporter while one of the authors
was present and listened in. Apparently he sought to demonstrate how highly
demanded his opinion was. Or, one interviewer worked at his PC briefly dur-
ing several times in interview his indispensability and thus signalled his gen-
erosity in agreeing to partake in an interview. The female interviewer should
probably just react with obvious disinterest and signal the admiration he
hoped for shall not arrive. Or the conversation might end with a guided tour
through the lab, whilst proudly showing the new DNA sequencer, which the
female scientist knows only to appreciate in part, because it reminds her of
the “one-armed bandit from the amusement park.”
The secondary analysis shows that interviews with female respondents are
not free of such effects either: “X told a lot about her many public discus-
sions with rivals and irrational fears of genetic engineering.” The analysis
shows too, that we felt superior in the conversational situation most of the
time once the profile-effect surfaced. We could decide when to applaud the
respondent, agree with him or her or “punish” them with disinterest. An
interview memo, for example, recorded “childish behaviour,” because the
respondent wanted to impress with supposedly “manifold important con-
tacts to the elite in politics and the economy.” The memo documents the
interviewer feeling superior.
150 Gabriele Abels and Maria Behrens

Another interview memo points out that the respondent was encouraged
to make his or her mark several times during the interview. The informa-
tional gain from such conversations is questionable, one must distinguish
between “real” information and exaggerations or even intentionally false
information. A mixed-sex project-team can contribute to avoiding such
gender-specific interaction-traps, the disadvantage being that the positive
effects may not be used in one’s favour then.

6.5 Conclusion

The depicted experiences do not permit any generalizations, but allow

merely the formulation of hypotheses that should be systemically exam-
ined in future research. The perspective on the relevance of interaction-
effects could be sharpened: Aside from age, professional status, background
experience and cultural factors, gender is a significant factor. Reports con-
cerning gender-specific interaction-effects clarify that expert interviews are
not partner-like discursive talks between equals. Despite differing research
areas, questioning-styles and experts, we discovered common ground in our
individual conversation contexts. On the one hand, these similarities refer
to the behaviour of the (mostly male) respondents towards us (as women),
but also allude to the findings of the secondary analysis, particularly how
we were “doing gender” in the interview situation and reacted to the behav-
iour by our counterpart. Simultaneously the result of our secondary analysis
showed that in part of the talks with female respondents we experienced
and felt – as several interview memos state – a “sisterhood” or “understand-
ing each other in a ‘womanly fashion’ ” on the “men’s clique” in science and
academia. This is also a gender construction.
Gender-specific biases, whether positive or discriminating, coin the rela-
tionship of interaction. They can be used to her benefit by the female sci-
entist. Defence mechanisms (iceberg-effect) are encountered seldom; other
effects such as the paternalism or profile-effect can be derived to informa-
tional gain. Often damaging to the interview was the often encountered
catharsis-effect, when the respondent switched to the private person.
A diachronic secondary analysis clarifies just how difficult it is to weigh
each of the different social factors that are intertwined and interact in relation
to the complex cause-and-effect-relationship. This points to a considerable
need to continue the development of research methods. In open and semi-
structured expert-interviews there is also the general problem that political
scientists are interested in revealing real-life procedures, while respondents
might seek to cover facts behind a veil and aim at creating political myths,
because “organisations, too, work in self-interest toward a most positive and
consistent depiction of their operations” (Hucke and Wollmann, 1980, p. 224;
our translation). One must assume that “political desirability” impinges on all
methods of data collecting in every controversial field of politics.
 Interviewing Experts in Political Science 151

From a methodologist point of view we call for a triangulation of tech-

niques. Detecting traps in methodology does not mean to disapprove the
instrument of research as a whole. Expert interviews are indubitably of def-
inite interest, especially with a field that is yet to be explored. Standardized
methods are often inadequate for seeking expert-knowledge (for example
the Delphi survey, cf. Aichholzer in this volume), because the required
reduction of reality is contrary to the expert’s experience.
According to Meuser and Nagel (1991, p. 133f.) the cognitive gain from
expert-interviews are more in line with a sociology of knowledge approach.
Hence, they lie on a level of interpretation-patterns, orientation-patterns
and patterns of normality; regarding political operations they aim at “struc-
tures of relevance and knowledge patterns in the political system.” Also,
political activities are “things with an own quality” (Hucke and Wollmann,
1980, p. 220; or translation). Grasping of every day political operational
practice can only be reconstructed in detail on the basis of these interviews.
Therefore “iterative expert-interviews” are especially common in policy
studies, in particular for a qualitative analysis of networks (Voelzkow, 1995,
p. 55). Documents, such as, for example, legislative bills, press releases, state-
ments or protocols however are – depending on their type – usually only
“landmarks,” albeit of relevance; but the empiric complexity of political
actions are only in part mirrored in them.
Experts are problematic informants (Dean and Whyte, 2006) leading to
pitfalls attached to this research technique. When confronting these issues,
we suggest the following possibilities: it is ubiquitous to combine expert-
interviews with other methods of data-collection, holding many interviews
with experts corresponding to different organizational and institutional
contexts in order to validate the data by means of a thematic comparison.
The validity and reliability would greatly benefit from being carried out
by a team, or at least if they were evaluated by one. Furthermore supervi-
sion should render the possibility of reflecting on each individual “rela-
tional interview-space” and to analyse its meaning for data collection and
evaluation.
Concerning gender effects we see significant need for methodological
research in order to improve research techniques: they can be recon-
structed in an ex post secondary analysis using transcripts and interview
memos. In the interest of an increase in methodological self-reflection
secondary analyses should be carried out by scientists who are not part
of the project-team. However, here we encounter the dilemma of hand-
ling confidential information (for a comparison cf. van den Berg, 2005).
Also a comparison between experiences from male colleagues, but also
including expert-interviews with female respondents, might be insightful.
Nevertheless, much fails as political operative spheres are still dominated
by male actors, and herewith gender-specific aspects already exist during
the sampling-phase.
152 Gabriele Abels and Maria Behrens

Mixed-sex research groups in particular grant the opportunity for a system-

atic comparison. Unfortunately these proposals fail – due to the economic
reasons – but also due to a “receptive block” (Dackweiler, 2004, p. 51; our
translation) in political science towards gender-theoretic approaches, such
as a consequent opening towards methodological questions. In the interest
of a methodological “duty of care,” this resistance should be overcome.

Notes
1. In the year 2003 an ad-hoc group “empirical methods in political science” of the
German Political Science Association (DVPW) was founded, which has the status
of a permanent research group since 2006. Furthermore the thematic sections and
working groups regularly debate on methods.
2. In our understanding the English terms “data-making” or “data-creation pro-
cedures” allude precisely to methods primarily producing data.
3. The term refers to individuals, which “[are] part of the operative field, whose prob-
lems are sought to be solved.” In this sense our interviewees would belong to the
functional elite. The term elite and the processes to identify the members are how-
ever subject to controversy; cf. Meuser and Nagel, 1994, pp. 181–4.
4. Inverting a formulation by Becker-Schmidt (1985, p. 95) formulated for the fem-
inist debate it may be questioned for expert-interviews if the object status of the
expert is missed because their subjectivity is denied.
5. For lack of space we only mention the most important information on the
projects, for a more detailed description see Abels and Behrens, 2002. One pro-
ject reconstructed political processes on the EC-Human Genome Programme
(Abels, 2000). For this purpose many actors from project development from sci-
ence and research administrations were interviewed. The other project (Behrens
and others, 1996) compares the implementation of research programmes in the
field of biotechnology in The Netherlands and The Federal Republic of Germany
and analysed the process of societal context-formation of genetic procedures for
food stuff production (Behrens and others, 1997 and 2001). Aside from political
experts, company representatives were interviewed. The majority of conversations
was recorded, in single cases this was not possible or the respondents did not agree
to being recorded. In total 127 interviews were conducted. Most of the 107 male
respondents were between 50 and 60 years old, 22 of female respondents between
35 and 45 years of age. Almost all respondents had an academic background,
many had acquired a PhD, some even habilitated. Most had studied the natural
sciences, many were lawyers or economists. Most of the time actors corresponded
to the lower and mid-levels of the hierarchy, and were included in the preparing of
decisions and their implementation (for example as subject specialists) and there-
fore had detailed knowledge, and were only in very rare cases part of the leading
elite. The authors/interviewers are political scientists and were in their early and
mid-thirties, striving for their PhDs, when conducting the interviews.
6. On problems of interview in a conflict-laden context see Kacen and Chaitin,
2006.
7. By consciously using gender-specific biases to attain information, one risks
strengthening existing gender stereotypes. However, it is fairly common to make
use of biases held by the experts towards the scientists, not only concerning gen-
der-specific questions, but also concerning, for example, political attitudes that
 Interviewing Experts in Political Science 153

are expressed with help of certain clothes (Schmid, 1995). Whether or not the
end justifies the means is a question of research ethics.
8. There are also forms of cultural and ethical paternalism, as an interview-note
proves: “XY was (as is typical for Brits!) very sceptical of EU-Programmes.”
9. We only ended a conversation earlier than intended in two cases.
10. Furthermore we assume that also in conversations between men a profile-effect
takes place in relation to conflicts over status and competencies. This one is also
gender-specific, due to this being a display of masculinity. The staging of gender
can also be found in the paternalism-effect in the figure of the fatherly type.

Further readings
Bryant, L. and Hoon, E. (2006) “How can the intersections between gender, class,
and sexuality be translated to an empirical agenda?” in International Journal of
Qualitative Methods 5(1), http://www.ualberta.ca/~iiqm/backissues/5_1/pdf/ bryant.
pdf, accessed 14 May 2007.
Corti, L. and Thompson, P. (2004) “Secondary analysis of archive data” in Seale, C.,
Gobo, G., Gubrium, J. F. and Silverman, D. (eds) Qualitative Research Practice
(London: Sage), 327–43.
Sarantakos, S. (2004) Social Research, 3rd edn. (Houndsmill: Palgrave).

References
Abels, G. (2000) Strategische Forschung in den Biowissenschaften. Der Politikprozess zum
europäischen Humangenomprogramm (Berlin: Edition Sigma).
Aberbach, J. D., Chesney, J. D. and Rockmann, B. A. (1975) “Exploring elite political
attitudes: some methodological lessons” in Political Methodology 2(1), 1–27.
Alemann, U. von (ed.) (1995) Politikwissenschaftliche Methoden. Grundriss für Studium
und Forschung (Opladen: Westdeutscher Verlag).
Alemann, U. von and Forndran, E. (1990) Methodik der Politikwissenschaft, 4th edn
(Stuttgart, Berlin, Köln: Kohlhammer).
Becker-Schmidt, R. (1985) “Probleme einer feministischen Theorie und Empirie in
den Sozialwissenschaften” in Feministische Studien 4(2), 93–104.
Behnke, C. and Meuser, M. (1999) Geschlechterforschung und qualitative Methoden
(Opladen: Leske & Budrich).
Behnke, J., Baur, N. and Behnke, N. (2006a) Empirische Methoden der Politikwissenschaft
(Paderborn, München, Wien, Zürich: Schöningh).
Behnke, J., Gschwend, T., Schindler, D. and Schnapp, K. -U. (eds) (2006b) Methoden
der Politikwissenschaft. Neuere qualitative und quantitative Analyseverfahren (Baden-
Baden: Nomos).
Behrens, M. (2001) Staaten im Innovationskonflikt. Vergleichende Analyse staatlicher
Handlungsspielräume im gentechnischen Innovationsprozess Deutschlands und den
Niederlanden (Frankfurt am Main: Peter Lang).
Behrens, M. (2003) “Quantitative und qualitative Methoden in der Politikfeldanalyse”
in Schubert, K. and Bandelow, N. C. (eds) Lehrbuch der Politikfeldanalyse (München,
Wien: Oldenbourg), pp. 205–38.
Behrens, M., Meyer-Stumborg, S. and Simonis, G. (1996) Gentechnik und die
Nahrungsmittelindustrie (Opladen: Verlag für Sozialwissenschaften).
Behrens, M., Meyer-Stumborg, S. and Simonis, G. (1997) GenFood: Einführung und
Verbreitung, Konflikte und Gestaltungsmöglichkeiten (Berlin: edition sigma).
154 Gabriele Abels and Maria Behrens

Behrens, M. and Hennig, E. (2009) “Qualitative Methoden der Internationalen

Politik” in Wilhelm, A. and Masala, C. (eds) Handbuch der Internationalen Politik
(Wiesbaden: VS Verlag) (forthcoming).
Berg, H. v. d. (2005) “Reanalyzing Qualitative Interviews from Different Angles: The
Risk of Decontextualization and Other Problems of Sharing Qualitative Data” in
Forum Qualitative Sozialforschung/Forum: Qualitative Social Research 6(1), art. 30,
http://www.qualitative-research.net/fqs-texte/1-05/05-1-30-e.htm, date accessed
16 May 2007.
Brosi, W. H., Hembach, K. and Peters, G. (1981) Expertengespräche. Vorgehensweise
und Fallstricke. Arbeitspapier des Forschungsprojekts Berufliche Bildung und regionale
Entwicklung (Trier: Universität).
Bryant, L. and Hoon, E. (2006) “How can the intersections between gender, class,
and sexuality be translated to an empirical agenda?” in International Journal
of Qualitative Methods 5(1), http://www.ualberta.ca/~iiqm/backissues/5_1/pdf/
bryant.pdf, date accessed 14 may 2007.
Corti, L. (2000) “Progress and Problems of Preserving and Providing Access to
Qualitative Data for Social Research − The International Picture of an Emerging
Culture” in Forum Qualitative Sozialforschung / Forum: Qualitative Social Research 1(3),
http://qualitativeresearch.net/fqs/fqs-eng.htm, date accessed 14 May 2007.
Corti, L. and Thompson, P. (2004) “Secondary analysis of archive data” in Seale, C.,
Gobo, G., Gubrium, J. F. and Silverman, D. (eds) Qualitative Research Practice
(London: Sage), pp. 327–43.
Dackweiler, R.-M. (2004) “Wissenschaftskritik – Methodologie – Methoden” in
Rosenberger, S. K. and Sauer, B. (eds) Politikwissenschaft und Geschlecht (Wien: UTB-
WUV), pp. 45–63.
Dale, A., Arber, S. and Proctor, M. (1988) Doing Secondary Analysis (London: Unwyn
Hyman).
Dean, J. P. and Foote Whyte, W. (2006) “How do you know if the informant is tell-
ing the truth?” in Dexter, L. A. Elite and Specialized Interviewing (Colchester: ECPR
Press), pp. 100–7.
Deeke, A. (1995) “Experteninterviews – ein methodologisches und forschungsprak-
tisches Problem” in Brinkmann, C., Deeke, A. and Völkel, B. (eds) Experteninterviews
in der Arbeitsmarktforschung. Diskussionsbeiträge zu methodischen Fragen und prak-
tischen Erfahrungen. Beiträge zur Arbeitsmarkt- und Berufsforschung Nr. 191 (Nürnberg:
Bundesanstalt für Arbeit), pp. 7–22.
Dexter, L. A. (1970/2006) Elite and Specialized Interviewing (Colchester: ECPR Press).
Ebbecke-Nohlen, A. and Nohlen, D. (1994) “Feministische Ansätze” in Nohlen, D.
(ed.) Lexikon der Politik, vol. 2: Politikwissenschaftliche Methoden (München:
Beck), pp. 130–37.
Goodwin, J. and O’Connor, H. (2006) “Contextualising the Research Process: Using
Interviewer Notes in the Secondary Analysis of Qualitative Data” in The Qualitative
Report 11(2), 374–92.
Gurney, J. N. (1985) “Not one of the guys: The female researcher in a male-dominated
setting” in Qualitative Sociology 8(1), 42–62.
Heinzel, F. (1997) “Biographische Methode und wiederholte Gesprächsinteraktion.
Ein Verfahren zur Erforschung weiblicher Politisierungsprozesse” in femina politica:
Zeitschrift für feministische Politik-Wissenschaft 6(1), 96–104.
Helfferich, C. (2004) Die Qualität qualitativer Daten. Manual für die Durchführung quali-
tativer Interviews (Wiesbaden: Verlag für Sozialwissenschaften).
 Interviewing Experts in Political Science 155

Hoffmann-Riem, C. (1980) “Der Datengewinn” in Kölner Zeitschrift für Soziologie und

Sozialpsychologie 32, 339–72.
Hucke, J. and Wollmann, H. (1980) “Methodenprobleme der Implementationsforschung”
in Mayntz, R. (ed.) Implementation politischer Programme: empirische Forschungsberichte
(Königstein/Ts.: Anton Hain), pp. 216–35.
Kacen, L. and Chaitin, J. (2006) “ ‘The Times They are a Changing’: Undertaking
Qualitative Research in Ambiguous, Conflictual, and Changing Contexts” in The
Qualitative Report 11(2), 209–28.
Kriz, J., Nohlen, D. and Schultze, R. -O. (eds) (1994) Lexikon der Politik, vol. 2:
Politikwissenschaftliche Methoden (München: Beck).
Lamnek, S. (1988/1989) Qualitative Sozialforschung, vols 1 and 2 (München: Psychologie
Verlags Union).
Littig, B. (2005) “Interviews mit Experten und Expertinnen. Überlegungen aus
geschlechtertheoretischer Sicht” in Bogner, A., Littig, B. and Menz, W. (eds) Das
Experteninterview – Theorie, Methode, Anwendung, 2nd edn (Wiesbaden: Verlag für
Sozialwissenschaften), pp. 191–206.
McKee, L. and O’Brien, M. (1983) “Interviewing men: taking gender seriously” in
Gamarnikow, E., Morgan, D. H. H., Purvis, J. and Taylerson, D. (eds) The Public and
the Private (London: Heinemann), pp. 147–61.
Meuser, M. and Nagel, U. (1991) “Experteninterviews als Instrument zur Erforschung
politischen Handelns” in Bering, H., Hitzler, R. and Neckel, S. (eds) Politisches
Handeln/Experteninterviews. Dok. Nr. 1 des AK Soziologie politischen Handeln (Bamberg:
University press), pp. 133–40.
Meuser, M. and Nagel, U. (1994) “Expertenwissen und Experteninterview” in
Hitzler, R., Honer, A. and Maeder, C. (eds) Expertenwissen. Die institutionalisierte
Kompetenz zur Konstruktion von Wirklichkeit (Opladen: Westdeutscher Verlag),
pp. 180–92.
Meuser, M. and Nagel, U. (1997) “Das ExpertInneninterview – Wissenssoziologische
Voraussetzungen und methodische Durchführung” in Friebertshäuser, B. and
Prengel, A. (eds) Handbuch Qualitative Forschungsmethoden in der Erziehungswissenschaft
(Weinheim: Juventa Verlag), pp. 481–91.
Meuser, M. and Nagel, U. (2005) “ExpertInneninterviews – vielfach erprobt, wenig
bedacht. Ein Beitrag zur qualitativen Methodendiskussion” in Bogner, A., Littig, B.
and Menz, W. (eds) Das Experteninterview – Theorie, Methode, Anwendung, 2nd edn
(Wiesbaden: Verlag für Sozialwissenschaften), pp. 71–93.
Padfield, M. and Procter, J. (1996) “The effect of interviewer’s gender on the inter-
viewing process: a comparative enquiry” in Sociology 30(2), 355–66.
Patzelt, W. (1991) “Politikwissenschaft” in Flick, U., Kardorff, E. von, Keupp, H.,
Rosenstiel, L. von and Wolff, S. (eds) Handbuch Qualitative Sozialforschung (München:
Psychologie Verlags Union), pp. 53–5.
Patzelt, W. (1994) “Qualitative Politikforschung” in Kriz, J., Nohlen, D. and
Schultze, R.-O. (eds) Lexikon der Politik, vol. 2: Politikwissenschaftliche Methoden
(München: Beck), pp. 395–98.
Schendelen, M. P. C. M. van (1984) “Interviewing members of parliament” in Political
Methodology, 10(3), 301–21.
Schmid, J. (1995) “Expertenbefragung und Informationsgespräch in der
Parteienforschung: Wie föderalistisch ist die CDU?” in Alemann, U. von (ed.)
Politikwissenschaftliche Methoden. Grundriss für Studium und Forschung (Opladen:
Westdeutscher Verlag), pp. 293–26.
156 Gabriele Abels and Maria Behrens

Semmel, A. K. (1975) “Deriving perceptual data from foreign policy elites: a

methodological narrative” in Political Methodology 2(1), 29–49.
Tietel, E. (2000) “Das Interview als Beziehungsraum” in Forum Qualitative
Sozialforschung/Forum: Qualitative Social Research [Online Journal], 1(2), http://
qualitative-research.net/fqs/fqs-d/2-00inhalt.htm, accessed 13 May 2007.
Voelzkow, H. (1995) “ ‘Iterative Experteninterviews’: Forschungspraktische Erfahrungen
mit einem Erhebungsinstrument” in Brinkmann, C., Deeke, A. and Völkel, B. (eds)
Experteninterviews in der Arbeitsmarktforschung. Diskussionsbeiträge zu methodischen
Fragen und praktischen Erfahrungen. Beiträge zur Arbeitsmarkt- und Berufsforschung
Nr. 191 (Nürnberg: Bundesanstalt für Arbeit), pp. 51–57.
Vogel, B. (1995) “ ‘Wenn der Eisberg zu schmelzen beginnt ...’ – Einige Reflexionen
über den Stellenwert und die Probleme des Experteninterviews in der Praxis der
empirischen Sozialforschung” in Brinkmann, C., Deeke, A. and Völkel, B. (eds)
Experteninterviews in der Arbeitsmarktforschung. Diskussionsbeiträge zu methodischen
Fragen und praktischen Erfahrungen. Beiträge zur Arbeitsmarkt- und Berufsforschung
Nr. 191 (Nürnberg: Bundesanstalt für Arbeit), pp. 73–83.
Ware, A. and Sánchez-Jankowski, M. (2006) “New Introduction” in Dexter, L. A. Elite
and Specialized Interviewing (Colchester: ECPR Press), pp. 1–11.
Warren, C. A. B. (1988) Gender Issues in Field Research (London: Sage).
Williams, C. L. and Heikes, E. J. (1993) “The importance of researcher’s gender in the
in-depth interview” in Gender and Society 7(2), 280–91.
7
Expert Interviews on the Telephone:
A Difficult Undertaking
Gabriela B. Christmann

7.1 Introduction: reason and methodical context of the study

The topic of this contribution is the method chosen for a sub-project, which
was carried out in the context of a more extended study on “Demographic
Change at Universities,” Saxony being the focal point.1 For the sub-project
qualitative expert interviews were used, part of them being conducted on
the telephone. For the time being, qualitative telephone (expert) interviews
have seldomly been discussed in literature. By this contribution the few
indications offered by literature and our own methodical experiences will
be reflected on.
As already indicated, the sub-project was integrated into an extended
project – consisting of five sub-projects and multi-methodically planned –
in the field of university research (see Figure 7.1). The starting point of
the entire project was the question of which effect demographic change
will have on the university system and the academic labour market par-
ticularly in Saxony. Based on existing statistics, the first sub-project was
about presenting predictions on the future numbers of university begin-
ners in Saxony and on estimating the offer of academic graduates until
2035. The second one predicted the need in numbers of academically edu-
cated workforce on the Saxonian labour market until 2020. To give rea-
son to predictions and estimations, a quantitative survey among experts
by way of the Delphi method was carried out in the course of the third
sub-project. By way of given items, experts from Saxony and the other
federal states were asked to give their estimations on expected trends
on the labour market, in the field of universities, and in educational
policy. Accordingly, experts came from the fields of “business,” “educa-
tional/university system,” and “politics.” As it had to be assumed that
steps by decision-makers at universities and in the field of the politics
of the federal state may considerably influence the demand for studies,
more deep-reaching studies by way of qualitative expert interviews with
decision-makers2 were carried out by a fourth and a fifth sub-project.

157
158 Gabriela B. Christmann

Project “Hochschulen im demografischen Wandel. Die Lage in Sachsen (Demographic

Change at Universities. The Situation in Saxony)”
Asking about the effects of demographic change on the university system and the labour market for
academics

sub-project I sub-project II sub-project III sub-project IV sub-project V

Prediction on Prediction on Giving reason to Giving reason to the Giving reason to the
the number the need of the predictions of predictions of SP I as predictions of SP I
of university academic SP I and II by well as a description as well as a
beginners workforce way of trend of concrete strategies description of
and on the estimations by in the action field concrete concepts in
offer of experts (Delphi “university” the action field
academic Survey) (qualitative expert “politics”
graduates interviews) (qualitative expert
interviews)

Step 1
Nation-wide written
survey among all
universities and
colleges on planning
strategies (short
questionnaire)

Step 2
Selecting experts on
the basis of
universities / colleges
answering and on the
basis of homepage
analyses of
universities / colleges
not answering

Step 3
Qualitative expert
interviews (some
face-to-face, some on
the telephone)

Figure 7.1 The research design in the context of the extended project
Source: Developed by the author.

With them, the questions of which concrete strategies are developed at

universities to be fit for competition in the future (sub-project IV) and
which steps are planned by political actors to organize the university sys-
tem (sub-project V) were in the fore. Ideas from other federal states were
supposed to be included into both sub-projects, due to which decision-
makers from the entire Federal Republic were taken into consideration.
Thus, the overall project pursued an application-related goal: founda-
tions for university-planning and educational-political acting were sup-
posed to be provided.
 Expert Interviews on the Telephone 159

For this contribution the method of the sub-project is in the fore, which
dealt with strategies in the action-field of universities (see the specially
marked column in Figure 7.1).
The sub-project was meant to analyse if and – if yes – how (in times of
demographic change) universities make their institutions attractive for
future students. “Best practice” examples of demography-according univer-
sity planning should be identified and described. On the one hand results
served for giving reason to assumptions by the prediction model of sub-
project I particularly for Saxony, and on the other hand for providing infor-
mation on further strategies of nationwide university planning.
By a first step, a written survey among universities by way of a short
questionnaire was carried out. The questionnaire addressed university
leaderships.3 It asked about essential topics of university planning by way of
standardized and open questions. Questions were derived from expert lit-
erature on university research and from the public discourse on university
development in Germany. A total of 268 institutions of academic education
received this questionnaire. After a “reminding action” it was possible to
achieve a total return of 41 per cent.
Based on this, at first information on basic strategies of universities was
gained. It served for selecting universities by a second step which seemed
to be promising for the then following – now: qualitative – round of expert
interviews. As by the short questionnaire universities were also asked to
name contact partners for a possible qualitative expert interview, it was
not difficult anymore to win over concrete interview partners. However,
we did not always follow the suggestions (see Chapter 2). Selecting those
universities, which were supposed to be included into the sample of the
qualitative expert survey was not only based on those universities as having
participated in the written survey. It was additionally based on homepage
analyses (of offers of studies, models) of those universities, which had not
answered in the context of the written survey. On the whole, such uni-
versities were chosen which according to our analysis looked particularly
active and innovative in respect of strategy development. In retrospect it
turned out that a great part of universities chosen this way also belonged to
the group of those universities as being chosen for support in the context
of the initiative of excellency of the Federal Government and the federal
states. By a third step, a total of 22 qualitative expert interviews were carried
out.4 As already indicated, these expert interviews were meant to achieve
deeper insight into how institutions of academic education prepare for the
future. The experts were asked which strategies of university planning their
respective university had actually developed and was expected to develop
and which experiences they had made with previous steps.5
There were 14 face-to-face and eight telephone interviews. The decision
for expert interviews on the telephone was motivated by research-economic
reasons – in respect of the project’s chronological and financial resources.
160 Gabriela B. Christmann

Our assumption was that if telephone interviews were an appropriate way

at all in the context of qualitative analyses, this would most probably be
the case with a project like ours, as the interview partners were academ-
ically educated experts holding positions with university leadership/
administration, who – as we could assume – showed the following core
competences: a high ability to think in the abstract, strong orientation at
topical criteria, high level of competence in respect of giving explanations,
experience with presenting themselves to the outside world6 by way of
interviews, even on the telephone, most of all towards journalists but also
towards fellow scientists. Furthermore, our research interest was not directed
at the reconstruction of life world or implicit knowledge, that is a kind of
knowledge being “difficult to ask about,” but rather at finding out about
estimations, experiences, suggestions and intended ways of action within
a field of practical acting and thus at explicit, reflected knowledge, a good
deal of which is characterized by inner-university negotiation processes and
to a certain extent by public discourses on the development of universities.
Given this target group and these questions, we assumed that telephone
interviews were suitable for extending those data as being collected by way
of face-to-face interviews and that we would not be confronted with more
serious methodical problems.
As intended, we were able to topically extend our study by way of tele-
phone interviews. Also experts interviewed on the telephone provided us
with much valuable information about their actual strategies of university
planning; it was even possible to find out about new aspects in this way.
However, interviews happened in a less unproblematic way than expected
by us. This will be described in the following (see Chapter 4). Before this,
however, there will be reflecting on which kind of interview was carried
out or which kind of interaction situation was typical for collecting data
(see Chapter 2). Furthermore, there will be a critical look at the literature on
methods of telephone (expert) interviews (see Chapter 3).

7.2 Explorative-systematizing expert interview at eye level

It has already been mentioned that experts were (not only but most of them)
recruited by way of written questionnaires. There we had asked for nam-
ing contact partners for possible oral expert interviews. This request was
met by almost all universities answering. Usually, universities named their
vice presidents as contact partners, colleges always named their presidents.
Sometimes both universities and colleges recommended the president’s
personal assistant, the head of the department of planning and develop-
ment, the head of the department of academic matters, or one of their study
advisers. If study advisers had been named as contact partners, in most
cases we did not follow the recommendation, as we were rather interested
in people being included in university planning at a higher level. In this
 Expert Interviews on the Telephone 161

context we assumed that experts are found not only at the level of immedi-
ate decision-making (members of university leaderships) but also at the level
of preparing decisions (assistants and heads of departments).7
Always the first contact was made by e-mail, that is both for face-to-face
interviews and for telephone interviews (on this see also Gläser and Laudel,
2006, pp. 153–61; Mieg and Näf, 2006, p. 26). Potential interview partners
were shortly informed about the research project. Most of all they were asked
if they were ready to give an interview and if they agreed with an audio-
recording. Also, at the same time they were sent the interview guideline,
after first experiences had shown that experts (that is particularly interview
partners intended for telephone interviews) connected their agreement to
being informed about actual questions in advance and in more detail (see
Chapter 4). After their agreement the dates for the interviews were fixed on
the telephone, sometimes via secretaries.
Bogner and Menz (2005b, see also Bogner and Menz, in this volume) for
their typology distinguish between the explorative, the systematizing, and
the theory-generating interview (see also Vogel, 1995, p. 74). On the one
hand, our project might be counted among the category of the systematiz-
ing interview because it was about finding out about “action knowledge
and experience-based knowledge gained by practical work and being reflex-
ively available and spontaneously communicatable.” On the other hand,
however, it was not that a strict “topical comparability of data in the fore,”
due to which this attribution is not completely adequate. It would also be
inappropriate to call our expert interviews purely explorative, even more as
according to Bogner and Menz (2005b, see also Bogner and Menz, in this
volume) the latter only serve “for preliminary orientation in a topically new
or confusing field.” However, one maxim of explorative interviews was def-
initely a guideline for us: the interviews were supposed to be as open as pos-
sible, and were supposed to find out about the universities’ action or, to have
it more exactly: planning emphasis. Thus in respect of method, interviews
practiced by us are of a mixed type, here we like to call them explorative-
systematizing expert interviews.
The basis of the interview was an interview guideline. This guideline con-
sisted of four main questions which were very openly designed, were aimed
at inner-university strategies, and which were asked all interview part-
ners. They were followed by additional questions, which addressed a total
of 11 subjects and for the purpose of improved clarity were divided into
three blocks. For – as shown by the analysis of our short questionnaires –
not every university has worked out concepts on every topical field, not for
every interview was it necessary to “work off” all additional questions. As
already mentioned, the study did not aim at all interviews to be compar-
able but at finding out about “best practice” examples from certain fields.
As the interview guideline was previously given to the interview partners
by e-mail, they were told at the same time that they would not have to deal
162 Gabriela B. Christmann

with every aspect of the additional questions but only with those being
relevant for their university/college or with those they considered relevant.
This way it was guaranteed that on the one hand the experts perceived
our range of interests. On the other hand, however, they could lay their
emphasis on those fields in which they were able to talk about concepts and
experiences. As Vogel (1995, p. 76) has it, this was important not at least for
atmospheric reasons, for the interviewees were supposed to feel they were
being taken seriously as a competent partner (status!). Thus, the experts
were provided with enough leeway for those aspects as being important for
them. Within a certain frame, in respect of answering the questionnaire
they were able to structure their use of time according to their own pref-
erences. However, indeed further questions by the interviewers had a cer-
tain effect. Both for face-to-face and for telephone interviews this strategy
proved to be practical.
Different from what Bogner and Menz (2005b, see also Bogner and Menz,
in this volume) state to be typical for the type of the systematizing inter-
view, the interviewer was in a position to have expert knowledge of the
topical fields she was asking about. She was familiar with cross-university
discourses and relevant research literature influencing universities’ internal
debates. However, she was not herself included into the experts’ respective
contexts of university planning and decision-making, but she was provided
with basic knowledge. Thus, she approached the experts not as a “lay-
woman” but to a certain degree as a “co-expert.”8 This is not to say that the
expert interviews took on the nature of debates among experts (see Bogner
and Menz, 2005b, see also Bogner and Menz, in this volume). Rather, the
expert’s knowledge advantage in his or her particular field and context was
taken seriously. But in any case it was possible to conduct the interviews “at
eye level” (Pfadenhauer, 2005, see also Pfadenhauer, in this volume).
The strategy for additional questions was based on the “question technique
between pre-knowledge and naivety” as suggested by Walter (1994, p. 274).
Walter pointed out to the fact that it is favourable if interviewers include a
certain degree of pre-knowledge into the interview. Apart from the fact that
this will increase the interviewer’s reputation, it challenges the expert to
deal with the topic more intensively and in more detail. At the same time
it will also be advantageous if naïve questions are asked. This will provoke
answers “which do not in any case fit to the interviewer’s expectations, and
it will bring up aspects which rather express the interviewee’s point of view”
(Walter, 1994, p. 274; on this see also Gläser and Laudel, 2006, p. 129f).

7.3 Entering methodically “unknown territory”:

the qualitative-oriented expert interview on the telephone

In several respects, extended literature research on qualitative telephone

interviews in the Anglo-American and German-speaking countries produce
 Expert Interviews on the Telephone 163

a disappointing result: the number of methodical-systematic contributions is

extremely small, furthermore there are only essays with a low number of
pages (five to eight pages), and in respect of content one misses fundamental
reflections on the suitability of this kind of interview for the field of quali-
tative social research.
However, this is not meant to say that qualitative telephone (expert) inter-
views are not conducted. According to statements by Opdenakker (2006,
Abstract), this kind of interview has become common: “Face-to-face inter-
views have long been the dominant interview technique in the field of
qualitative research. In the last two decades, telephone interviewing became
more and more common.”
However, this statement may be supposed to be true most of all for Anglo-
American countries. There, qualitative interviews are used rather frequently
and also very pragmatically (see for example studies by Brody, Geronemus
and Farris, 2003, Galster, 2006, Mitchell and Zmud, 1993). These works are
very much topic-related and provide only little information about the actual
methodical way of proceeding. Discussing methods is completely missing.
In German-speaking countries qualitative telephone interviews are com-
paratively less common. However, a number of entries on current research
projects in the “Informationssystem Sozialwissenschaften” (FORIS) indicate
that the use of qualitative telephone interviews is increasing. Some entries
make obvious that qualitative telephone interviews with decision-makers
and people having special knowledge – that is experts – are intended for the
respective projects. Nevertheless, interviews in the context of these projects
are not called expert interviews. Thus it becomes obvious that qualitative-
oriented expert interviews on the telephone are definitely used but that this
methodical practice is not reflected on.
The few contributions, which are explicitly meant as methodical con-
tributions to qualitative telephone interviews (see Burke and Miller, 2001,
Busse, 2000) do not offer anything else than practical-technical instruc-
tions on how to proceed. There are no (explicit) considerations on the
question of which kind of target group or question this method is suit-
able, or not.
However, it is remarkable that Busse (2003, p. 28, own translation) at
the beginning of his contribution calls the telephone interview an “expert
interview on the telephone.” But this is done only in passing. Thus, impli-
citly he restricts the application field of the qualitative telephone interview
to expert interviews, without explaining in more detail or giving reasons.
Instead, Busse speaks of the advantage of the telephone interview: “The
guideline-based telephone interview, that is the expert interview on the
telephone by help of a question catalogue, is a very efficient and economic
kind of qualitative data survey, as neither travelling is necessary nor are
there any expenses if an interview shall be conducted” (Busse, 2003, p. 28,
own translation).
164 Gabriela B. Christmann

In the following paragraphs the author will explain in detail how a guide-
line must be constructed, how contact is made with potential interview
partners, and how a telephone interview is practically done. Other than it
is mostly the case with contributions on qualitative face-to-face interviews,
he recommends a strongly structured guideline which “is similar to a stand-
ardized questionnaire, the only difference being the fact that it includes
more so called “open questions” than the latter” (Busse, 2003, p. 30, own
translation).
By the way, here a phenomenon becomes obvious which Bogner and Menz
(2005b, this volume) state also for face-to-face interviews with experts: even
if interviews are conducted as partly structured interviews, they must not be
considered “genuine representatives of the qualitative paradigm.”
According to Busse (2003), the guideline should be organized for an
interview of about 45 minutes. For making contact he recommends a
three-step-procedure: By way of a letter – purposefully addressing the poten-
tial interview partner – the expert shall at first be informed about the goals
and the subject of the project. This letter should also announce a first tele-
phone call. This call should be some days later and inform the expert about
what to expect in respect of time expenditure and content, to then talking
about his readiness for an interview. By a third step, the actual telephone
interview is conducted.
In the context of conducting the actual interview Busse (2003) also
emphasizes the advantages, which in his opinion are connected to the
guideline being formulated and structured as best as possible. The latter
is said to be the best means to prevent problems of understanding which
may occur particularly with telephone interviews, as “during the interview”
one “is restricted to the verbal part of communication” (Busse, 2003, p. 31,
own translation). It is outspokenly important, Busse (2003, p. 31, own
translation) says, “to formulate questions as clearly and precisely as pos-
sible.” Finally the author discusses the necessity of further asking in case
of uncertainties, of the audio recording, and of the interview record, in the
context of which it stays unclear if by this “record” he means a complete
word-for-word transcript.
On the whole, a certain sensitiveness for the special nature of the tele-
phone interview in comparison to the face-to-face interview becomes obvi-
ous in Busse (2003). As early as at the beginning of his contribution he states:
“However, one disadvantage of this method is in the fact that in contrast
to the common interview procedure, when one sits face to face with one’s
interview partner, quite an important part of human communication (ges-
tures, facial expressions) is lost: one is restricted to the interview partner’s
voice as a source of information, and he or she only hears the interviewer’s
voice” (Busse, 2003, p. 28, own translation).
As we have already seen, however, the author suggests the interview to be
strongly structured as a possible solution for potential interaction problems.
 Expert Interviews on the Telephone 165

This way advantages are given away which are promised by (more) openly
structured interviews, something, which Busse (2003) does not discuss,
however.
In Burke’s and Miller’s (2001) contribution discussions in respect of the
technically mediated kind of interaction of the telephone interview are
completely missing. Explicitly, the authors understand their explanations
to be “practical recommendations,” most of all they address novices in the
field of qualitative interviews (Burke and Miller, 2001, p. 3). Against this
background they deal in very much detail with subjects such as the pre-test,
recording technology, making contact, and scheduling.
Burke and Miller (2001, p. 4) structure the interview process by three
phases: the phases before, during, and after the interview. In the context
of the preparatory phase the authors recommend to make preliminarily
contact with the interview partner by a letter, which already informs about
actual questions. This way it becomes implicitly obvious that in the con-
text of telephone interviews as described by them they must have certain
kinds of questions and a certain type of interview partners in mind: “We
found it useful to communicate our interview questions ahead of time to
participants, along with a general introductory letter about our study. This
would be especially relevant if you are researching a topic that is abstract,
such as intuitive decision-making. Participants need time to reflect and
think about their responses, and we found that this padding of time ultim-
ately yielded more thick, rich descriptive data from participants” (Burke
and Miller, 2001, p. 7).
Practically, the authors derived their methodical recommendations from
interviews with managers, which were focused on the latter’s decision-making
behaviour. Thus, the research interest was in the professional way of acting of
people who may be considered experts. This means that Burke’s and Miller’s
methodical contribution – just as also Busse’s – actually deal with expert inter-
views on the telephone, without the authors explicitly saying so.
As already indicated, in the context of the “pre-interview phase” Burke and
Miller (2001, p. 5) deal in detail most of all with questions of making con-
tact as well as with technological possibilities of audio recordings. During
the phase of conducting the interview the interviewer’s way of talking is in
the fore. Methodical recommendations are similar to relevant instructions
for qualitative interviews. It is explained in detail how interview partners
can be motivated to go on with talking by help of signals from the recipi-
ent. It is then surprising in this context that Burke and Miller suggest a
strongly structured interview guideline, which apart from open questions
also provides closed ones. The answers to open questions are said to have
the function of providing an explanation horizon for the answering behav-
iour with closed questions: “Ensure you have a mix of open-ended and
close-ended questions. It is helpful to have some questions where people
respond, for example, in a specific Likert scale fashion (that is, close-ended
166 Gabriela B. Christmann

response options), so that you have some easy-to-score data. The open-ended
questions will then provide you with the rich filler to elaborate upon such
responses” (Burke and Miller, 2001, p. 21).
Recommending a semi-standardized way of proceeding is similar to Busse,
the latter – as already explained – offering this as a solution for difficult
interaction situations during a telephone interview, whereas Burke and Miller
do not give any reason for their way of proceeding.9
In respect of the duration of the telephone interview, the authors talk
about their experience that during the pre-test many interview partners
considered a duration of 15 to 20 minutes too long. Thus, in the context
of the actual field phase individual durations of interviews were agreed on
with the respective interview partners. According to the interviewees’ time
budget and the study’s maximum need of information, the interview guide-
line was individually composed.
Ideas for the “post-interview phase” are comparatively short. They are
restricted to mentioning mainly content-analytical or categorizing ways of
proceeding.
Both in Busse’s and in Burke’s and Miller’s contributions it is conspicu-
ous that there is no systematic comparison of telephone and face-to-face
interviews. This is done by Opdenakker (2006) who compares qualitative
face-to-face interviews and technology-mediated ones, while for the latter
category he takes telephone, MSN messenger, and e-mail interviews into
consideration. The comparison is done very fundamentally in respect of the
chronological and spatial (a-) synchronicity of interaction situations (see
Opdenakker, 2006, p. 4). According to this, the telephone interview has in
common with the MSN messenger and also with the face-to-face interview
that the interaction happens in a chronologically synchronous way, in con-
trast to the chronologically asynchronous e-mail interview. On the other
hand, in respect of space the telephone interview is characterized by an asyn-
chronous interaction situation which by the way is also true for the MSN
messenger and the e-mail interview, in contrast to the spatially synchron-
ous face-to-face interview. Opdenakker particularly discusses the advantage
connected to spatial synchronicity in respect of method: spatially co-present
interaction partners are able to perceive social signals (“social cues”) in the
form of non-verbal elements. The author assumes that the importance of
non-verbal elements of interviews varies according to the kind of questions
and the target group of a study: “Due to this synchronous communication,
as no other interview method FtF [face-to-face; G.C.] interviews can take its
advantage of social cues. Social cues, such as voice, intonation, body lan-
guage and so on of the interviewee can give the interviewer a lot of extra
information that can be added to the verbal answer of the interviewee on a
question. Of course the value of social cues also depends on what the inter-
viewer wants to know from the interviewee. If the interviewee is seen as a
subject, and as an irreplaceable person, from whom the interviewer wants
 Expert Interviews on the Telephone 167

to know the attitude towards for example the labour union, then social
cues are very important. When the interviewer interviews an expert about
things or persons that have nothing to do with the expert as a subject, then
social cues become less important” (Opdenakker, 2006, p. 7).
It is remarkable that here Opdenakker also refers to expert interviews, in
the context of which he estimates that for the target group of experts the
importance of “social cues” is less than with others. This would mean that
sheer “voice-to-voice” communication during qualitative telephone inter-
views (Ball, 1968, p. 61) may at best be considered unproblematic if it is an
expert interview.
However, possibly Opdenakker underestimates that “social cues” alone
are of great importance for smoothly maintaining the atmosphere of the
interview.10 Conversation-analytical studies on telephone calls in private
and institutional contexts, which for the time being have focused on open-
ing and terminating a conversation (on this see the “classics”: Schegloff and
Sacks, 1973, Schegloff, 1977), showed that telephone interaction – due to
lacking visual perception – means much more coordination efforts for the
interaction partners even if it is only about identifying each other at the
beginning or about finishing the conversation.11 In the latter case the con-
versation partners must carefully prepare each other for the receiver being
replaced. Finishing the conversation too abruptly is not a good option, as
according to cultural conventions on maintaining social contacts it would
result in considerable irritation (see also Höflich, 1989, p. 205f). The assump-
tion suggests itself that even during the phase between starting and finish-
ing the conversation the lack of non-verbal signals may result in increased
coordination efforts and may also result in the organization of the conver-
sation being disturbed, even more as this phase is much less ritualized than
those of starting and finishing.
This assumption is contrasted by results of psychological laboratory
experiments showing that compared to direct conversation situations
telephone-mediated conversations do not prove to be more considerably
affected in respect of organizing a conversation (see for example Butterworth,
Hine and Brady, 1977, Cook and Lalljee, 1972). However, it must at once be
added that for these laboratory examinations interaction situations from
daily life and work contexts were simulated. In these contexts it is mostly
no problem if there happens the – wide-spread – short-time overlapping of
speech and negotiations on the right to speak. This is much more problem-
atic in situations of scientific interviews, particularly in contexts of quali-
tative research where a different level of conversational skill is demanded
than in everyday life. The interviewer tries to act according to the method-
ical imperative of the “minimum-invasive” way of proceeding. He or she is
asked not to interrupt his or her opposite. Most of all he or she should avoid
disturbing the interview partner’s thoughts during pauses for thought.
Thus, to be able to correctly judge an interaction situation the interviewer
168 Gabriela B. Christmann

depends highly on non-verbal signals (on this see also Chapter 4 as well as
Jordan, Marcus and Reeder, 1980, p. 217). Thus the following is valid: “the
entire stock of non-linguistic signs of gestures, facial expressions, posture is
not an unnecessary concomitant of speaking which might be blended out
(...) but is constitutive for producing unambiguity for an act of speaking”
(Bülow, 1990, p. 307, own translation).
Now it is remarkable that in methods-related literature on standard-
ized interviews the possibilities and limitations of telephone interviews
are discussed much more critically than in literature on qualitative
interviews.12 One is highly aware of the fact that the significance of non-
verbal elements may not be underestimated (on others see most of all
Friedrichs, 1990).
It is considered to be proven that telephone interviews are only suitable
for interviews with simple question patterns and guidelines for answers
(see Anders, 1990, Friedrichs, 1990, Hillmann, 2007, p. 890, Hippler and
Schwarz, 1990, Schnell, Hill and Esser, 1999, p. 345). Against this back-
ground Noelle-Neumann and Petersen (2000, p. 184) state the fear that after
all the wide-spread telephone surveys will result in methodical “impover-
ishment.” The debate is contradictious in respect of the question of in how
far the standardized telephone interview implies lower or higher quota of
refusals. Hillmann (2007, p. 890) assumes that telephone interviews are
regarded to be more anonymous and less personally embarrassing, from
which he derives a lower quota of refusals. Noelle-Neumann and Petersen
(2000, p. 193), however, point to the telephone interview being regarded “as
less interesting and a bigger disturbance and more stress” in comparison
to face-to-face interviews. This way they explain the then evidently higher
quota of refusals of telephone interviews. The assumption that telephone
interviews should be rather short (see for example Hillmann, 2007, p. 890) is
also debated. Frey, Kunz and Lüschen (1990, p. 22) point out to the fact that
longer telephone interviews may definitely be successful (on this see also
Schnell, Hill and Esser, 1999, p. 351, Cockerham, Kunz and Lüschen, 1990,
p. 405, own translation): “Also, in case of a short interview the interviewee
will rather have the impression that in case of only a few questions the
project cannot be respectable. Experience shows that in case of appropriate
preparation, previous information, and a good and interesting course of
the interview which is closely connected to the topic of the project tele-
phone interviews of 30 minutes or even more are possible.”
Interesting is the result that interviewees perceive (standardized) inter-
views as “less interesting” and short interviews as “not respectable.” It is
an open question in how far these results can be transferred to qualitative-
oriented expert interviews on the telephone. At least it cannot be completely
ruled out that these observations might also – and particularly – be true for
experts, even more as this target group (as we may assume) is provided with
a particularly high level of reflexivity.
 Expert Interviews on the Telephone 169

7.4 Problems previous to and during expert interviews

on the telephone

Bogner and Menz (2005a) state that usually it is easier to win experts over
for an interview than people from an “unfiltered public” because they are
professionally interested in their own field and they are basically more open
towards research. Indeed, in the course of our study we were able to find
that it is easy to win experts over if it is a face-to-face interview, but not if
it is a telephone interview. Whereas experts meant for an interview at the
place were quick to agree and as fast with offering a date for an interview,
potential telephone interview partners were considerably reluctant with
agreeing. This happened independently of which type of academic educa-
tion institution these experts belonged or which function they had. A total
of 12 telephone interviews was intended. However, due to delays – beyond
the time limit – it was only possible to realize eight.
This is surprising in so far as the face-to-face interview was announced
to be 60 minutes long and the telephone interview to be 20 or 30 minutes
long. We assumed that experts, permanently suffering from time pres-
sure, would be much more ready for interviews if these required compar-
ably little time. But this was a misjudgement. Interestingly, also Burke
and Miller (2001, p. 11) talk about the experience that fixing dates for
telephone interviews was a lengthy thing; however in their case things
were somewhat less problematic, as dates were fixed on the telephone:
“We found participants more responsive to setting up interview appoint-
ments via the phone, compared to email. But this process can also become
a scheduling quagmire that takes as much time as conducting the inter-
views themselves.”
We conclude that with experts a shorter interview does not automatically
work as a factor motivating for agreeing to an interview and fixing a date. In
respect of its motivating effect, the attention paid to an expert by the inter-
viewer’s more or less time-consuming journey (across all of Germany) and a
one-hour interview must not be underestimated. In this respect, the request
for a rather short telephone interview is a different signal.
However, also it may be supposed that to play an important role the short
period of 20 to 30 minutes for answering important questions of university
planning is only a minor incentive for the expert, as it offers only little lee-
way for explaining or it forces him or her to give a compressed explanation.
Thus, a rather short interview is not necessarily a relief. Possibly, in such
a situation experts feel the need to previously prepare themselves (more
intensively).
This consideration is supported by the observation that most telephone
interview partners were only ready to agree after they had seen the inter-
view guideline, something which due to lack of time was not always imme-
diately possible. Further problems resulted from some telephone interview
170 Gabriela B. Christmann

partners using the telephone call’s basic flexibility with regard to sched-
uling. Compared to the face-to-face interview, there is no need for the
interviewer to make a journey, appropriately it may be fixed or changed
at short notice. One expert right at the beginning suggested a spontan-
eous date at very short notice – depending on an unforeseen gap in his
schedule. Two interview partners changed a fixed date at short notice when
the interviewer called them at the agreed time. Thus, telephone dates are
less binding than face-to-face dates, as the telephone contact may be made
again without effort at any other time. The flexibility expected from the
interviewer in this context may be supposed to be a challenge for every
research team.
But not only fixing dates for interviews was challenging. Also actually
conducting the telephone interviews turned out to be a difficult matter.
Mitchell (1984, p. 249) rightly points to the fact that due to lacking non-
verbal elements interaction partners must organize the interaction process
exclusively by way of language and voice, due to which conversations on the
telephone require much more attention in respect of what is happening lin-
guistically than immediately personal interaction: “The voice must express
all that there is to say. It must cover both the content of the message and the
necessary nonverbal instructions on how to interpret that content. Thus,
talking by telephone demands a good deal more attention to vocal nuance
than face to face communication does.”
This statement is particularly valid for scientifically motivated tele-
phone interviews (see also Chapter 3). In the following, by the example of
selected transcript segments from three different telephone interviews we
like to show interaction problems happening several times or being typical.
Although the first phenomenon was not as often and frequently found in
the data material as the second one, it could at least be observed in four
cases, due to which it deserves attention.

Transcript segment 113

01 IP1 Also es war ja die Frage danach so wie sich so ((nennt die
02 Hochschule)) positioniert,
03 I Mhm,
04 IP1 ähm (--) gibt’s denk ich schon: zu einem gewissen Antei:l, äh::
05 den Punkt, dass man sagt ähm ‘hh äh (.) wir sind da nicht
06 vergleichbar,
07 I Mhm,
08 IP1 un:d was man nicht gut vergleichbar- vergleichen kann, das ist
09 auch schwierig miteinander in Wettbewerb zu bringen.
10 I Mhm,
11 IP1 Das hat ja auch (.) soweit erstmal eine gewisse Logik.
12 I Ja.
13 (--)
14 IP1 Ähm:: (2,25) ja. ‘hh Ähm: (---) ansonsten, (--) hat die ähm (.)
 Expert Interviews on the Telephone 171

15 ((nennt Hochschule)) sich auch beteiligt an den Eliterun den

16 I Mhm,
17 IP1 äh zu bewerben, also in der ersten Runde und jetzt auch in
18 der nächsten Runde, ’hh und hat (--) ‘hh sich insofern: also
19 I Mhm,
20 IP1 erstmal von der Forschungsschiene her (--) sehr stark (-) damit
21 beschäftigt, wo sind ‘hh unsere Kernfelder, gleich ob das dann
22 Forschungsschwerpunkt oder Forschungscluster heißen soll,
23 I Mhm,
24 IP1 eines von beiden äh:: gibt’s da, hat das also für sich (--)
25 definiert, (1,5) un::d- (3.0) *was war noch die Frage*?
26 I Ähm
27 IP1 ‘hh ach ja (.) um- um:: sich auch in diese Richtung dann zu
28 bewe gen.
29 I Mhm, mhm,
30 IP1 Ähm (--) ich denke dass- (2,25) ‘hh ähm:: (2,0) ansonsten war
31 die- ((nennt Hochschule)) die letzten zwei Jahre und sicherlich
32 auch das nächste halbe oder ganze Jahr (.) ‘hh sehr stark damit
33 beschäftigt, diese Modularisierung der Studiengänge
34 voran zutreiben.
35 I Mhm,
Translation:
(Due to language differences the transcripts only reflect the original
(German) transcript to provide the reader with an idea of what the inter-
views were like.)
IP1 Well, the question was how ((names university)) positions
itself, after all,
I uhm,
IP1 well, (--) I think this exists to a certain degree, ehm there
is the point when you say ‘hh well (.) we can’t be compared to
this,
I uhm,
IP1 and what you can’t really compare is also difficult to see
competing
I uhm,
IP1 after all, this is (.) also logical to a certain degree.
I yes.
(--)
IP1 Ehm, (2.25) yes. ‘hh Ehm (---), also (--) ((names university))
has taken part in the elite rounds
I uhm,
IP1 ehm, to apply, that is during the first round and now also in
the second round, ‘hh and (--)’ hh in so far has under the
research aspect (--) been dealing very much (-) with, where are
172 Gabriela B. Christmann

‘hh our essential fields, no matter if you call it research

focus or research cluster,
I uhm,
IP1 it is one of these, ehm there is- has defined that for (--)
himself, (1.5) and- (3.0) what was the question again?
I Ehm
IP1 ‘hh Oh yes (.) to- to move towards this direction then.
I uhm, uhm,
IP1 well (--) I think that – (2.25) ‘hh ehm (2.0) otherwise during
the last two years and surely also the next half or the whole
year ((names university)) ‘hh was very much occupied with
driving on this modularization of courses.
I uhm,

With segment 1, a number of linguistic irregularities (re-formulations, break-

ings off) and longer breaks are conspicuous (lines 04–08, 14–17, 24–8, 30).
Indeed, linguistic irregularities are not at all unusual with oral language,
particularly if it is a daily life conversational situation. They even appear
with very formal conversational contexts; however usually this happens
to a very low degree. Accordingly, in our case the experts we interviewed
(including the above quoted one) were in the overwhelming majority of
interviews able to realize a rather “written” style. Compared to the remain-
ing data material, the above quoted segment is thus conspicuous. In line 25
it is additionally the case that according to his statement the interview part-
ner has forgotten the question (for a short time). On the whole it becomes
obvious that the expert is not concentrating.
As already mentioned, the situation was similar also with three more tele-
phone interviews. This raises the question if the interview partners were
distracted by a third person, such as the secretary entering the room, and/
or by a short-time activity (signing a document, looking for some informa-
tion among the documents held ready for the interview). In the literature
on methods of qualitative interviews one is aware of the fact that external
disturbances may have effects on the content of the interview. Due to this
there is trying to avoid situations prone to disturbances as far as possible.
However, during a telephone interview it is not possible to make sure that
the expert will exclusively pay attention to the interview (see also Engel and
Behr 2008). In this respect, interviewers have no control over the interview
situation.
But even more serious is the fact that often external disturbances cannot
be identified as such, except the interview partner does so. This means that
conspicuous linguistic irritations in the context of telephone interviews
cannot be conclusively interpreted. It is not possible to judge if irritations
are caused by external factors (as explained above), are due to topical rea-
sons, or are caused by a “sheer” concentration deficit.
 Expert Interviews on the Telephone 173

In our research context one reason for linguistic irritations might be that
on the one hand the interview partners want to be cooperative and want to
provide their opposite with useful information but that on the other hand,
due to reasons of competition, they must at the same time very much con-
sider which information to give without talking about a “trade secret.” With
some interviews there are explicit indications that for reasons of competi-
tion one does not want to further explain a subject.
If, on the other hand, linguistic frictions are “only” due to a temporary
concentration deficit, there is the question of how far this is due to lack
of support by the interviewer by way of non-verbal attention signals. Not
without reason, literature on methods of qualitative face-to-face interviews
emphasize that gestures addressing the interview partner, eye contact, and
nodding – apart from linguistic reception signals – are of essential import-
ance for the course of the interview.
Furthermore non-verbal aspects, particularly in the form of eye contact,
are significant also for the interviewer, and that is when it must be ascer-
tained if the interview partner has finished his answer, that is if should
be a change of speakers and the next question may be asked. Segments 2
and 3 show which insecurities develop if these non-verbal elements are
lacking:

Transcript segment 2
01 IP2 Was ich sehr kritisch- oder was ich m::h gewisserweise
02 beklage, ist äh das Akkreditierungsgeschäft,
03 I Ja,
04 IP2 ‘hh
05 (--)
06 IP2 das ist natürlich ähm: (1,25) sehr auf wändig, und ‘hhh (.)
07 I Mhm,
08 IP2 äh: ich sag mal trotzdem nicht immer ähm (1.0) streng rational,
09 I Mhm, (.) mhm,
10 (1,25)
11 I Ja. (--) ‘hh Ja; neben Bachelor und Masterstudiengängen (...)
Translation:
IP2 What I complain about very critically or ehm to a certain
degree is the ‘hh accreditation business,
I yes,
IP2 ‘hh
(--)
IP2 of course, this needs ehm (1.25) very much effort, and ‘hhh (.)
I uhm,
IP2 well, I’d say nevertheless it is not always ehm (1.0)
strictly rational,
174 Gabriela B. Christmann

I uhm, (.) uhm,

(1.25)
I yes. (--)’hh well, apart from BA and MA courses (...)

The interview partner of segment 2 answers the question about which

experiences were made at his university with the introduction of BA and MA
courses. He was asked to formulate advantages and disadvantages of these new
kinds of courses and, if necessary, to discuss necessities of change. Accordingly,
in the previous sequences he mentioned various aspects, which he considers
positive or negative. For example, he mentions the accreditation procedure,
which is connected to the new courses. After the completion of line 02, men-
tioning the “accreditation business” will probably be completed by giving a
reason, for one may expect an explanation why the expert judges “critically”
on the procedure. In line 03, during a break by the expert the interviewer at
first reacts by a reception signal. After taking breath and another short break
(l. 04/05) the expert starts giving reason for his judgement (l. 06). At the end
of the statement in line 08 on the one hand a thought has been completed,
but on the other hand, given the weakly rising contour of intonation, one
might think that reason-giving will be continued. The interviewer must make
sure if the expert makes a break to further structure his speech or if he con-
siders his contribution to be finished and is waiting for the next question.
According to experience, in face-to-face situations eye contact is helpful with
solving such cases of doubt. During a telephone interview this is not possible.
Carefully, the interviewer reacts by a second “Mhm,” the first signal being
divided from the second one by a micro pause (l. 09). Then there is a break
(l. 10): obviously still the interviewer is not sure about which kind of situation
she is confronted with; in any case she does not dare interfering (possibly too
early). But there is no change: no reaction by the interview partner becomes
obvious. Reluctantly, the interviewer at first says “yes,” followed by a short
break, then, finally, she starts her next question (l. 11).
A similar situation is shown by segment 3 with another interview
partner.

Transcript segment 3
01 IP3 Äh: das andere ist, dass wir äh: temporär Schwankungen hatten,
02 es gab also (--) Einbrüche als wir auf Bachelor Master
03 I Mhm,
04 IP3 umgestellt haben,
05 I Mhm,
06 IP3 äh:: die waren aber temporär, und haben sich jetzt ins
07 Gegenteil umgekehrt, also ‘hh äh (--) äh: ja das- äh das sind
08 I Mhm,
09 IP3 eben auch statistische Pro zesse; (--) n e,
10 I Ja, ja, mhm,
 Expert Interviews on the Telephone 175

11 IP3 ‘h h
12 I Mhm,
13 IP3 Gut.
14 I Gut.
15 (1,0)
16 I Sie wollten noch zur Internationalisier ung etwas sagen?
17 IP3 Ja. Inter- also
18 I ‘hh
19 IP3 Internationali sierung, (...)
20 I Mhm,
Translation:
IP3 Ehm the other aspect is that we ehm had temporary
fluctuations, that is there were sharp falls when we changed to
Bachelor Master,
I uhm,
IP3 ehm but those were temporary, and have now reversed, that is
I uhm,
IP3 ‘hh ehm (--) ehm well, these- ehm these are also statistical
processes; (--) you see,
I yes, yes, uhm,
IP3 ‘hh
I uhm,
IP3 well.
I All right.
(1.0)
I You were also going to say something about
internationalization?
IP3 Yes. Inter- Internationalization, (...)
I uhm

Also here it is about experiences with BA and MA courses. The interview

partner talks about sharp declines in numbers of applicants after the change
to the new courses, but he emphasizes that these had only been temporary
and that meanwhile the process had reversed. In line 09, discussing this
aspect is completed both in respect of thought and intonation. However, it
stays an open question if the expert intends to further explain this aspect
or if he intends to give several other aspects. The interview partner audibly
taking breath in line 11 may be interpreted as indicating his speech to be
continued. In line 12 the interviewer by the reception signal “Mhm” indi-
cates her readiness to listen further. However, the expert by taking breath
has not made himself ready for further explanations but for saying the word
“well” which due to its strongly declining intonation is of a concluding
nature (l. 13). This indicates an end of speech. Against this background the
176 Gabriela B. Christmann

interviewer takes up the word “well” by saying “all right” (l. 14), this way
ratifying the end of speech by the expert. By way of a following break, how-
ever, she leaves her opposite on the telephone the option of beginning to
speak again (l. 15). Then she changes to another topic (l. 16). She reminds
the expert of something which he has marked as a topic which must be
dealt with in his previous statements and thus formulates a conversation-
immanent request to narrate.
The comparative analysis of the change of situations of speech with
face-to-face interviews produced the result that coping with this was much
faster and easier.
On the whole, the interviewer subjectively perceived telephone interviews
much more exhausting than immediate interview situations at the place,
although the former were shorter than face-to-face interviews. Even if with
face-to-face interviews the degree of attention is unquestionably higher, it
is still even higher with telephone interviews, as lacking non-verbal signals
must be compensated by much more work of interpreting what has been
said, and that is in respect of structuring speech both topically and for-
mally. Rightly so, Gläser and Laudel (2006, p. 169) emphasize that by inter-
vening with the “apparent emptiness” of a pause for thought one will miss
“important information.” Thus, almost all the time the interviewer was in
a situation of insecurity. On the one hand she was aware of the danger of
interpreting a break not as an inner-speech break but as an indication of
ending speech and that against this background she might interfere too
early with the interview partner’s thought. On the other hand she felt to be
under the pressure of not allowing too long breaks. Not at all – if there was
a speech vacancy – she wanted the interview partner to have the impression
of lacking attention and to make him feel uncertain. It is also that periods of
silence on the telephone, as shown by Hess-Lüttich (1990, p. 286, own trans-
lation), are risky for another reason: “Being silent together on the telephone
is much more difficult to go on with than with direct conversation because
both sides interpret it as a danger or even an interruption of the contact
(“Hallo,” “Are you still there?”) (...) Thus, activities of securing the contact
are more typical for the type of telephone conversation than for example
activities of securing understanding.”
However, Hess-Lüttich (1990) looks at daily-life telephone calls. But this
thought is also – and particularly – of significance for telephone interviews,
as for methodical reasons any uncertainty of interview partners must be
avoided. This may also explain why with telephone interviews much more
often the interviewer produced reception signals in the form of “yes” and
“uhm” than was the case with direct interviews. On the one hand this was
the only way to indicate attention, on the other hand these signals served
also for communicating to the opposite that the telephone-technical con-
tact was still intact. Also, vice versa strategies by the experts of reducing
breaks became obvious, as conspicuously more often than their colleagues
 Expert Interviews on the Telephone 177

during face-to-face interviews they produced fillers by way of lengthening

and other lengthening of syllables.

7.5 Conclusion

In respect of method, qualitative telephone interviews are a very demand-

ing and thus difficult undertaking. This is not to say that due to this they
are generally useless. It may depend on the kind of questions of a study
whether telephone interviews are useful. In the context of the here pre-
sented “explorative-systematizing” study (see Chapter 2), which was first of
all on finding out about examples of “best practice” in the field of univer-
sity planning, we were able to gain outspokenly useful information by help
of telephone interviews. However, the more strongly expert interviews are
directed at generating theory, the more problematic the telephone method
may be supposed to be. Due to the lack of non-verbal elements, the inter-
action process may be disturbed much more than with face-to-face inter-
views. Furthermore, potential external disturbances (such as the entering of
third parties) are usually not recognized. This means that interviewers are
not able to know if the interview partner’s attention is exclusively directed
at the interview. Thus, telephone interviews generate a kind of data mater-
ial, which for the following interpretation process may leave many ques-
tions open. This may considerably obstruct theory generation.
There would have to be a much more systematic analysis of the question
whether telephone interviews should be rather short – that is not longer
than 20 to 30 minutes – or whether they may be much longer. Statements
by methods-related literature are contradictious. Also our own empirical
experiences do not allow for any clear statement: on the one hand there
were indications that short telephone interviews, due to fewer possibilities
of presentation and stronger necessities to compress, must be considered
rather unattractive and not being relieving per se. This would suggest longer
telephone interviews. On the other hand, probably not only for the inter-
viewer telephone interview situations are particularly “attention-intensive”
and thus exhausting. This might increase in the case of longer interviews.
Thus, this might provoke unwanted interview mistakes and loss of concen-
tration with interview partners.
What concerns the interaction problem of change of speakers during
telephone interviews, this may be reduced by an interview strategy. In
the context of the above presented analysis it could be shown that during
potentially speech-vacant breaks the interviewer (due to her fear of block-
ing the interview partner by her next question) at first reacted carefully by
reception signals and breaks, this way successfully leaving the interview
partner enough freedom for answering the questions. This unconscious
behaviour by the interviewer might be developed to become an interview
strategy which would be: breaks by interview partners must be systematically
178 Gabriela B. Christmann

accompanied by recipient signals and short breaks, and this way one must
wait for further reactions until the situation is clear.

Notes
1. The client was the Saxonian State Ministry of Arts and Sciences, the conducting
institution was the “Zentrum demographischer Wandel (Centre of Demographic
Change)” of the Technical University of Dresden, heads of the project were Karl
Lenz and Winfried Killisch.
2. Following Meuser and Nagel, by an expert we understand most of all some-
body who “influences decisions and problem-solving beyond the routines of
decision-making” (see also Bogner and Menz 2005a, Pfadenhauer 2005, see also
Pfadenhauer, in this volume, and Köhler, 1992, p. 319f.). Meuser and Nagel con-
sider expert interviews to be particularly interesting as “they inform about those
action concepts and knowledge stocks as controlling, driving, and retarding proc-
esses of social change and the modernization of society.” (Meuser and Nagel,
2005b) See also the most extended volume for the time being on “Expertise and
Expert Performance” by Ericsson and others (2006), and see the volume “Eliten
am Telefon” (Elites on the Telephone) by Martens and Ritter (2008).
3. Every state and Church university and college including art and music colleges as
well as all vocational academies in the Federal Republic of Germany were written to.
4. Twelve interviews were conducted at universities, nine at colleges, and one at a
vocational academy. Ten experts were from the so called “old” federal states, 12
from the “new ones.” Thirteen federal states were represented, some of them sev-
eral times, particularly Saxony, as the emphasis of the overall project was there.
5. Experts were concretely asked about their experiences for the time being with BA
and MA courses and how they were judging on them in respect of their future
attractivity. It was also asked which particular kinds of courses are and will be
important for the respective university and for what reasons, and which target
groups institutions of academic education try to win over as university beginners.
Not least there was the question of which developments universities and colleges
expect for the Federal Republic’s university system: if they assume the university
system to be de-differentiated (in respect of the three traditional “columns” of
university, college, vocational academy), how they judge on future competitive
behaviour among universities, and if in respect of relationships between univer-
sities they rather count on competition or on co-operation.
6. Also Bogner and Menz (2005a) are of this opinion when writing that “often” the
expert “is used to acting in a publicity-effective way and close to the public.”
7. On choosing experts see also Deeke (1995, p. 17) and Meuser and Nagel (2005a).
8. The status of co-expert was supported in so far as the interviewer, being a
“Privatdozent” (having successfully completed a habilitation thesis but has not
been offered a chair yet), was close to the expert at least in respect of the academic
degree.
9. The semi-standardized method not even connected by the authors to their ori-
ginal target group – that of managers – who, as made clear by Trinczek (2005, see
also Trinczek, in this volume), in the context of expert interviews do not expect
leeway for extended narrations but a clearly structured guideline and a rapid game
of questions and answers. Giving such a reason would have been plausible in so
far as the communication process during an expert interview should possibly be
“adjusted to the interviewee’s cultural context” (Gläser and Laudel 2006, p. 110).
 Expert Interviews on the Telephone 179

10. According to Reid (1977, p. 397), “social cues” have the following functions for a
conversation: “1. Mutual attention and responsiveness (to provide evidence that
the other person is attending). 2. Channel control (to indicate the way partici-
pants should take turns in speaking and listening). 3. Interpersonal attitudes (to
indicate attitudes and intentions). 4. Illustrations (to accompany and illustrate
what is being said – for example by gesture). 5. Feedback (to indicate whether
the other person understands, believes, or disbelieves, is surprised, agrees or dis-
agrees, is pleased or annoyed).”
11. See also the numerous linguistic works on telephone and language in situations
of daily life conversations, such as by Bülow (1990), Hahn (1990), Hess-Lüttich
(1990), Wiegmann (1990).
12. On this see Noelle-Neumann and Petersen (2000, p. 183) who write: The
seemingly unstoppable increase of telephone surveys since the end of the
1970s is accompanied by the continuously rising number of books and art-
icles in expert magazines where the advantages and disadvantages of tele-
phone interviews are compared to those of oral-personal “face to face”
interviews. See for example the works by Anders (1990), Brückner and others
(1982), Buschmann (2001), Cockerham and others (1990), Friedrichs (1990),
Frey and others (1990), Henkel (2001), Mc Cormick and others 1993, Noack
(2003), Noelle-Neumann and Petersen (2000), Schenk (1990), Schnell and
others (1999), Waleczek (2003). However, in the latest volume about tele-
phone interviewing “elites” (edited by Martens and Ritter 2008) most of the
articles only deal with the techniques of interviewing experts in the con-
text of telephone surveys. It seems as if critical aspects fade into the back-
ground.
13. Transcript conventions: IP = interview partner, I = interviewer; [ = beginning
of an overlapping in case of speaking at the same time; (.) = short stop; (-) (--)
(---) = breaks shorter than 1 sec., each (-) representing about 0.25 sec.; (1.5) =
breaks being 1 sec. long or longer; *yes* = speaking in a low voice; ‘hh = audible
breathing in; punctuation marks, and ? symbolize a weakly or strongly rising
contour of intonation; punctuation marks ; and . symbolize a weakly or strongly
falling contour of intonation; yes::: = lenghtening; the number of colons some-
what symbolizes the lengthening, each : symbolizing about 0.25 sec.; mayb- =
unfinished statement; (( )) = remarks by the one making the transcript; (...) =
omission in the transcript.

Further readings
Burke, L. A. and Miller, M. K. (2001) “Phone Interviewing as a Means of Data
Collection: Lessons Learned and Practical Recommendations” in Forum Qualitative
Sozialforschung / Forum: Qualitative Social Research 2, available at: http://qualitative-
research.net/fqs/fqs-eng.htm, date accessed 11 June 2006.
Ericsson, K. A., Charness, N., Feltovich, P. J. and Hoffman, R. R. (eds) (2006) The
Cambridge Handbook of Expertise and Expert Performance (Cambridge: Cambridge
University Press).
McCormick, M. C., Workman-Daniels, K., Brooks-Gunn, J. and Peckham, G. J. (1993)
“When You’re Only a Phone Call Away. A Comparison of the Information in
Telephone and Face-to-Face Interviews” in Journal of Developmental and Behavioral
Pediatrics 14, 250–55.
180 Gabriela B. Christmann

References
Anders, M. (1990) “Praxis der Telefonberatung” in Forschungsgruppe Telefon-
kommunikation (ed.), pp. 426–36.
Ball, D. W. (1968) “Toward a Sociology of Telephones and Telephoners” in Truzzi, M.
(ed.) Sociology and Everyday Life (Englewood Cliffs, NJ: Prentice-Hall), pp. 59–75.
Bogner, A. and Menz, W. (2005a) “Expertenwissen und Forschungspraxis: Die
modernisierungstheoretische und die methodische Debatte um die Experten.
Zur Einführung in ein unübersichtliches Problemfeld” in Bogner, A., Littig, B.
and Menz, W. (eds) Das Experteninterview – Theorie, Methode, Anwendung, 2nd edn
(Wiesbaden: Verlag für Sozialwissenschaften), pp. 7–30.
Bogner, A. and Menz, W. (2005b) “Das theoriegenerierende Experteninterview.
Erkenntnisinteresse, Wissensformen, Interaktion” in Bogner, A., Littig, B. and
Menz, W. (eds) Das Experteninterview – Theorie, Methode, Anwendung, 2nd edn
(Wiesbaden: Verlag für Sozialwissenschaften), pp. 33–70.
Brinkmann, C., Deeke, A. and Völkel, B. (eds) (1995) Experteninterviews in der
Arbeitsmarktforschung. Diskussionsbeiträge zu methodischen Fragen und praktischen
Erfahrungen (Beiträge zur Arbeitsmarkt- und Berufsforschung: BeitrAB 191)
(Nürnberg: Institut für Arbeitsmarkt- und Berufsforschung der Bundesanstalt für
Arbeit).
Brody, H., Geronemus, J., Roy, G. and Farris, P. K. (2003) “Beauty Versus Medicine:
The Nonphysician Practice of Dermatologic Surgery” in Dermatologic Surgery 29,
319–24.
Brückner, E., Hormuth, St. W. and Sagawe, H. (1982) “Telefoninterviews – Ein
alternatives Erhebungsverfahren? Ergebnisse einer Pilotstudie” in ZUMA
Nachrichten 11, 9–36.
Bülow, E. (1990) “Sprechakt und Textsorte in der Telefonkommunikation” in
Forschungsgruppe Telefonkommunikation (ed.), pp. 300–12.
Burke, L. A. and Miller, M. K. (2001) “Phone Interviewing as a Means of Data
Collection: Lessons Learned and Practical Recommendations” in Forum Qualitative
Sozialforschung / Forum: Qualitative Social Research 2, available at: http://qualitative-
research.net/fqs/fqs-eng.htm, date accessed 11 June 2006.
Buschmann, N. (2001) Auswirkungen von Befragungsmodi. Andere Antworten bei
Telefoninterviews als bei Face-to-face-Umfragen und Fragebogenerhebungen (München:
Grin-Verlag).
Busse, G. (2003) “Leitfadengestützte, qualitative Telefoninterviews” in Katenkamp, O.,
Kopp, R. and Schröder, A. (eds) Praxishandbuch Empirische Sozialforschung (Münster
u.a.: LIT), pp. 27–33.
Butterworth, B., Hine, R. R. and Bradyleen, K. D. (1977) “Speech and Interaction in
Sound-Only Communication Channels” in Semiotica 20, 81–99.
Cockerham, W. C., Kunz, G. and Lüschen, G. (1990) “Sozialforschung per Telefon:
BRD und USA im Vergleich” in Forschungsgruppe Telefonkommunikation (ed.),
pp. 400–12.
Cook, M. and Lalljee, M. G. (1972) “Verbal Substitutes for Visual Signals in Interaction”
in Semiotica 6, 212–21.
Deeke, A. (1995) “Experteninterviews – ein methodologisches und forschungsprak-
tisches Problem” in Brinkmann, C., Deeke, A. and Völkel, B. (eds) Experteninterviews
in der Arbeitsmarktforschung. Diskussionsbeiträge zu methodischen Fragen und prak-
tischen Erfahrungen. Beiträge zur Arbeitsmarkt- und Berufsforschung Nr. 191 (Nürnberg:
Bundesanstalt für Arbeit), pp. 7–22.
 Expert Interviews on the Telephone 181

Engel, T. (2008) “Das eine tun, ohne das Andere zu lassen – Experteninterviews
mit Managern am Telefon. Nutzen, Anforderungen, Praxis” in Martens, B. and
Ritter, T. (eds) Eliten am Telefon: neue Formen von Experteninterviews in der Praxis
(Baden-Baden: Nomos), pp. 75–94.
Ericsson, K. A., Charness, N., Feltovich, P. J. and Hoffman, R. R. (eds) (2006) The
Cambridge Handbook of Expertise and Expert Performance (Cambridge: Cambridge
University Press).
Forschungsgruppe Telefonkommunikation (ed.) Telefon und Gesellschaft. Band 2:
Internationaler Vergleich – Sprache und Telefon – Telefonseelsorge und Beratungsdienste –
Telefoninterviews (Berlin: Spiess).
Frey, J. H., Kunz, G. and Lüschen, G. (1990) Telefonumfragen in der Sozialforschung.
Methoden, Techniken, Befragungspraxis (Opladen: Westdeutscher Verlag).
Friedrichs, J. (1990) “Gesprächsführung im telefonischen Interview” in
Forschungsgruppe Telefonkommunikation (ed.), pp. 413–25.
Galster, G. C. (2006) “What’s the Hood Got to Do With It? Parental Perceptions
about How Neighborhood Mechanisms Affect Their Children” in Journal of Urban
Affairs 28, 201–26.
Gläser, J. and Laudel, G. (2006) Experteninterviews und qualitative Inhaltsanalyse als
Instrumente rekonstruierender Untersuchungen (Wiesbaden: VS).
Hahn, W. von (1990) “Telefon und Sprache. Einleitende Zusammenfassung” in
Forschungsgruppe Telefonkommunikation (ed.), pp. 275–80.
Henkel, J. (2001) Das Telefoninterview (München: Grin-Verlag).
Hess-Lüttich, E. W. B. (1990) “Das Telefonat als Mediengesprächstyp” in
Forschungsgruppe Telefonkommunikation (ed.), pp. 281–99.
Hillmann, K. -H. (2007) Wörterbuch der Soziologie (Stuttgart: Kröner), p. 890.
Hitzler, R., Honer, A. and Maeder, C. (eds) Expertenwissen. Die institutionalisierte
Kompetenz zur Konstruktion von Wirklichkeit (Opladen: Westdeutscher Verlag).
Höflich, J. R. (1989) “Telefon und interpersonale Kommunikation – Vermittelte
Kommunikation aus einer regelorientierten Perspektive” in Forschungsgruppe
Telefonkommunikation (ed.), pp. 197–220.
Jordan, L. A., Marcus, A. C. and Reeder, L. G. (1980) “Response Styles in Telephone and
Household Interviewing: A Field Experiment” in Public Opinion Quarterly 44, 210–22.
Köhler, G. (1992) “Methodik und Problematik einer mehrstufigen Expertenbefragung”
in Hoffmeyer-Zlotnik, J. H. P. (ed.) Analyse verbaler Daten. Über den Umgang mit
qualitativen Daten (Opladen: Westdeutscher Verlag), pp. 318–32.
Martens, B. and Ritter, T. (eds) (2008) Eliten am Telefon. Neue Formen von
Experteninterviews in der Praxis (Baden-Baden: Nomos).
Mayring, P. (2000) Qualitative Inhaltsanalyse. Grundlagen und Techniken (Weinheim:
Deutscher Studienverlag).
McCormick, M. C., Workman-Daniels, K., Brooks-Gunn, J. and Peckham, G. J. (1993)
“When You’re Only a Phone Call Away. A Comparison of the Information in
Telephone and Face-to-Face Interviews” in Journal of Developmental and Behavioral
Pediatrics 14, 250–55.
Meuser, M. and Nagel, U. (1994) “Expertenwissen und Experteninterview” in Hitzler, R.,
Honer, A. and Maeder, C. (eds) Expertenwissen. Die institutionalisierte Kompetenz zur
Konstruktion von Wirklichkeit (Opladen: Westdeutscher Verlag), pp. 180–92.
Meuser, M. and Nagel, U. (2005a) “ExpertInneninterviews – vielfach erprobt, wenig
bedacht. Ein Beitrag zur qualitativen Methodendiskussion” in Bogner, A., Littig, B.
and Menz, W. (eds) Das Experteninterview – Theorie, Methode, Anwendung, 2nd edn
(Wiesbaden: Verlag für Sozialwissenschaften), pp. 71–93.
182 Gabriela B. Christmann

Meuser, M. and Nagel, U. (2005b) “Vom Nutzen der Expertise. ExpertInneninterviews

in der Sozialberichterstattung” in Bogner, A., Littig, B. and Menz, W. (eds) Das
Experteninterview – Theorie, Methode, Anwendung, 2nd edn (Wiesbaden: Verlag für
Sozialwissenschaften), pp. 7–30.
Mieg, H. A. and Näf, M. (2006) Experteninterviews in den Umwelt- und
Planungswissenschaften. Eine Einführung und Anleitung (Lengerich and others:
Pabst).
Mitchell, G. (1984) “Some Aspects of Telephone Socialization” in Thomas, S. (ed.)
Studies in Communication (Norwood NJ: Ablex), pp. 249–52.
Mitchell, V. L. and Zmud, R. W. (1999) “The Effects of Coupling IT and Work Process
Strategies in Redesign Projects” in Organization Science 10, 424–38.
Noack, G. (2003) Online-Befragung und Telefoninterview (München: Grin-Verlag).
Noelle-Neumann, E. and Petersen, T. (2000) “Das halbe Instrument, die
halbe Reaktion. Zum Vergleich von Telefon- und Face-to-face Umfragen” in
Hüfken, V. (ed.) Methoden in Telefonumfragen (Wiesbaden: Westdeutscher Verlag),
pp. 183–200.
Opdenakker, R. (2006) “Advantages and Disadvantages of Four Interview Techniques
in Qualitative Research” in Forum Qualitative Sozialforschung / Forum: Qualitative
Social Research 7, available at: http://qualitative-research.net/fqs-texte/4-06-4-11-e.
htm, date accessed 14 March 2007.
Pfadenhauer, M. (2005) “Auf gleicher Augenhöhe reden. Das Experteninterview –
ein Gespräch zwischen Experte und Quasi-Experte” in Bogner, A., Littig, B. and
Menz, W. (eds) Das Experteninterview – Theorie, Methode, Anwendung, 2nd edn
(Wiesbaden: Verlag für Sozialwissenschaften), pp. 113–30.
Reid, A. A. L. (1977) “Comparing Telephone with Face-to-Face Contact” in Sola
Pool, I. de (ed.), pp. 386–414.
Schegloff, E. A. (1977) “Identification and Recognition in Interactional Openings” in
Sola Pool, I. de (ed.), pp. 415–50.
Schegloff, E. A. and Sacks, H. (1973) “Opening up Closings” in Semiotica 8, 289–327.
Schenk, M. (1990) “Das Telefon als Instrument der Sozialforschung” in
Forschungsgruppe Telefonkommunikation (ed.), pp. 379–85.
Schnell, R., Hill, P. B. and Esser, E. (1999) Methoden der empirischen Sozialforschung
(München/Wien: Oldenbourg).
Sola Pool, I. de (ed.) The Social Impact of the Telephone (Cambridge u.a.: The MIT
Press).
Trinczek, R. (2005) “Wie befrage ich Manager? Methodische und methodolo-
gische Aspekte des Experteninterviews als qualitative Methoden empirischer
Sozialforschung” in Bogner, A., Littig, B. and Menz, W. (eds) Das Experteninterview –
Theorie, Methode, Anwendung, 2nd edn (Wiesbaden: Verlag für Sozialwissenschaften),
pp. 209–22.
Vogel, B. (1995) “ ‘Wenn der Eisberg zu schmelzen beginnt ...’ – Einige Reflexionen
über den Stellenwert und die Probleme des Experteninterviews in der Praxis der
empirischen Sozialforschung” in Brinkmann, C., Deeke, A. and Völkel, B. (eds)
Experteninterviews in der Arbeitsmarktforschung. Diskussionsbeiträge zu methodischen
Fragen und praktischen Erfahrungen. Beiträge zur Arbeitsmarkt- und Berufsforschung
Nr. 191 (Nürnberg: Bundesanstalt für Arbeit).
Waleczek, T. (2003) Das Telefoninterview in der empirischen Sozialforschung (München:
Grin-Verlag).
Walter, W. (1994) “Strategien der Politikberatung. Die Interpretation der
Sachverständigen-Rolle im Lichte von Experteninterviews” in Hitzler, R., Honer, A.
 Expert Interviews on the Telephone 183

and Maeder, C. (eds) Expertenwissen. Die institutionalisierte Kompetenz zur Konstruktion

von Wirklichkeit (Opladen: Westdeutscher Verlag), pp. 268–84.
Wiegmann, H. (1990) “Zur Rhetorik telefonischer Kommunikation” in Forschungsgruppe
Telefonkommunikation (ed.), pp. 313–18.
Williams, M. L. (1993) “Measuring Business Starts, Success and Survival. Some
Database Considerations” in Journal of Business Venturing 8, 295–300.
8
Expert versus Researcher: Ethical
Considerations in the Process of
Bargaining a Study
Vaida Obelenė

8.1 Introduction

This chapter explores the dilemmas faced when interviewing experts by

reflecting my own experiences in relation to the ideals of ethical research.
In general, ethical issues encountered when doing a study are multiple; they
arrive prior to, during and after a study (for example Mauthner and others,
2002, Homan, 1991, Kimmel, 1988, Kvale, 1996, Ali and Kelly, 2004). Ethical
dilemmas are also acknowledged by researchers who work on powerful and
knowledgeable research subjects (for example Odendahl and Shaw, 2001,
Hertz and Imber, 1995, Moyser and Wagstaffe, 1987, Dexter, 1970). Against
this background, it is the purpose of this chapter to reflect the extent to
which the propositions of the democratic research literature – that is:
research that argues in favour of assigning equal rights to research subjects
by turning them into co-researchers and, accordingly, engaging them at all
levels of a study process – are relevant to an expert researcher. Furthermore,
I seek to reflect the dilemma of the researcher who has a commitment to
protect the study from harm and simultaneously owes a duty to protect the
research subject from harm.
The argument is structured in the following way. As a first step I elabor-
ate the understanding of the expert as a research subject, and review prop-
ositions derived from the literature on democratic and ethical research. As
a second step, I discuss whether the democratic practices are at all appropri-
ate for an expert researcher that simultaneously has to protect the research
subject and the study from harm. The expert researcher has to manage rela-
tionships with powerful research subjects, and simultaneously he or she
has to find a position of control in order to secure the purpose of schol-
arly investigation. On the basis of my own experiences of bargaining with
experts I conclude that these practices do not necessarily mean more ethics
in research, and that the participatory ideals and practices, however noble

184
 Expert versus Researcher: Ethical Considerations 185

they are, may not be transferable to and even inappropriate for research
based on expert interviewing. The question remains whether the researcher
is able to accomplish such a twofold task of protecting the study and the
expert, and how? The third part addresses this issue by discussing how the
powerful experts are themselves vulnerable to impacts a study may have.
Two issues can be distinguished. On the one hand, there is the question of
how researchers may exploit information on strengths and weaknesses of
an expert to their advantage when assuming a strong bargaining position
and when motivating experts to engage with a study in a role assigned to
them by the researcher. On the other hand, there is the question of how
the researcher, who is committed to minimizing harm to a subject in a
study context, may use such information on strengths and weaknesses of an
expert in following the ideals of ethical research.

8.2 The expert, democratic and ethical research

The question is, in which way the interests of researcher and the expert
should be balanced in the context of a study? Just as researchers have a
duty of protecting their subjects from harm, so researchers have the duty of
protecting themselves from harm; the tension which may turn to be resolv-
able only as “either one or the other” choice. Needless to say, a study poten-
tially is linked to a variety of harms. Harm reduction targets to minimize
negative consequences of participation in a study for a research subject.
But most importantly it is the idea of benefit that facilitates their decision
to take part in a study. It thus becomes important to acknowledge that the
problem of harm is not resolved by minimizing the negative influence of
the study: harm also occurs if the subjects do not benefit from the study in
the way they anticipated. On the other hand, ethical stance is about secur-
ing the interests of scholarly investigation. Therefore, as another tension
point in this discussion I want to consider the interests of the researcher
that represent academic freedom to study phenomena and report findings.
To sum up: the researcher owes a duty to ensure the purpose of the schol-
arly investigation and a simultaneous duty to protect the research subject
from harm.
To do a study, among other things, means to embark on a process of nego-
tiating and constructing the researcher’s own position and the position to be
assumed by the research subject at all levels of a study. The literature about
managing relations with respondents suggests many professional tricks to
avoid causing harm to the respondent and to achieve rapport and disclosure
in an interview situation. Furthermore, there is advice on the need to foster
and maintain the relationship with subjects or, in this case, co-researchers
in a study beyond the interview situation. For example, the respondents
ought to have the right to access their data and be given a chance to fix the
transcripts; they should be provided a chance to look into reports prior to
186 Vaida Obelenė

their dissemination and once more invited to give their “informed consent”
to proceed. Respondents should even be offered opportunity to withdraw
from a study at any point. In the meantime, the practices of doing research
such as those conceived by the term “communicative validity” (Kvale, 1996)
invite the researcher to engage in a process of validating the interpretation
with the interviewees. All in all, the democratic research literature may
suggest that such democratic practices facilitate research tactically, meth-
odologically and are a requirement for ethical research (Hammersley, 1995,
pp. 48–9). In the meantime, the literature specialized in knowledgeable and
powerful respondents indicates that expert researchers may find themselves
in a situation to pursue such good practices due to the nature of their sub-
ject. For example, the interviewees may claim their rights in a study as far as
censoring the writings. Altogether, the survey of this literature and its argu-
ments may be overwhelming and disempowering rather than instructive for
arranging the practical aspects of research. For instance, by following the
emphasis on the power of the subject and the necessity of permanent nego-
tiation one may lose sight of defending the original research purpose.
It is true, however, that the literature warns the researcher on particu-
larities of studying the powerful and knowledgeable. For example, femin-
ist researchers, altogether proponents of democratic research practices, to
refer to the hierarchical order in research arrangements acknowledged the
difference between “studying up” and “studying down” (Harding, 1987,
p. 8). There are situations, specifically they may occur when “studying up,”
that “imply important qualification to the feminist commitment to non-
hierarchical research techniques” (Hammersley, 1995, p. 56). Hammersley
furthermore argues that these situations may require from researchers to
“exploit whatever resources they have to exert control over the relationship,
on the grounds that in present circumstances the only choice is between
being dominant or being dominated” (1995, p. 56). In a similar way, Luff by
reflecting her own experience of “studying up” suggests that research prac-
tices developed in “studying down” “may be not transferable, indeed may be
counter-productive” (1999, p. 692) when “studying up.”
This chapter, for the most part, reiterates those themes and concerns
expressed by researchers with regard to studying powerful people. Yet it is
important to remind oneself of the particularity of the experts as research
subjects, including their potentially dominating stand in the interaction
with the researcher. Experts are knowledgeable and with power capacity
(Bogner, 2005, see also the contributions of Bogner, Menz and Littig, in this
volume). The expert researcher will encounter highly literate and knowledge-
able people capable of offering their interpretation of reality, who, however,
are also capable of exerting influence on a study. The question is, should
the researcher and the expert be assigned equal power and at all levels of
the research process? What happens when the conflict of interests ensues?
It is common to regard a study in the perspective of power dynamics: such
 Expert versus Researcher: Ethical Considerations 187

images of a study as a battlefield may become highly relevant for a study

based on expert interviews. For example, as much as knowledge is perspec-
tival and value implicated, there may be no means to resolve the interpret-
ative conflicts. In this context, the question to ask is, how the choices the
researcher makes in managing the relationships with the expert serve to
cause such conflicts? While ethical stance, as described above, means a pro-
cess of balancing study decisions in such a way that the harm to the sub-
ject and the researcher is minimized, which is a difficult but nevertheless
worthwhile ideal to achieve, based on my own experiences in the following
section I seek to demonstrate that the democratic research practices may be
counterproductive and do not immediately mean that the research is more
ethical.
This chapter is based on a reflection of my experience of interviewing
people who were communist functionaries 15–19 years ago. In the mean-
time, they moved ahead to establish themselves in relation to new forms of
knowledge and power. It is true, that the reasons to propose such a study
as an instance of expert interviewing are not immediately obvious. While
many people I studied may be described as those who may “influence the
freedom of others to act” (Littig, in this volume), these capabilities ensue
not immediately from the know-how I studied. This, however, merely draws
attention to the constructed nature of the expert’s position: the expert
researchers would grant the experts’ status in relation to the present day
roles in society to the same people I was studying. The underlying purpose of
the study was to contribute to the understanding of the communist recruit-
ment system but also to explore the construction of the devalued condition
of a communist functionary. In other words, the study was concerned with
a sensitive and devalued topic; accepting the expert’s status in such a study
meant that individuals admit belonging to a blameworthy group in society.
This implicated that the study had to overcome their reluctance to recog-
nize themselves as experts. Besides, it had to expect that the experts would
claim the control position in a study in order to protect themselves from
harm in a context of such a topic.

8.3 The process of bargaining in practice

While the literature may propose the democratic research practices as a

means to avoid ethical dilemmas, and possibly may even argue that the
ethical research consists of such practices, it is important to note that such
practices may not only facilitate but also thwart the research efforts, includ-
ing the researcher’s duty to pursue research in an ethical way. In this section
on the basis of my own experiences of bargaining with experts I seek to
reflect the democratic research practices as problematic choices for research
based on expert interviewing, particularly by anticipating their potentially
dominating stand in the interaction with the researcher. However, it is
188 Vaida Obelenė

important to understand that whatever arrangement with the research sub-

ject the researcher achieves – the exclusion or invitation of the expert to
become co-researcher – the tensions and dilemmas will follow from either
approach. Finally, “bargaining,” in the way I mean, does not imply aban-
doning the researcher’s position, on the contrary, the simple point I wanted
to make here is about the researcher in a strong bargaining position. This
emphasis on bargaining, however, communicates my experience that a
study of experts will be a form of bargained research and the interests of
both parties will have to be considered.
Initially, when entering the field, I did not problematise this power relation-
ship between me as researcher and the respondents in terms of their impact on
my further work (for example, when writing on data). While I acknowledged
the risks of giving in to the interests of the respondents, I was predisposed
to secure the most urgent objectives. In particular, there was uncertainty
whether, given the sensitivity of the topic and the small population that con-
stituted the field, I will be able to secure my objectives in terms of data col-
lection. For example, I had a list of 60 experts in Lithuania drawn from the
archive sources going back as far as 16–19 years. Locating these people after
so many years did not mean yet that they would be available for interview-
ing. For instance, the most radical case of accessing (although not exceptional)
is represented by interview negotiation extending across three months with
19 answered phone calls. Furthermore, I identified about one third of those
60 experts as “irreplaceable” in relation to two criteria: on the one hand, it was
a unique know-how this particular individual could offer; on the other hand,
there were experts I had to access in order to generate credible grounds for the
study. Besides, I was learning about inaccessible people (for example, death,
illness, imprisonment, emigration). The pressure caused by my commitment
to meet the study objectives was enormous. At these moments of negotiating I
thought least about how I will have to write on the data I was accessing “at any
cost” and by accepting everything that experts were offering.
For example, the “irreplaceable” expert informed me that he cannot
accept taking part in a study as long as there is a communist perspective:
either the topic is wrong or “I think I am a wrong subject for you.” During our
40-minute negotiation process he recurrently tried to make me understand:
“This was just a job for me nothing more. I worked there like in many other places
during my life.” He, however, not merely disagreed to support the main inter-
est of the study but engaged with contemplation what could be a topic of
his interest. When he accepted the interview, it was clear that he did not
change his understanding on being a wrong subject. Besides, the topics to
be addressed during the interview were reworked to consider his interests.
Furthermore, it was obvious that my choices in analysis, interpreting and
writing will have to develop a way to represent his opinion. This example
shows how the expert was capable of influencing the study by exploiting
my vulnerability ensuing from pressure to get access to data.
 Expert versus Researcher: Ethical Considerations 189

It is in light of this uncertainty that I embarked on the process of negoti-

ating the study with the experts and making promises to them with regard
to which benefits the study will generate and how I would minimize harms
the study may cause. The practices of democratic research at that moment,
among other things, appeared to be offering a valuable bargaining chip. I
thought that such arrangements, for example, would generate sense of con-
trol and predictability on the part of the experts (for example, this seemed
important given their social prominence and the fact that communist func-
tionary is a devalued condition), and in this way eventually would motivate
the experts to accept the invitation to take part in the study. However, dur-
ing the first interviews I acknowledged the possibly thwarting implications
of the democratic research practices. Partially in order to facilitate the inter-
view interaction, partially in order to assess the validity of my interpret-
ation, during the interview I was returning the question by interpreting the
interviewee’s preceding response in terms of the study’s research questions.
My efforts to interpret their words by “reading through or beyond the data”
(Mason, 2002, p. 149) happened to be evaluated by “I might have been not
clear enough” or “That’s not what I said” phrases. Already at this point the
most accurate interpretations were responded by using “Yes, but” construc-
tion. In other words, the likelihood of “interpretative conflict” (Yow, 2005,
p. 142) was immediately obvious.
It was, for example, entirely obvious that I would not be able to engage
with each research subject into a “discussion, emerging as each of us granted
the other interpretative space and stretched to understand the other’s perspective”
(Borland, 1998, p. 331). Moreover, such conflicts are not easy to be resolved
by appealing to academic standards of analysis, and interpretation as devel-
oped by the researcher is always at risk of being discounted as value impli-
cated. Most importantly, what happens when the subject is not interested to
understand the researcher’s perspective? I certainly was experiencing how
some of the interviewees were assuming the roles of “teachers” to explain
that the questions I was asking were “wrong” which implied that my per-
spective and thus ability to understand the phenomena I was studying was
also “limited.” For example, when I sought information about hierarchical
structure in response it was explained to me that I was observing an entirely
irrelevant factor. Instead I was advised to focus on functional structure. This
certainly was a valuable insight but I was not given information on hierarch-
ical structure because the interviewee did not accept it was important (with-
out even considering explaining why it was not important). Furthermore,
there were instances when interviewees refused to accept my perspectives,
even when corroborated with factual data. For example, I asked one of them
whether he could explain why a rather significant proportion of people
who were recruited by the system were those who simultaneously had an
alternative option of engaging with academic career. “There were not many
like that among us,” he refused to accept the unanticipated perspective that
190 Vaida Obelenė

I was offering. My second attempt to elaborate on data I had did not change
his stance. Yet now the most interesting part was to begin: “I can’t get what
are you getting at by asking this question,” he eventually requested I deliv-
ered an explanation. The innovative perspective dislodged the expert from
the position of control: he was not able to predict the consequences his
reply may entail, but most importantly, he realized that he was not cer-
tain what the study’s purpose was. These examples on how those people
who experience themselves as knowledgeable respond, invite projecting the
response the researcher could expect in case they are invited to “validate”
propositions and conclusions of a study. Most importantly, what could be
the consequences of a conflict over the interpretation given that the expert
is not only capable of offering her or his interpretation of reality but may be
capable of obstructing an alternative account proposed by the researcher?
One has to realize that when accessing “irreplaceable” and dominating
experts, the researcher may experience pressures to comply with requests
from experts, which may harm the interests of a study. Above all, it is import-
ant to understand that various potentially counterproductive arrangements
are even necessary in order to protect research subjects from harm. It is
true that the practices of engaging experts at all levels of a study may be
the only way to go about the study (for example Raab, 1987). However,
bargaining behaviour researchers have argued that bargainers might be
offering more than necessary to motivate the other party to accept the con-
tract (for example Corfman and Lehmann, 1993) and this is certainly a
valuable insight for an expert researcher who faces a challenge of defend-
ing the interests of study and not only those of an expert. For example, I
rather quickly during the fieldwork assumed a stance that experts are not
to be engaged at all levels in a study process. By defending this stance I
interviewed 36 experts in Lithuania and only four declined but even not
for the reasons of lacking control. I received five offers “to help” but only
if I wanted. However, partially because of my choices to accept the topics
from the “irreplaceable” experts by making their concerns a part of the
study; partially because of interesting exploratory opportunities offered by
the unanticipated topics that were emerging from data which I collected;
and partially because of what my commitment to protect the anonymity of
experts and confidentiality of their data requested (the challenge which I
did not anticipate to the full when distributing promises), I ended up doing
a rather different study from what I designed initially. Certainly, my advan-
tage was that I could allow this flexibility in research decisions.
However, eliminating the subject from the participation at all levels of
research merely means that one problematic issue is replaced with another:
the beliefs on the necessity of engaging subjects in harm and benefit man-
agement have their good reasons. For the researcher, inferring the subject’s
perceptions of harm is not easy, if possible at all. Making sure that subjects
are not harmed will require a high sensitivity; such an awareness will also
 Expert versus Researcher: Ethical Considerations 191

result from the researcher’s empathic ability to see things from another’s
perspective (Stewart, 1955, p. 132). But the risk exists that such percep-
tions are merely “a kind of subjective colonialism” (Nealon, 1998, p. 32).
Furthermore, even if the expert leaves the study at the researcher’s own
discretion, it is still likely that the writings of the researcher will have to go
through scrutiny of the subjects as audience and any mismanagement of
data may bear immediate implications on the prominent research subjects.
While there are plenty of possibilities to consider the experts as vulner-
able, studying the experts entails a range of mechanisms, which invite the
researcher to consider the impact of the researcher’s work, including how
the researcher could be harmed. The expert researcher has to have a per-
ception of not only entering but also leaving and once again returning to
the field. In light of these considerations, the expert may be defined as the
subject the researcher can never completely disengage from, the expert is
highly vulnerable to the impact of a study and the expert has the capacity
to protect oneself from harm or, alternatively, is capable to push for benefit
in a study context. Given that in my work I chose not to engage the experts
as co-researchers, I had to tackle a difficult task given such a demanding
research subject: first, the researcher has to motivate the prospective sub-
jects to take part in a study, and then the researcher has to find a way to
motivate them to leave the rest at the researcher’s own discretion. Therefore,
in the next section I turn to explore how I engaged with expert’s motivation
management. Simultaneously I aim to illuminate on the issues of expert’s
strength and vulnerability assessment.

8.4 Managing relationships with a motivated subject

What is the researcher’s role in producing the motivated research subject?

Dodge and Geis (2004) present an example of a study which failed to recruit
a single participant on account of having provided unmotivating informa-
tion. It is true that the reasons why an expert is motivated to take part in a
study may be different and, for example, not once I experienced that fac-
tors such as their sense of duty to serve as a research subject on the topic
were motivating the experts. In particular, this section aims to consider the
researcher’s responsibility in producing motivation of an expert to request
a position of control in a study. I want to look into two types of motivation
by offering them as something specifically an expert researcher should be
cautious about. On the one hand, there is motivation, which is directed
by the idea that participation in a study will provide benefits to an expert.
When reflecting on this type of motivation I particularly want to emphasize
the importance of understanding the researcher’s own role in producing a
perception of shared agenda in a study context. I also want to emphasize
the ethical interviewing as a choice that eventually may result in irresolv-
able tensions when also pursuing the ideal of ethical writing. On the other
192 Vaida Obelenė

hand, there is motivation, which is directed by fear that participation in a

study will cause harm to the subject. Learning strengths and vulnerabilities
of research subjects remains an important task if the researcher wants to do
a study in a way that complies to criteria of being ethical. Such information
also enables the researcher to predict resulting motivations on the part of
the research subject and helps to understand strengths and weaknesses of
the researcher’s position in negotiating and implementing the study.

8.4.1 The benefit motivation

This type of motivation may be enhanced by offering incentives which lead
experts to believe that their participation in a study will contribute to gen-
erating a range of beneficial outcomes as seen by them. I felt it was crucial
to understand the expectations of experts with regard to the study but also
more generally what type of study would be most relevant to answer their
urgent concerns. This knowledge gave me a broader understanding of the
expertise I chose to observe and allowed me to accommodate the study I
wanted to do with reference to the urgent issues in the domain of a given
expertise. Now I was able to speak about the study in their language of
motives by clarifying to which extent the study will be able to fulfil such
expectations. While it was certainly not applied or expert purpose driven
research that I intended, the ability to “translate” the objectives of the study
in terms of experts’ vocabulary was a particularly valuable tactic when talk-
ing to experts with pragmatic interests and concerns.
From my experience in this process of negotiation it was vital to begin
with the package of propositions with regard to the impact of the study
that would appear attractive but also intriguing and somewhat amusing for
experts. For example, when negotiating with former functionaries in order
to emphasize a public moment of their experience and to imply their duty
to take part in a study, I was sharing with them my belief that they have a
duty to history writing due to their unique position in the past. This argu-
ment was something that was just too much for some of them to make
any sense (considering that the communist past is devalued). One of them
bitterly commented: “You have to start writing history 10 years after everybody
is dead,” and almost without a pause he added: “Shall we start?” Indeed he
might doubt the high ambition of the study, which pursues the controver-
sial and highly devalued purpose in society and he warned me about the
possible implications if such data – when used “too early” – may have on his
life. But, among other things, his choice to accept the role of the research
subject may mean that when his condition for history writing is met there
is a chance that also his views will be considered.
Experts themselves facilitate the impression of shared agenda. I was asked
perpetually by them when negotiating the interview: “What is there in it
for me?” or “Why should I care?” If I failed to provide satisfying answers,
the experts would themselves suggest what type of study interested them.
 Expert versus Researcher: Ethical Considerations 193

Furthermore, experts might try to show they are not only knowledge-
able subjects but that they also are able to cope with issues as complex as
research questions. Dexter (1970) shares his experience of elite as people
who like to “teach.” However, the decision to open a space for an expert to
reflect the study implicates that the expert develops his or her own ideas
about how the study should be done. It is true that evading a response to
such questions or failing to satisfy the expert’s desire to “teach” is not the
best strategy for a researcher who has to recruit an irreplaceable expert or
achieve consent to use data. However, once the researcher answers the ques-
tion such as “what is there in it for me?” convincingly enough she or he might
have recruited somebody who thinks of himself as not merely a research
subject but a partner.
This brief excursus into what motivates experts to take part in a study
gives an understanding that the researcher should be cautious that there are
established topics and concerns in the field to such an extent that working
on the topics as desired by the researcher might be difficult, if possible at
all. Among other things, success depends on creating a sense of interest and
support to the study on the part of the research subjects. While research-
ers experience difficulties in gaining the experts’ cooperation in research
efforts, they will have to think about the ways to motivate the reluctant
to contribute, disinterested or highly busy experts to take part in a study.
And, of course, the opposite of this is also true: the expert researcher may
encounter highly motivated individuals who have a particular interest or
strong feelings about the topic and want to participate actively in a study.
Moreover, they may have their own ideas about the benefit the study will
generate. But here await decisions to engage or not engage in study bargain-
ing, as discussed in preceding sections.
In my work I discerned the interview interaction as a particularly prob-
lematic source in building the notion of shared agenda. The interview nego-
tiation is explicitly marked with the purpose of shaping a shared agenda,
but in the interview interaction this explicitness typically will be lost. The
issue to contemplate for an expert researcher is that the fieldwork ethics
(for example Ryen, 2004) might turn to be another choice that will put a
further spoke in the researcher’s wheel. Literature typically emphasizes the
importance of maintaining a balance between different roles: researchers
find themselves experiencing tension between professional obligation to
seek the best information possible and interpersonal obligation to respect
the subject’s privacy and well-being. Advice on ethical fieldwork typically
considers the issue of informed consent and focuses on building the rapport
and on the “tricks” of maintaining it throughout the interview. The subject
will not be judged even if the researcher disagrees with the subject’s views;
instead the researcher seeks to understand and displays an understanding.
Clearly it may only mean that these encouragement techniques are inter-
preted by the subject in terms of agreement and shared views. It was not
194 Vaida Obelenė

once during my fieldwork that while experts accepted reluctantly to engage

with the study, it was throughout this investment in the interview inter-
action that they developed their ideas of the agenda the study pursued and
came to care about the outcome it produced: by contemplating the subject
matter they understood the purpose and usefulness of the study (but in
their own way).

8.4.2 The harm motivation

Certainly there are actions the researcher can take to limit and regulate
the experts’ exposure to vulnerability in a study. On the one hand, under-
standing the expected harm perceptions on the part of the experts invites
the researcher to engage with management of motivation such as the steps
targeted at reducing the collection of potentially harmful data, particularly
those which are not even necessary for a study. On the other hand, collect-
ing data on sensitivities, competence and social embeddedness of experts
in my work proved to be a valuable source of information when projecting
which of my choices when analysing, interpreting and writing may cause
harm to experts. In the following I seek to discuss these issues in relation to
three images of an expert, namely the sensitive subject, the competent sub-
ject and the socially embedded subject.

8.4.2.1 The sensitive expert

The way the researcher delivers an image of a trustworthy individual and a
credible professional capable of preventing the harm is but only one aspect
the researcher ought to be cautious about. An equally important task during
my fieldwork was reducing the collection of sensitive material in the context
of the sensitive topic, including the task of not asking unnecessary and pro-
vocative questions. Some of these tasks I was tackling certainly may be con-
sidered as an instance of self-censoring but my unwillingness to endanger the
interviewees won out. My decisions reduced the problems of collecting such
data but it is also true that these sensitive and often provocative topics were
entering the study in other ways. For example, I refrained from addressing the
topics of property stealth, one of the crimes that the stigma theory tends to
assign to the people I was studying in Lithuania, by instead establishing that
I am interested in production and transfer of know-how. However, the inter-
viewees were addressing the sensitive issues, for example, by accounting how
others were stealing and by expressing their regrets that I was not interested
in the issues which, according to them, were the topics of the utmost import-
ance. While I explicitly omitted the possibility that the study will tackle such
topics, my perception was that this information invited the experts to think
that they could openly talk about such urgent for them topics without even
occasionally bothering to enter “off-the-record” passage.
Nevertheless during the interviews there were always sensitive issues to
address and they tended to vary unpredictably among the people I was
 Expert versus Researcher: Ethical Considerations 195

interviewing. To my relief, it often happened that when after a lengthy hesi-

tation I accumulated courage and asked, the experts appeared to be highly
eager to explore an “interesting” perspective. When reflecting on what mis-
leadingly appeared to be the sensitive issues to me, I realized that often such
perceptions resulted from my own mismanagement of the study: I expe-
rienced the issues as potentially sensitive when I felt that the interviewee
would think that the topic is not a part of an agreement about the interview.
I realized that the issues were sensitive only when the interviewees asked to
stop the recording or when they were moving closer to me, as if they wanted
to tell a secret.
Yet it is true that the experts grow increasingly comfortable during the
interview; they become engaged and committed to the study. At these
moments they are ready to go to great lengths to make sure the researcher
will be equipped to do the study. Moreover, the experts do not only breach
the limits of their own privacy, they also breach the privacy of others and
disclose the secrets of organizations. Accounts of expert researchers are full
with descriptions of such “miracles” when their respondents “open up” and
this is how I experienced the process. In that context, it was important for
me to understand how the length of the interview was producing the expos-
ure of experts to vulnerability. My perception of an interview dynamics was
that after a very desirable development such as “opening up” the moment
of “breaking down” follows. I achieved understanding that two hours of
interaction almost certainly meant that a productive time of an interview
was exhausted and the experts by now were wandering into issues which
increased their exposure to harm and, among other things, requested that I
give away the researcher’s role.
I collected plenty “off-the-record” or “not for press” episodes during the
interview. However, there was also plenty of apparently sensitive informa-
tion without such markers involved. It is obvious that when at first the
researcher struggles to recruit respondents amid the risk they may decline
the sensitive topic, now the researcher faces decisions how to go about
the sensitive data collected, possibly by wrongly assessing which issues
are sensitive. The guidelines the experts give may be vague. What should,
for example, the phrase “I trust you” pronounced recurrently throughout
the interview entail for what to do with data? To understand the possible
impact of my decisions of working with data, I sought to explore whether
the specific topics appeared to be sensitive in the context of other inter-
views. In my work such consideration proved to be a valuable tool, albeit
still limited, in developing sensitivity to vulnerabilities of the experts and
an additional source of judgement when taking decisions.

8.4.2.2 The competence of the expert

The reason to assign the research subjects with the status of an expert alone
draws the attention to the fact that they are knowledgeable and invites the
196 Vaida Obelenė

researcher to assume their competence. I, for example, anticipated from

the beginning that being of the age which means that I was too young
to experience the past events might generate an impression of my incom-
petence. I anticipated a certain patronizing and, accordingly, exhaust-
ive explanations. However, the experts accounted by mentioning names
of people without explaining who they were, they used the slang words
without explaining them, they were interrupting my questions by imply-
ing that they understand what I want to know, and they spoke about past
events without contextualizing them as if I was taking part in them. The
experts assumed my competence to understand such highly specific refer-
ences and they assumed they understand my purpose. But as much as such
a format was difficult to follow, it seems, it also demonstrates disregard of
the experts to the competence of the researcher. This illuminates the prob-
lem the researcher has to face: the experts set interaction and produce data
under the false assumptions of the researcher’s competence.
I found it was crucial to shape the understanding of different senses of
competence that characterize the researcher and the expert: it allowed to
ease domination relationships by shaping perception of difference and spe-
cialization but simultaneously provided a chance of informing the experts
about the different perspectives into the topics. For example, it was import-
ant to think about how I could articulate the questions I wanted to ask in a
way that would make the experts perceive them as innovative perspectives
into the topic. When the expert would say, “I never though about this in such
a way,” I felt that I was recognized as competent.
Creating a sense of difference in competence of the interacting parties
may help to offset experts’ desire to teach or overpower, but, of course, the
opposite response might be the case. Still such strategies on revealing dif-
ferent competence may helpfully invite experts to express a more profound
interest in the study’s purposes and, if (like in the example I present below)
such perception of incompetence results too late, there is still a chance for
them to inform the researcher on the status of the data they delivered under
the false assumptions. For example, one expert devoted seven hours of his
time to the study (this certainly was one of the exceptionally long inter-
views which extended from “less than one hour”). Only after five hours had
elapsed he, an engineer by education, realised that there was not a single
“yes” or “no” question asked. He then tried to figure out how I was going
to produce tables with percentages and graphs from the passages that he
delivered in the course of our interaction. “I only hope you do not set me up,”
he then whispered for a goodbye when I was finally leaving. Ideally the
principle of informed consent rules out the possibility of covert research
and deception. However, regardless of my earnest attempts to inform the
experts about the study, I had the sensation of covert research all through-
out the fieldwork, similarly like that described by Luff (1999). Not once did
I come across experts who felt competent and rushed impatiently through
 Expert versus Researcher: Ethical Considerations 197

introductory stages by delivering their consent without informing them-

selves about the study. In fact I got so used to that pattern that I felt highly
disturbed when the nineteenth interviewee spent about 30 minutes “exam-
ining” the study and carefully taking notes.
All in all, the researcher cannot expect that experts, however knowledge-
able they are, will be able to assess the meaning and implication of their
account in relation to the interpretative perspectives of the study. When
experts assume that they understand hazards and benefits pertaining to
their participation in a study, it merely means that they take their point of
view for the researcher’s point of view. People are competent in their own
ways: “there can be no universal criteria of ascertaining competence” (Sin,
2005, p. 280). The examples of the experts, who assume their competence in
a study context as discussed above, may mean that they deliver data by con-
trolling for entirely different risks than the study represents. The respon-
sibility that awaits the researcher to protect the experts from harm may
turn out to be a challenging task, particularly that there just does not seem
to be another way to interpret their data than by ascribing the meanings
they assign. Beyond the strategies to encourage the experts to acknowledge
different senses of competence, I proposed earlier that my solution to this
problem was to find a way to motivate the experts to leave the study at the
researcher’s own discretion.

8.4.2.3 The social embeddedness of the expert

Regardless of how researchers conceive experts in terms of their know-how,
the experts are implicated into social hierarchies and associations. Experts
are few, they may be interconnected, and they take part in unique events.
This signals that securing the anonymity of the expert is a likely problem an
expert researcher will face. Furthermore, their close interconnection means
that some may harm others by disclosing information others would con-
sider private or secret. For example, by choosing to work on people who
worked in one and the same high office, I entered a network of friends
and foes; some of these people have not been in contact for as many as
20 years. Talking about their own experience, which is also the experience
of others requires a high sensitivity and a skill from the expert of how to
go about the topics if others should not be harmed by what a given expert
says. And while some failed this task trespassing upon the privacy of others
or breaching the secrets of organizations unintentionally, there also were
those who took their chance to take revenge on their foes or punish those
whom they considered offenders. Furthermore, the questions they asked
about others were incredibly direct (even such as, for example, “is he still a
heavy drinker?,” “is he still a womaniser?” or other references to the “dark,”
immoral, or sometimes embarrassingly private side of other experts inter-
viewed). These experts’ stories, gossips and questions about others, how-
ever, in my work became an interesting perspective in developing awareness
198 Vaida Obelenė

about sensitive topics. For example, there was one expert who spoke about
how a human being is responsible for protecting the devalued accomplish-
ment from destruction. I was interpreting this passage, which he delivered
in an abstract and somewhat metaphorical way, as a valuable insight into
the knowledge transfer from one system into another. It was with the help
of “gossipers,” however, that I was able to realize that the criteria in terms
of which the expert described the “obligation” correspond to the business
project which not once publicly was considered in terms of theft. This cer-
tainly influenced me to become cautious in choosing how to exploit such
a passage in the text. Furthermore, the expert’s skill in delivering the sensi-
tive message by abstracting it to the level where it loses sensitivity inspired
some of the choices I made when writing about sensitive topics.
Expertise implicates as a range of beliefs and stereotypes; it may entail
a devalued and stigmatized condition. A considerable proportion of the
69 people I interviewed in two societies may be described in terms of being
prominent people and some of them, for example, established their post-
communist lives without “highlighting” their former communist involve-
ment. The participation in a study may immediately bear implications on
the socially embedded life of an expert. Moreover, I certainly experienced
myself to be much less a “disinterested observer” than I wanted to be. While
the interaction with research subjects starts from the ideas about what type
of being I am going to encounter, this is yet another reason why it is helpful
to foster somewhat different images of experts: the images would help the
researcher to be assertive enough to enter into a negotiation with an expert
and seek the data for the purpose of own scholarly investigation; but simul-
taneously the image of an expert would help the researcher to be cautious
and attentive to vulnerabilities of the research subjects.

8.5 Conclusion

In this chapter and drawing on my experiences of studying powerful research

subjects, I suggest discussing democratic research practices as a set of possibly
counter-productive research tips for expert researchers. While the proponents
of democratic research may argue that such practices are also immediately
ethical, it is, in my view, important to question this proposition. For example,
the discussion of the possibility of interpretative conflicts between researcher
and experts as subjects of research suggests that there is a risk that the study
may reach an ethical impasse. Instead, the expert researcher needs to antici-
pate and carefully negotiate the tensions which may follow from the decision
to adopt recommendations in the literature on turning the research subjects
into co-researchers. The reflection of the status of powerful research subjects
has long been an issue in research. Yet it is important to remind oneself of the
particularity of the experts as research subjects, including their potentially
dominating stand in the interaction with the researcher.
 Expert versus Researcher: Ethical Considerations 199

Whatever arrangement the researcher achieves – the exclusion or invita-

tion of the expert to become co-researcher – the question remains what kind
of action the researcher could take in order to anticipate the tensions and
dilemmas that will follow from either approach. It should be obvious from
my discussion that the solution I adopted in my own work is conventional
and follows the first strategy: beyond the interview situation the research
subject is not invited to participate in the process, while the researcher
assumes responsibility with regard to the representation of the interests of
the research subjects. While this is certainly not the only solution I conclude
on the basis of my experiences and the specific framework of my study, that
such a model can sufficiently secure the purpose of the investigation and
simultaneously fulfil the duty to represent the interests of the research sub-
jects. Such arrangement, however, implicates that the researcher alone is
responsible for decisions with regard to harm and benefit management in
a study context. This is an uneasy task given that the expert, among other
characteristics, is a research subject difficult to disengage from and is vul-
nerable to the impact of a study.
In any case, a study of experts will always be a form of bargained research:
it will be an outcome of negotiations between the researcher and the expert;
the interests of both parties have to be considered. In order not to be cor-
rupted by the experts’ interests the researcher needs to stick to the research
agenda and to remain assertive enough in defending his or her desired role
as well as the topics of the study that need to be supported by the experts.
Finally, while underlining the commitment to the ideal of ethical research, the
chapter invites caution with regard to motivating the experts to take part in
the research. The expert as a respondent is vulnerable to the impact of a study
or may be seduced by its apparent benefits. Here I discussed how the expect-
ation and anticipation, on the part of the respondent, of harm and benefit
resulting from the study is actually facilitated by the researcher’s own tech-
niques of engaging the respondent, techniques that provoke the wish of the
respondent to assume control and, eventually, the status of a co-researcher.

Acknowledgements

I wish to thank Alexander Bogner and Herwig Reiter for their comments.

Further readings
Dexter, L. A. (1970) Elite and specialized interviewing (Evanston: Northwestern
University Press).
Raab, C. (1987) “Oral history as an instrument of research into Scottish educational
policy making” in Moyser, G. and Wagstaffe, M. (eds) Research methods for elite
studies (London: Allen and Unwin), pp. 109–25.
Luff, D. (1999) “Dialogue across the divides: ‘Moments of rapport’ and power in
research with anti-feminist women” in Sociology 33, pp. 687–703.
200 Vaida Obelenė

References
Ali, S. and Kelly, M. (2004) “Ethics and social research” in Seale, C. (ed.) Researching
Society and Culture (London: Sage), pp. 115–28.
Bogner, A. (2005) Grenzpolitik der Experten. Vom Umgang mit Ungewißheit und Nichtwissen in
pränataler Diagnostik und Beratung (Weilerswist: Velbrück Wissenschaft).
Borland, K. (1998) “ ‘That’s not what I said’: Interpretative conflict in oral narrative
research” in Perks, R. and Thomson, A. (eds) The Oral History Reader (London:
Routledge), pp. 320–31.
Corfman, K. P. and Lehmann, D. R. (1993) “The Importance of Others’ Welfare in
Evaluating Bargaining Outcomes” in The Journal of Consumer Research 20, pp. 124–37.
Dexter, L. A. (1970) Elite and specialized interviewing (Evanston: Northwestern
University Press).
Dodge, M. and Geis, G. (2004) “Fieldwork with the elite: interviewing white- collar
criminals” in Hobbs, D. and Wright, R. (eds) The Sage handbook of fieldwork (London,
Thousand Oaks, New Delhi: Sage Publications), pp. 79–92.
Hammersley, M. (1995) The Politics of Social Research (London, Thousand Oaks,
New Delhi: Sage Publications).
Harding, S. (1987) Feminism and Methodology (Bloomington, Milton Keynes: Indiana
University Press, Open University Press).
Hertz, R. and Imber, J. B. (eds) (1995) Studying elites using qualitative methods (Thousand
Oaks, CA: Sage Publications).
Homan, R. (1991) The Ethics of Social Research (London: Longman).
Kimmel, A. J. (1988) Ethics and Values in Applied Social Research (Newbury Park: Sage
Publications).
Kvale, S. (1996) Interviews: An Introduction to Qualitative Research Interviewing (Thousand
Oaks, CA: Sage Publications).
Luff, D. (1999) “Dialogue across the divides: ‘Moments of rapport’ and power in
research with anti- feminist women” in Sociology 33, pp. 687–703.
Mason, J. (2002) Qualitative Researching (London, Thousand Oaks, CA: Sage Publications).
Mauthner, M., Birch, M., Jessop, J. and Miller, T. (eds) (2002) Ethics in Qualitative
Research (London, Thousand Oaks, CA, New Delhi: Sage Publications).
Moyser, G. and Wagstaffe, M. (eds) (1987) Research methods for elite studies (London,
Boston: Allen and Unwin).
Nealon, J. T. (1998) Alterity politics: ethics and performative subjectivity (Durham, NC,
London: Duke University Press).
Odendahl, T. and Shaw, A. M. (2001) “Interviewing elites” in Gubrium, J. F. and
Holstein, J. A. (eds) Handbook of Interview Research: context and method (Thousand
Oaks, London, New Delhi: Sage Publications), pp. 299–316.
Raab, C. (1987) “Oral history as an instrument of research into Scottish educational
policy making” in Moyser, G. and Wagstaffe, M. (eds) Research methods for elite
studies (London: Allen and Unwin), pp. 109–25.
Ryen, A. (2004) “Ethical issues” in Seale, C. (ed.) Qualitative Research Practice (London;
Thousand Oaks, CA: Sage Publications), pp. 230–47.
Sin, C. H. (2005) “Seeking informed consent: reflections on research practice” in
Sociology 39, pp. 277–94.
Stewart, D. A. (1955) “Empathy, common ground of ethics and of personality theory”
in Psychoanalytic review 42, pp. 131–41.
Yow, V. R. (2005) Recording Oral History: A Guide for the Humanities and Social Sciences
(Walnut Creek, CA, Oxford: AltaMira Press).
Part III
Fields of Application: Applications
of Expert Interviews in Different
Fields of Research
This page intentionally left blank
9
How to Interview Managers?
Methodical and Methodological
Aspects of Expert Interviews as a
Qualitative Method in Empirical
Social Research*
Rainer Trinczek

9.1 Introduction

In discussions on methodology, purists frequently raise objections to the

expert interview1 on grounds of it being a “dirty” method. Expert inter-
views, they claim, operate in a “no-man’s-land” somewhere between the
qualitative and quantitative paradigm2 devoid of much profound methodo-
logical reflection. Flexibly applied in empirical research, relying more or
less on an interview schedule depending on the research interests and the
specific research question involved, conducted in a more or less open fash-
ion, the data prepared, analysed, and interpreted in some obscure fashion
according to homemade recipes, expert interviews, in the eyes of method-
ology gurus from the ranks of qualitative as well as quantitative method-
ology, fail to meet the standards of either one of the paradigms. The social
researchers M. Meuser and U. Nagel have experienced this with the quali-
tative research community: “Whenever the word “guided interview” falls,
relentless scepticism from hard-line advocates of “soft” methods is guaran-
teed. Resort to concepts such as semi-structured or focused interview does
little to remedy this – quite to the contrary” (Meuser and Nagel, 1989, p. 8,
translated from German).
Such discontent with guided expert interviews is also reflected in the rele-
vant literature on methodology, which rarely deals with this method in
any detail. Whereas, in the German context, the “narrative interview” and
“objective hermeneutics,” which tend to be viewed as particularly refined,
carefully considered, and highly sophisticated methods of collecting and
analysing data respectively, have attracted overwhelming interest, the

203
204 Rainer Trinczek

guided interview has been relegated to the sidelines, even though it is actu-
ally much more frequently used in empirical social research.3
This article is an attempt in defence of the expert interview as a useful
and suitable instrument for data collection. It will be discussed with refer-
ence to the qualitative paradigm, since guided expert interviews are typic-
ally, and rightfully so, associated with qualitative methodology. Drawing
on interviews with managers, the problem of the ideal interview setting for
subjective structures of relevance to freely emerge will be discussed, which –
as we all know – is a key issue in qualitative methodology.4

9.2 Methodological basics

With recourse to symbolic-interactionist and phenomenological approaches,

standardized methods of interviewing have been criticized on grounds of
systematically shutting the door on opportunities to get at what interviewees
“really” think, since such approaches to interviewing fail to adequately con-
sider the specific interview situation as well as the constitutive features of
everyday life and everyday communication along with the subjective attri-
bution of meaning that takes place therein. The interview situation, thus the
reasoning, is of course no exemption in terms of the significance of ascrib-
ing meaning: in light of their everyday knowledge and structures of rele-
vance, interviewees attribute certain meanings to the questions asked and
respond based on those subjective interpretations. Accordingly, Cicourel,
in his classical text on “Method and Measurement in Sociology” (1970),
stresses: “If it is correct to assume that persons in everyday life order their
environment, assign meanings or relevances to objects, base their social
actions on common-sense rationalitites, then one cannot engage in field
research or use any other method of research in the social sciences without
taking the principle of subjective interpretation into consideration. (...) To
ignore this point is to render both the questions (or conversations) and the
answers received as problematic and/or meaningless.” (Cicourel, 1970, p. 61)
And: “Standardized questions with fixed-choice answers provides a solution
to the problem of meaning by simply avoiding it” (Cicourel, 1970, p. 108).
The high degree of sensitivity to context is rightfully regarded as a cru-
cial advantage of qualitative interviews over standardized survey methods;
this sensitivity pertains to the context of interviewees’ everyday lives and
the respective structures of relevance orienting their actions as well as
to the context of the interview situation. With this in mind, it has been
argued that social scientific research must systematically “adapt to the
rules of everyday communication that exist prior to the research process”
(Schütze and others, 1981, p. 434, own translation) as an indispensable
methodological prerequisite for successful interviewing. The terms “open-
ness” and “communication” refer to two basic norms guiding the process
of data collection (Hoffmann-Riem, 1980) that, at the methodical level, are
 How to Interview Managers? 205

conceived to enable an interpretive sociology to effectively make good on

its methodological promises.
However, in the further course of the debate on methodology, a peculiarly
one-sided specification of these “prior rules of everyday communication”
has taken hold. The fundamental openness of qualitative interviewing for
the specific area of research in question, with the key advantage of allowing
the researcher to flexibly adapt to the large diversity of differently structured
social contexts, has been increasingly displaced by a set of best practice
instructions defining “good” qualitative interviewing practice.
With reference to the supposedly largely narrative structure of everyday
communication, which research is required to adjust to, it has become a
widely unquestioned commonplace that the best possible interview set-
ting for encouraging interviewees to disclose their subjective attributions of
meaning and structures of relevance is one in which the interviewer largely
refrains from intervening in the interview. The interviewer is called upon
to exercise pronounced restraint – such the logically perfectly consistent
reverse conclusion – as a means of respecting the “prior rules of everyday
communication in the field.”5
This line of reasoning, indeed, plays a significant role in nourishing the
scepticism with which hard-line advocates of qualitative methods view the
guided expert interview. In their perception, this method involves too much
interviewer intervention and structuring so that interviewees are not given
the opportunity to freely develop their own structures of relevance.
In society, however, there exist realms of social reality where every-
day communication for the most part follows completely different rules.
A company, for instance, represents such a world. Now, if we take the meth-
odological foundations of qualitative research seriously with an eye to com-
panies as areas of research, we must attempt to find ways of interviewing
that accord with the prevailing modes of everyday communication in this
segment of social reality.
In the following, I will make such an attempt drawing on two manage-
ment studies. One of the studies focuses on the company level and exam-
ines collective orientation patterns of managers regarding the issue of
codetermination (Trinczek, 1993). The other study is concerned with the
interface of company life and the private life of executives. It looks into
how executives (re-)arrange their lives to cope with changed demands in
their professional and private lives (cf. Ellguth and others, 1998, Behnke and
Liebold, 2000, Liebold, 2000). Both studies involve research objects that can
be considered suitable cases for the purpose of the argument proposed here,
since in both instances they are sufficiently “soft” to qualify as appropriate
cases for qualitative research, even in the eyes of the qualitative paradigm’s
most devout disciples.6
A comparison of the methods applied in both studies is interesting mainly
for the purpose of illustrating that due to the fact that managers on a daily
206 Rainer Trinczek

basis are engaged in worlds that are very differently structured in terms of
the prevailing modes of communication (company – private life), it follows
from the tenets of qualitative methodology that, depending on the research
question, very different techniques of data collection may be adequate to
the task.

9.3 The initial phase of the interview

Any interview with managers in a company environment at the outset faces

managers’ implicit and explicit expectations of such a conversation, irrespect-
ive of the specific research topic involved. These expectations are for the most
part shaped by the existing rules of everyday communication in the com-
pany. Accordingly, concerning the, for the success of an interview, crucial
opening sequence, managers expect the research team to confront them with
questions, which they are supposed to respond to. Once the greeting ritual is
completed, the “host” typically opens the conversation with words like the
following, which express the anticipation of clearly defined roles in the inter-
view setting: “Okay, fire away. What do you want to know? Feel free to ask.”
Such initial expectations obviously largely arise from the interviewee’s
everyday work experience. Being asked questions by superiors or asking the
right questions oneself are an integral part of managerial work. Managers
project expectations rooted in such experience onto the interview situation.
In the interview, the researchers are the ones to ask questions and they are
the ones to answer, just as they expect their subordinates to do or their
superiors expect them to do: briefly and precisely, to the point and with-
out wandering from the subject. Of course, few managers actually live up
to this ideal. When conducting interviews, we regularly come across inter-
viewees who, by constantly straying from the subject, make it difficult for
the researcher to successfully accomplish the interview. Nevertheless, the
expectations are still in effect; sometimes the interviewee even becomes
aware of deviating from this ideal and brings up the subject in the course of
the interview. One manager, for instance, before the interview, admitted to
having a tendency of getting too caught up in details and asked us to inter-
rupt him whenever he dwelt on a topic too long.
The guided interview, especially in the version employing a more closed
interview schedule, defines an interview situation that corresponds per-
fectly with managers’ prevailing expectations in the opening sequence of
an interview and therefore, undoubtedly, is the best choice for the opening
stage. Switching to other methods of interviewing first requires a cautious
process of overcoming managers’ predominant question-and-answer orien-
tation. This can take considerable time and demands a high degree of social
competence. Opening an interview with an executive by asking him or her
to engage in a lengthy narration runs the risk of dooming the whole inter-
view to failure – the reason being that such an interview situation is almost
 How to Interview Managers? 207

diametrically opposed to the everyday communication structures at the

workplace. In the eyes of managers, “time” – especially their own time – is
a scarce and valuable resource. In their perception, constantly lacking and
being pressured for time marks everyday action in the company setting.
Requesting managers right at the beginning of a conversation to take the
time for lengthy elaborations is likely to appear provocative in light of this
perception of their own work situation. This would create a considerable gap
between the interview situation and the rules of everyday communication
as they exist in the field and are reflected in managerial expectations that
the research team give the conversation a clear structure.

9.4 Developing the interview

Interview situations are not static; typically, they develop a life of their own
such that the initial expectations the parties to the interview bring to the
interview may change as the interview progresses. For instance, once the
interview situation is successfully established and initial insecurities are
overcome, it is a quite common experience in interviews with management
that the atmosphere of the conversation quickly becomes noticeably more
relaxed. Gradually, the interviewees begin to realize the difference between
the interview situation and the question-answer situation that is a regular
part of their daily work experience. As opposed to the social situation in the
company context, the interview puts nobody under scrutiny and no one has
to justify shortcomings or failure; in consequence, the initially noticeable
strategic handling of information can be relaxed.
This change in the managerial perception of the social situation “interview”
becomes apparent in the interviewees readjusting their initial expectations
of the distribution of roles in the interview setting. Once the interview
enters this stage, the initial, typified expectations of a question-answer-type
conversation no longer prevail; as a result, the nature of the interview can
change without running the risk of failure. This is precisely the situation
when an “openness” in conducting qualitative interviews, which not only
allows to flexibly adapt the style of interviewing to the situation but virtu-
ally demands to do so, proves its worth.
This is the time when systematic shifts to “other” forms of interviewing
can take place with potentially promising results. However, the researcher
in this situation is still not fully free to make use of the whole range of
interviewing techniques. Rather, there remain limits to what is feasible
that clearly have to do with the subject matter that one seeks to approach
through the interview. In the following, drawing on the above-mentioned
cases of “orientation patterns of managers concerning codetermination”
and “life arrangements,” I will show that the appropriate choice of interview
technique in the “second stage” of interviews with management depends on
the nature of the object of research.
208 Rainer Trinczek

The crucial difference between these cases with regard to the form of
interviewing is that, in the one case, the object of research solely involves
a manager’s company life; hence, when topics of this kind are addressed in
the interview, the typical rules governing everyday communication in this
setting determine the manager’s expectations of the interview – and the
interviewer is well-advised not to disregard them in the way the interview is
conducted. The second case (“life arrangements”), on the other hand, refers
to private life outside of the company, where other modes of everyday com-
munication prevail. In this case, it is indeed possible to overcome managers’
initial expectations and to shift to more narrative forms of interviewing.
The interviewee in the interview situation mentally “leaves” or (in the best
case) even “forgets” the company setting with its typically non-narrative
modes of communication – even if the interview is conducted in the inter-
viewee’s own office on company premises.

9.5 Interviewing managers about business issues

If the topic of the interview primarily relates to the company context, the
interview will typically not take on a narrative form even in the second
stage when the conversation becomes more relaxed. Attempts at stimulating
at least brief narrations in the course of an interview are rarely successful
with managers. This is lesser due to managers’ desire to not waste valu-
able time “chattering” – as they would call it. Rather, it has to do with the
fact that to the extent that managers’ expectations of the interview shift
from the familiar initial question-answer focus – which happens frequently
although not always – they mostly tend to switch to another structure of
everyday communication that is rooted in managerial experience in the
company context: the expert discussion, the basic communicative structure
of which managers are mainly familiar with from the open discussions in
project teams and other team-type working arrangements.
In the course of interviews on company matters, we can thus observe a
shift between differently structured forms of conversation, which are both
within the range of common company practice. “Good” qualitative inter-
viewing practice requires the interviewer to “go along with” this shift from
one form of communication common to a company setting to another.
The new interview situation represents a relaxed discussion setting, in
which, although the topics are still structured by a “competent” research
team, the interviewees voice their views on issues without any reservations,
allow them to be questioned, and critically reflect on them. Since the inter-
view situation is free from the demands of action and socially inconsequen-
tial, it occasionally nourishes a degree of candor and open self-reflection
on the managers’ part, which they, in this form, would normally not allow
themselves to engage in in the company context with its mostly strategic
communication and interaction style.7
 How to Interview Managers? 209

The fact that the interviewees at times willingly go considerably over the
time limit initially set for the interview, even though seeking just an add-
itional 15 minutes may have required a fair amount of haggling at the point
of arranging the interview, shows how attractive this inconsequential con-
versation situation is to them. It is not uncommon for the researchers to
finally take the initiative to end the conversation and not the managers.
Kern and others (1988) speak of a virtually “cathartic effect” in manage-
ment interviews and have essentially traced this back to the one- dimensional
“separation” of the strategic and communicative sides of action in company
life. “If even a social contact in an interview situation is experienced as
an opportunity for compensation, this must be a consequence of having
to suppress discursive communication in everyday life. The reason for this
may well be that being confined to what is subjectively perceived to be
“purely strategic” action acts to gag the strategist himself” (Kern and others,
1988, p. 93, own translation from German, emphasis in the original).
Under such conditions in a company, “the relationship offered by ‘under-
standing people from the outside’,” according to the authors, provides “short-
term relief” (Kern and others, 1988, p. 94, own translation from German);
the interview situation affords an opportunity to dissolve the “communi-
cation blockage.” Kern and others have apparently experienced that the
managers they interviewed were often inclined to seize the opportunity of
the interview as a stage for self-dramatization with cathartic effects, while
relegating the sociologists to the role of an audience or the straight man
providing the feeds for further elaborations. They report that attempts to
question managerial views and to point out problems regularly failed: “At
points where their views are in threat of being called into question in dis-
course, repression of other legitimate worldviews is repeated.” (Kern and
others, 1988, p. 94, own translation from German) In the view of Kern and
associates, such pathogenic forms of communication can occasionally inter-
fere with the task of “successfully” conducting an interview.
As indicated above, I have mainly had other experiences. In my
observation, managers appear to seek less a patient audience for exces-
sive self-dramatization, rather than being more interested in the social
researcher as an expert and a person to converse with, who takes a different
analytical and conceptional perspective.8 In fact, managers exhibit quite a
broad range of different interview behaviour. Some are very open-minded
and self-reflective in weighing different options and positions, carefully
argue their own views, display some willingness to give counterarguments
thoughtful consideration, and enquire about experiences the interviewers
may have made in other companies. Others are more inclined to sweep aside
objections, presenting their own position as ultimately the only reasonable
stance on an issue. However, such managers, too, are obviously also bound
by the tacit norms of discourse: Well-argued interviewer interventions are
not perceived mainly as annoying instances of interfering with the process
210 Rainer Trinczek

of self-dramatization; rather, it seems as if at times the interviewees virtu-

ally hope for objections to be raised, thus providing them with an oppor-
tunity to all the more effectively present their own arguments to repudiate
them as inadequate, out of touch with reality, or limited in perspective.
Irrespective of the specific behaviour displayed in the interview, the great
majority of managers on all accounts gives the impression that it perceives
the conversation as an opportunity to indulge its “appetite” for intellec-
tual exchange, reasoning and debating under conditions free from the
demands of action; and it seems to be precisely this condition of freedom
from the need to act characteristic of the interview situation that whets
this “appetite” to begin with. Freedom from the demands of action is also
the key difference to the expert discussion, which managers are familiar
with from daily experience and which provides a communication structure
that they can easily adopt in an interview situation. There is, however, no
difference – and this needs to be emphasized – in terms of the fundamental
principles of communication that are in effect in asking questions and giv-
ing answers, in reasoning and debating.
Interviewee expectations that in this way call for a discursive interview
structure require a specific type of interviewing that only to a limited extent
corresponds to the standards of qualitative research as regularly stressed in
the literature. These standards, as we all know, emphasize that the inter-
viewer takes a neutral and supportive stance in interviewing and under all
circumstances abstains from his or her own interventions to not run the risk
of interfering with the interviewees developing their own subjective struc-
tures of relevance. The contrary holds true when interviewing managers: in
light of the above outlined “prior rules of everyday communication” in this
specific field of research, an expert interview with managers must by neces-
sity be designed to be conducted in a discursive, argumentative fashion.
The researcher must of course allow interviewees time and sufficient
opportunity to give shape to their positions; on the other hand, the inter-
viewer regularly confronts the interviewees with opposing views during the
interview. Such a mode of “discursive, argumentative” interviewing does
not intend to suggest that such interviews must necessarily be “confron-
tational.” It is a matter of interviewer competence to find the appropriate
“dosage” when challenging the interviewee. The purpose of bringing up
objections and opposing views is as a matter of course not to get interviewees
to modify their positions during an interview; rather, the ultimate objective
of such interventions is to encourage interviewees to discursively develop
their structures of relevance by establishing a conversation structure that
is common to everyday company life beyond the “artificial” interview situ-
ation. In line with this intention, the interviewer should explicitly assume
the role of an “actor diabolus” when intervening in order to minimize the
risk of interviewees implicitly adjusting their responses along supposed lines
of social desirability possibly conveyed by the interviewer.
 How to Interview Managers? 211

Admittedly, a constructive discursive interview situation presupposes an

interviewer commanding a high level of competence, both socially and
pertaining to the subject matter in question. As a necessary condition for
managers to accept and engage in a discursive, argumentative, and for the
research project potentially productive interview situation, the interviewer
must be able to give the impression of being sufficiently compatible and on
a par with the interviewee. This, just to mention in passing, also frequently
has to do with age and status of the researchers. The higher a manager’s own
level of qualification, formal status, or the greater the area of responsibility,
the higher are his or her tacit expectations concerning the formal status of
the other. Young researchers who have yet to acquire their doctoral degree
hardly stand a chance to be accepted as a competent counterpart in the eyes
of managers, irrespective of their actual abilities. In their view, if the uni-
versity is not going to send a “real professor,” the researcher should at least
have a PhD.9
Irrespective of such status issues, for a discursive, argumentative expert
interview with management representatives to achieve its goal, the inter-
viewer is indeed required to be an expert himself: the more an interviewer
demonstrates knowledgeability during the interview by giving competent
assessments, stating reasons, and raising counterarguments, the more man-
agers in turn will be willing to offer their own knowledge and take a stance
on issues, thus disclosing their subjective structures of relevance and pat-
terns of orientation in absence of strategic considerations.

9.6 Interviewing managers about “life-world” issues

In the case just discussed, the interview technique was adjusted within the
range of the typical (non-narrative) forms of everyday communication char-
acteristic of the company setting. In management studies aimed more at the
lifeworld of managers (as the study on “life arrangements of executives”)
it is perfectly possible – and depending on the specific research interest it
might even be imperative – to use the transition stage of the interview fol-
lowing the opening sequence in order to switch to a narrative form of inter-
viewing by providing a stimulus encouraging the interviewee to engage in a
more extensive impromptu narration. This may or may not be successful. In
the course of the said project, there regularly were cases (although in total
not a large number!) where the interviewees had difficulty in veering away
from their professional roles and the respective forms of communication
in spite of the fact that the focus of the research project was on life out-
side of the company. The project team had approached them as managers,
thus in their role as persons bearing certain responsibilities, and it proved
extremely difficult for them to let go of this role even after a more relaxed
atmosphere of conversation had been established. In some cases, the fact
that the interviews were conducted in a company environment may have
212 Rainer Trinczek

further aggravated the difficulty of freeing oneself from the typical modes
of in-house communication. These interviews largely remained locked into
the question-answer mode, and the fact that the interview addressed a topic
at the interface between company and private life did not provide an oppor-
tunity for switching to a discourse “free from the demands of action” either.
The respective mode of communication associated with this subject matter,
in these cases, does not appear to be a part of the repertoire of everyday
communication, which the researcher can easily draw upon in conducting
the interview.

9.7 Conclusions

In sum, we may state that there is no single ideal recipe for conducting inter-
views with managers. Rather, here too the general methodological principle
applies that the method employed ought to be adequate to the object of
research. Research methods, in this case interviews, must be adapted to the
specific modes of communication characteristic of the social setting that
they seek to address. This is the only way to effectively live up to the meth-
odological demand of qualitative research that the research process respect
the existing structures of everyday communication in the field.
The typical question-answer structure characteristic of the interview situ-
ation in the opening sequence of expert interviews with managers, as well
as the “argumentative, discursive” structure of interviewing in the second
stage of the interview, in the first study, and the more narrative form of
interviewing, in the other study, all meet this key standard of qualitative
research: On the one hand, such an interview strategy avoids violating
interviewee expectations of the interview; on the other hand, it allows to
align the research process with the prior structures of everyday communi-
cation such that the process of developing subjective structures of relevance
in the interview as they are operative in everyday life is supported to the
best possible extent.
In case of topics revolving around company matters, the “success” of
question-answer-based and “argumentative, discursive” interviewing respect-
ively is to no small part due to the fact that the forms of interviewing corres-
pond with the situation managers face in the company when, for instance,
the works council or their superiors question their positions and they are
required to justify them. For this reason, although the interview situation
compared to the everyday situations of company life can be considered to be
more “open” and there is in principle less need for “tactical” behaviour, “argu-
mentative, discursive” interviewing in thematically focused expert inter-
views is the appropriate research method for this setting. The same is true for
the more narrative structure characteristic of lifeworld communication and
the respectively adjusted mode of interviewing employed in the second case,
where the interview topic aimed more at life outside the company.
 How to Interview Managers? 213

The pronounced methodological and methodical disinterest in expert

interviews even among the advocates of the qualitative paradigm, which
this method belongs to, was the starting point of these considerations. The
disregard shown for the guided interview as a “dirty” method – and this
was our initial thesis – is rooted in the qualitative paradigm’s fetish that
interviewing must involve as little interviewer interference as possible. As
demonstrated above, this tenet is an untenable case of generalizing and
rendering as absolute rules of everyday communication related to a very
specific social context. Resulting insistence that the interviewer deny the
self or at least remain reticent has been shown to be methodological and
methodical nonsense in certain research settings. Precisely when we take
the basic principles of qualitative research seriously – for instance, the rule
that Hoffmann-Riem has described as the “key principle” of the interpretive
paradigm, which states “that the researcher generally only gains access to
data disclosing meaning when entering a communication relationship that
respects the research subject’s system of communication rules” (Hoffmann-
Riem, 1980, p. 346 f., own translation from German) – it is sometimes neces-
sary to take a more active role by structuring the interview and engaging
in argument while maintaining an open interview situation. It still holds
true that no single best solution in the choice of interview method can be
derived from abstract methodological considerations. There is only one
appropriate criterion for such choice: adequacy! Adequacy with regard to
the research object.

Notes
* This text has a longer history. It is based on a presentation given at a workshop con-
ducted by the Institut für Arbeitsmarkt- und Berufsforschung (Institute for Labor Market
and Vocational Research – IAB) on “Expert Interviews in Labor Market Research”
and was published accordingly in the report on the workshop (Trinczek, 1995). In
the meantime, I have gained new insights in a research project on “ ‘Executives’
Ways of Life in light of New Challenges in Professional and Private Life,” mainly
conducted by Cornelia Behnke and Renate Liebold at the Institute of Sociology
at the University of Erlangen, Germany, and sponsored by the German Research
Foundation (DFG). Today, I have come to take a somewhat different view from my
initial position in the mid-1990s. Hence, while major parts of the current article
draw on the initial article, certain sections differ considerably.
1. Lamnek (1989) provides an especially instructive example of the ambiguity in
classifying guided interviews in terms of the qualitative-quantitative universe
(a dichotomy which has frequently been criticized but nevertheless still structures
the debate). Concerning the similar focused interview, he writes, “Although the
focused interview falls into the category of quali-tative survey methods, it is closer
to quantitative methodology than are other qualitative methods” (Lamnek, 1989,
p. 80, own translation from German).
2. The “expert interview” is essentially an interview whose specific nature is deter-
mined by the fact that the interviewee qualifies as being an “expert.” In the strict
214 Rainer Trinczek

sense, the expert interview hence cannot claim to be a method of data collection
in its own right, since there are various different ways of interviewing experts.
Nevertheless, there exists a kind of “tacit consensus” in the research commu-
nity that an expert interview is a “guided interview,” and I will keep with this
consensus: in the following, I will discuss the guided expert interview (for a more
general discussion, see also Liebold and Trinczek, 2002).
3. This assessment is deliberately somewhat exaggerated to clearly state the point.
Without doubt, there are some very thoughtful articles that deal with various
aspects of the guided interview, for instance, by Christel Hopf (1978), Meuser and
Nagel (1989, 1994, 1997); they are, however, exceptions from the mainstream.
And the debate in the English-speaking world has been dominated by Merton and
Kendall’s text (1946), which is now more than 50 years old (see also Merton and
others, 1956).
4. Numerous interviews with managers – mostly from industry – conducted by
the project team “Labor and Industry Research” at the Institute of Sociology in
Erlangen and, more recently, by the Chair of Sociology at TU Munich in the con-
text of various projects on issues concerning working hours, problems regard-
ing the modernization of production, on various aspects of exchange relations in
companies, as well as issues concerning work-life balance provide the empirical
background and data basis for my exposition.
5. This, in my opinion, simplified view of potential communication situations in
everyday life can probably only be explained as the result of a gradual, unnoticed
process of equating everyday communication with communication in the con-
text of private life associated with the lifeworld. In the discourse of the qualitative
research community, the rules of communication identified in such settings have
been tacitly generalized as the universal rules of everyday communication as such.
6. Guided expert interviews are also frequently conducted simply to obtain infor-
mation that might otherwise not be available – for instance, because the research
touches upon issues that are thought to be problematic. For most advocates of
qualitative research, this type of “information mining” does not represent an
application that makes effective use of the strengths of their methods.
7. Welch and others (2002) also state that successful interviews with “company
elites” tend to succeed in “encouraging elite interviewees to regard the interview
as an intellectual discussion very different in nature to company meetings and
briefings.”
8. The differences between Kern and others’ experiences and my own may also be
related to the fact that the interviews involved different levels of company hier-
archy. It may be assumed that behaviour as observed by Kern and others is more
frequently encountered at the very top of a company hierarchy than among middle
management. For interviews with this group of managers see also Thomas 1993.
9. The situation is similar concerning the ascriptive characteristic “age.” It is easier
for a researcher in his mid-fifties to interview a plant manager of similar age than
it would be for a colleague in her early thirties. However, it seems that age, to a
certain extent, can be compensated by academic titles and status.

Further readings
Bloom, N., Krabbenhöft, K. and Lamba, N. (2006) LSE – University of Cambridge –
Stanford University – Management Interview Guide, http://www.stanford.edu/~
nbloom/ManagementInterviewGuide.pdf.
 How to Interview Managers? 215

Welch, C., Marschan-Piekkari, R., Penttinen, H. and Tahvanainen, M. (2002)

“Corporate elites as informants in qualitative international business research” in
International Business Research Review 11, 611–28.
Thomas Th. R. J. (1993) “Interviewing important people in big companies” in Journal
of Contemporary Ethnography 22, 80–96.

References
Behnke, C. and Liebold, R. (2000) “Zwischen Fraglosigkeit und Gleichheitsrhetorik –
Familie und Partnerschaft aus der Sicht beruflich erfolgreicher Männer” in
Feministische Studien 18, 64–77.
Cicourel, A.V. (1964) Method and Measurement in Sociology (New York: Free Press).
Ellguth, P., Liebold, R. and Trinczek, R. (1998) “ ‘Double-Squeeze’ – Manager zwischen
veränderten beruflichen und privaten Herausforderungen” in Kölner Zeitschrift für
Soziologie und Sozialpsychologie 50, 517–35.
Hoffmann-Riem, C. (1980) “Die Sozialforschung einer interpretativen Soziologie. Der
Datengewinn” in Kölner Zeitschrift für Soziologie und Sozialpsychologie 32, 339–72.
Hopf, C. (1978) “Die Pseudo-Exploration – Überlegungen zur Technik qualitativer
Interviews in der Sozialforschung” in Zeitschrift für Soziologie 7, 97–115.
Kern, B., Kern, H. and Schumann, M. (1988) “Industriesoziologie als Katharsis” in
Soziale Welt 39, 86–96.
Lamnek, S. (1989) Qualitative Sozialforschung, Bd.2: Methoden und Techniken (München:
Psychologie Verlags Union).
Liebold, R. (2001) “Meine Frau managt das ganze Leben zu Hause ...” Partnerschaft und
Familie aus der Sicht männlicher Führungskräfte (Wiesbaden: Westdeutscher Verlag).
Liebold, R. and Trinczek, R. (2002) “Experteninterview” in Kühl, S. and Strodtholz, P.
(eds) Methoden der Organisationsforschung. Ein Handbuch (Reinbek: Rowohlt),
pp. 33–71.
Merton, R. K., Fiske, M. and Kendall, P. L. (1956) The Focused Interview: a Manual of
Problems and Procedures (Glencoe, Illinois: The Free Press).
Merton, R. K. and Kendall, P. L. (1946) “The focused interview” in American Journal
of Sociology 51, 541–57.
Merton, R. K. and Kendall, P. L. (1979) “Das fokussierte Interview” in Hopf, C.
and Weingarten, E. (eds) Qualitative Sozialforschung (Stuttgart: Klett Cotta),
pp. 171–204.
Meuser, M. and Nagel, U. (1989) “Experteninterviews – vielfach erprobt, wenig
bedacht. Ein Beitrag zur qualitativen Methodendiskussion.” Arbeitspapier Nr. 6 des
Sonderforschungsbereichs 186 der Uni Bremen, Statuspassagen und Risikolagen im
Lebensverlauf’ Bremen.
Meuser, M. and Nagel, U. (1997) “Das ExpertInneninterview – Wissenssoziologische
Voraussetzungen und methodische Durchführung” in Friebertshäuser, B.
and Prengel, A. (eds) Handbuch Qualitative Forschungsmethoden in der
Erziehungswissenschaft (Weinheim/München: Juventa), pp. 481–91.
Schütze, F. (1977) Die Technik des narrativen Interviews in Interaktionsfeldstudien –
dargestellt an einem Projekt zur Erforschung von kommunale Machtstrukturen
(Bielefeld: Manuskript der Universität Bielefeld).
Schütze, F., Meinefeld, W., Springer, R. and Weymann, A. (1981) “Grundlagentheoretische
Voraussetzungen methodisch kontrollierten Fremdverstehens” in Arbeitsgruppe
Bielefelder Soziologen (ed.) Alltagswissen, Interaktion und gesellschaftliche Wirklichkeit,
5th edn (Opladen: Leske + Budrich), pp. 433–529.
216 Rainer Trinczek

Thomas, R. J. (1993) “Interviewing important people in big companies” in Journal of

Contemporary Ethnography 22, 80–96.
Trinczek, R. (1993) Management und betriebliche Mitbestimmung. Eine Typologie kolle-
ktiver Orientierungsmuster (Erlangen: Habil-Schrift).
Trinczek, R. (1995) “Experteninterviews mit Managern: Methodische und meth-
odologische Hintergründe” in Brinkmann, C., Deeke, A. and Völkel, B. (eds)
Experteninterviews in der Arbeitsmarktforschung. BeitrAB191 (Nürnberg: IAB),
pp. 59–67.
Welch, C., Marschan-Piekkari, R., Penttinen H. and Tahvanainen, M. (2002)
“Corporate elites as informants in qualitative international business research” in
International Business Research Review 11, 611–28.
10
Expert Interviews in Interpretive
Organizational Research
Ulrike Froschauer and Manfred Lueger

Reputable and scientifically founded research strategies adhere to certain

basic methodological assumptions impressed upon the analysis procedures
by the specifics of the actual area of research. In this way, the underlying
methodology regulates the entire research process and assumes a reporting
function that covers everything from the definition of relevant research
issues and the characteristics of any material collected or produced through
to the interpretive adaptation of the analysis results. Against this back-
ground, this article focuses (first line of argumentation) on how best to view
the subject of “the organization” from an interpretive social research meth-
odological perspective to (second line of argumentation) determine the
consequences and relevance of different types of expertise and procedural
specifics on the characteristics and applicability of expert interviews. In the
process, three types of expertise are differentiated, each with its own spe-
cific functions in the research process. A case study is then used to address
the systematic inclusion of expert interviews in interpretive research design
(third line of argumentation).

10.1 Interpretive organisational research

In broad terms, interpretative organizational research is based on two

lead perspectives: (1) constructionism, which ties knowledge of reality to
communication and the way people are anchored in separate collectives
(cf. Baecker and others, 1992, Gergen, 1994, Frindte, 1998), and (2) the
general understanding of social reality as a social construct (Berger and
Luckmann, 1981). It also embraces those methodological positions, which
focus on the joint construction of reality as a condition for the creation
of collective forms of action (see for example Berger and Luckmann, 1981,
Gergen, 2000, Lueger, 2001, Luckmann, 2006). Thus, organizations are, at
the same time, both condition and consequence of the social world: they
might set the parameters for the construction of concepts of reality and
thereby structure the way people act in organizations and their relevant

217
218 Ulrike Froschauer and Manfred Lueger

environments, but these actions themselves, in turn, also constitute real-

ity as a subject-independent phenomenon. In this respect – as summed up
by Weick’s (2003, p.190ff.) notion of enactment – organizations do more
that just observe their environments: by acting with regard to them, they
not only develop themselves, they also develop and actively shape their
environments as well.
The collective focus and individual capacity to act developed through
our dealings with a social environment and its particular requirements for
social cooperation. The people involved in the social internalisation pro-
cess in a specific organizational environment develop standardized social
interpretation schemes (cf. Fleck, 1981), which establish the context for the
coordination of collectively binding action. The actual interaction, that
is the interplay between the communicative acts in a social setting, helps
shape the experiences gained through interaction into a mutually access-
ible worldview for use in the joint management of action requirements.
Collective patterns of meaning are therefore by no means simply attitudes,
they are contexts of argumentation regarding mutual action problems
(Knoblauch, 2005, p. 178f.).
In this way, the inter-individual and interactive construction of reality
organizes the construction of subjective structures of meaning and know-
ledge into common organizational practices. Weick (1995) refers consist-
ently in this context to a process of sensemaking (“Sinngenerierung”) in
organizations, which enables people to act by adding coherence to the
world and thus creating both system and expectability. In addition to this
sensemaking on an individual level, the collective also produces interaction
patterns – objective, latent structures of meaning which escape the atten-
tion of the individual consciousness and cannot be intentionally accessed
by individuals. From an organizational theory perspective, this develop-
ment process not only focuses the collective construction of reality as a
condition for joint action, it is also where this construction of reality is
interpreted as an expression of the context behind the action.
Consequently, interpretive organizational analysis is not based on an
organization’s quasi “objective” given “reality,” it draws on the condi-
tions and substantiation of the inter-subjective construction of the
views of the organization shared by so-called “meaning collectives”
(“Deutungsmuster,” cf. Fleck, 1981) and analysed in reaction to object-
ive problems. Neither events nor actions make sense here as such, they
draw their meaning and relevance from being embedded in the require-
ments for action and from the available frameworks of meaning. Since,
as Goffmann (1975) also notes, these frameworks determine the analysis
approach used for a specific object and set out the possible contextual
horizon of meaning for the interpretation, the contextual sensemaking
process forms a key component of interpretive organizational analysis.
The core elements here are communication as a sensemaking process,
 Expert Interviews in Organizational Research 219

meaning as a form of order in experience and action and structuring as a

production of order (Froschauer, 2006, p. 194f.).

● Communication: according to Luhmann (1995, p. 147ff.) communication

unifies three selection levels, namely information, utterance and under-
standing. Information is not simply communicated, it must first be recon-
structed from the utterance which made it observable and introduced it
into the communication process. This is why the information gathered
from an utterance can differ from its original intent. Simply consider-
ing the manifest meaning of what is said and done does not suffice to
understand what happens in organizations: the context-bound nature of
linguistic meaning also requires an analysis of the social embeddedness
of messages.
● Meaning: to be able to coordinate interaction, meaning is required as a
form of order in human experience (Luhmann, 1982, p. 31) endowed in
the consciousness as reference parameter (Schütz and Luckmann, 1984,
p. 13). However, the intended meaning of the individual actors nonethe-
less remains inaccessible to research. Consequently, other forms of mean-
ing come to the fore in interpretive analysis: (a) the type of subjective
meaning intended by the actors (Weber, 1978), (b) the objective mean-
ing embedded in the phenomenon as structuring principle (Schütz, 1972,
p. 31ff., p. 132ff.) and (c) the practical meaning (Bourdieu 1995) expressed
through customary practices of interpretation and action in a particu-
lar social field, which acts as a bridge between the understanding of the
world (subjective meaning) and the world itself (objective meaning).
● Structuring: in general, the actors in social systems face the apparently
contradictory task of constructing a level of complexity that enables them
to both develop and reduce their knowledge of the environment and their
capacity to create meaningful order through action. This order is based on
the establishment of security in expectations (for example as a result of
the generalizability of behavioural expectations) and rules for interpret-
ing observations within a horizon of meaning. Since it is the structures
that define how actual subjects or objects are integrated in topological
and relational systems (Deleuze, 1992, p. 15f.) and these structures – as
possible ways of expressing phenomena – remain invisible, only the
actual bearers of expression material can be observed. These, in turn, are
based on objective latent structures of meaning as products of unobserv-
able yet formal rule system algorithms, deducible as standard rules from
the material forms of expression (Oevermann, 1993, p. 114ff.).

Interpretive organizational analysis therefore adheres to methodological

frameworks which link an appropriate understanding of organizations
to several considerations: a dialectic (re)production of reality and actual-
ity (as individual and collective notions) constructed on communicative
220 Ulrike Froschauer and Manfred Lueger

sensemaking processes whose latent structure manifests itself in the actions

and views of the actors involved. In this context, the methodological deter-
mination of expert knowledge must also consider the communicative basis
of organizational dynamics, the specific form of organizational sensemak-
ing and the particular system of structural organizational integration in all
their complexity.

10.2 Expert interviews in an interpretive

organizational research context

In the interpretive social research context, it quickly becomes clear that the
key to understanding organizations lies in understanding how they produce
order (and hence meaning). At the core of this issue lie the communicative
information creation processes embedded and assembled in lifeworlds and
organizational contexts as orderly wholes in the form of knowledge. This
means there are two basic components to organizational analysis: (1) famil-
iarity with the knowledge creation process, and (2) the content of the know-
ledge created, which must be stabilized and made available in some way.
Experts can provide a research process with valuable information on both
components. However, to enable them to do so, clarification is first needed
of where such relevant expertise or skills lie in an interpretive organiza-
tional analysis context.

10.2.1 Types of expertise

According to popular definition, experts are equipped with explicit specialist
knowledge gained through specific training which provides them with an
in-depth understanding of a particular topic or field and enables them to
provide clarification or resolve specific issues or problems (cf. Hitzler, 1994,
see also the Five Stage Model of Skill Acquisition in Dreyfus/Dreyfus, 2005).
They are also confronted with a number of expectations regarding their
knowledge that legitimize their expert status. For example, their know-
ledge must be different from everyday knowledge (otherwise it would not
be expert knowledge), is not accessible to everyone (otherwise we would
have no need for experts), is superior to everyday knowledge (otherwise we
could all rely on our own knowledge) and must avail itself of a theoretical
perspective (otherwise it would have no explanatory value). Consequently,
their expertise is characterized not only by the content of their knowledge
but also by a specific system of knowledge representation (cf. De Sombre
and Mieg, 2005, p. 59f.). Experts can thus provide assistance in situations
of uncertainty (for example in court) or help solve complex problems
(for example organizational consultants). To enable them to fulfil these
functions, their expert status must be recognized, backed up in general-
izable terms by a relevant qualification and communicated convincingly
in each individual situation (cf. Honer, 1994, p. 49ff.). Nonetheless, in an
 Expert Interviews in Organizational Research 221

interpretive organizational analysis context, such expertise can only con-

tribute to a limited extent towards understanding the way an organization
and its members work, because its relationship to the research object oper-
ates on a purely knowledge-based level. As a result, the notion of expertise
must be expanded further in this context.
Meuser and Nagel (1991, p. 442f.) consider participation in the actual area
of expertise a prerequisite for expert status, whereby such status is accorded
by the researchers in line with their field of research (relational expert
status). Accordingly, an expert is someone “who is responsible in some way
or another for the development, implementation or monitoring of a prob-
lem or who has privileged access to information about people or decision
processes” (Meuser and Nagel, 1991, p. 443, own translation). This results in
a twofold definition of expert status, namely through the according of expert-
ise by the researchers and the assumption of a disproportionate distribution of
knowledge seen as the sedimentation, storage and availability of privileged
experience.
Furthermore, according to Schütz and Luckmann (1973, p. 99ff.), lifeworld
knowledge is bound to the situatedness of a subject’s personal history and
is used to deal with everyday situations. Bourdieu takes this a step further
with his “habitus,” which serves as a matrix of perception, interpretation
and action and, as such, forms a generative principle of common practice
(Bourdieu, 2005, p. 78ff., 1995, p. 55). By this concept, human behaviour can
be aligned without direct communication by the homogenizing of human
habitus forms as a result of similar life circumstances and conditions. Thus,
although individual action is by no means purely subjective, it is not bound
exclusively to objective factors either. Instead, it follows the common sense
developed from real life experience. Knowledge is incorporated in the hab-
itus as a structured principle and serves generatively as a structure-giving
principle. This kind of knowledge is thus always practical and is linked not
in the consciousness but through practical activities.
To embrace these standpoints, empirical social sciences research must
focus primarily on the knowledge held by the people involved with an
organization or its relevant environments. For the interpretive research pro-
cess, this leads to a distinction between three types of expertise, each with
an increasingly abstractive degree of distance to practical knowledge of the
research field (subject matter knowledge).

● Subject matter know-how: this type of expertise is based primarily on

experience drawn from participating in activities in the field of research.
Hörning (2001) describes such people as “everyday experts.” Their
expertise is grounded to a great extent in direct first order observation
and generally stored as tacit knowledge in the way they perceive, think
and act (Polanyi, 1983, Oevermann, 2001). In the shaping of social prac-
tices based on this type of expertise, social differentiation can be seen
222 Ulrike Froschauer and Manfred Lueger

in different fields of activity and in the interplay between these fields.

Consequently, this type of expertise is essential to the development of
a deeper understanding of the research object and the creation of new
knowledge. This knowledge is extremely heterogeneous because the sedi-
mentation of experience is accompanied at a subjective level by the life-
world background and relevance structures.
● Reflective subject matter expertise: this kind of expertise rests on more far-
reaching primary and secondary experience and encompasses more than
“just” procedural knowledge or know-how. It is developed primarily by
those actors who depend in structural terms on second order observation
(to observe how others observe) and whose interactions systematically
cross internal and external field boundaries. Because people with reflect-
ive expertise act as organizational hubs, they are usually perforce mind-
ful observers of the organizational context who piece the various partial
perspectives together to form a whole. Thus, this type of knowledge is
more relational, reflective and abstract than concrete know-how. It can be
explicated above all with regard to the organization’s official view, but is
also subject to multiple thematic restraints.
● External expertise: this incorporates sound relevant theoretical knowledge
supported by secondary experience and second order observation and can
throw light on various aspects of a subject (both intra- and interdiscip-
linary). It is up to the researchers to integrate such reflective knowledge
into their research, where it can contribute to planning activities, provide
information on specific contextual assumptions or be used in the test-
ing, classifying or contrasting of results. External experts with system-
atic, abstract special knowledge can only be accorded marginal relevance
in interpretive organizational analysis because they lack practical know-
how and/or knowledge gained through experience. This kind of expertise
reproduces the knowledge already available and snatches control of the
topic away from the research.

Much of the expert knowledge relevant for interpretive organizational ana-

lysis is therefore derived from elements of the existing social knowledge base
imparted through observation, interaction or as part of the socialization
process (Schütz and Luckmann, 1973, p. 111ff.). The experiential knowledge
used in organizations is thus socially acquired knowledge, which reflects the
internal differences between groups of people whose actions are based on
internally comparable yet externally different situations. The corresponding
structures of meaning are both the reason for and consequence of internal
cooperation relationships and demarcations and represent the actual distri-
bution of knowledge in the organization. The holders of such subject matter
know-how are not allocated a “lay” status in the research context because
it is not their everyday knowledge that is relevant but rather the specialist
knowledge they require to act successfully in a specific organizational field.
 Expert Interviews in Organizational Research 223

In contrast, the distribution of the primarily scientific knowledge held by

external experts is based on the institutionalized acquisition of skills encoun-
tered in the academic world (that is inter- and intra-disciplinary, by the-
matic focus, and so on). This incorporates a basic set of skills – textbook
knowledge – which represent a relatively homogeneous basis of generally
accepted views. The underlying collective of meaning is found in the cor-
responding discipline or sub-discipline (where accreditation rules are also
defined) but only bears a limited relation to the actual organization under
analysis. One such particular form of knowledge is research expertise, which
focuses on the methodology behind the planning, organizing and carrying
out of research processes and deals with the systematic process of obtaining
contextual knowledge of a research field. It concentrates on second order
social sciences (re)construction based on the reconstruction of a social field
in which expert knowledge and experts appear as everyday phenomena
(“Experts for us”; cf. Aman, 1994, p. 35). This type of expertise is not fac-
tored into the subsequent discussion, because it refers to methodological
procedures not contextual knowledge generation.
All three types can also be further differentiated according to specializa-
tion, where the focus lies on the application of specific expertise to a spe-
cific problem (cf. Hitzler, 1994, p. 22f.). Since the particular competences
involved serve to resolve specific functional issues, this type of knowledge
is of limited relevance in understanding the organization under analysis:
external specialists rarely have an overview of the entire subject, and their
knowledge is thus only of limited use for establishing an overview and con-
trasting results; while internal specialists can provide important informa-
tion on specific issues, this usually concerns the manifest content level, not
the level of latent sensemaking that is of far greater importance in under-
standing organizational dynamics. Nonetheless, it may prove pertinent to
the research (for example from a specific evaluation or interpretation per-
spective) to also integrate specialists in the course of the project.

10.2.2 Procedural consequences for expert interviews

Different types of expertise place different requirements on survey and
interpretation procedures. The following section examines the relevance of
the three types of expertise described above for various interview and evalu-
ation techniques and methods.

● Subject matter know-how: this assumes a central position in interpretive

organizational analysis. Since this type of knowledge is linked to indi-
viduals or groups of individuals, is activated in specific situations and can
only be explicated (unconsciously) to a limited extent, it cannot be made
topical by means of exact questioning in the interview setting but must
instead be largely reconstructed from the form of knowledge representa-
tion. Since it is not so much the manifest content but rather the form of
224 Ulrike Froschauer and Manfred Lueger

expression used to describe organizational practices that is important in

interviews of this kind, open, narrative interview forms prove particularly
suitable for bringing this kind of expertise to the fore (see for example
Schütze, 1977, Glinka, 1998, Froschauer and Lueger, 2003, Holstein and
Gubrium, 2003, Küsters, 2006). The freedom afforded by this type of inter-
view technique forces interviewees to structure their information, while
this structure, in turn, serves as an expression of organizational dynam-
ics and offers important clues for understanding the organization. Group
interviews are also a good method of revealing the dynamics of com-
munication in organizations (Froschauer and Lueger, 2003, p. 55ff.) and
generally focus on the day-to-day situation and practices in organizations
and their environment. As far as the interpretation of results is con-
cerned, methods suited to the analysis of latent structures of meaning,
such as those applied in social sciences related hermeneutics (cf. Hitzler
and Honer, 1997, for variations on objective hermeneutics cf. Oevermann
and others, 1979, Oevermann, 1984, Lueger, 2000, Froschauer and Lueger,
2003) are also suitable here. To avoid receiving one-sided views, consid-
eration must also be given to the organization’s social complexity in the
selection of appropriate experts.
● Reflective subject matter expertise: this follows on seamlessly from the
above, with one key difference, namely a shift in emphasis away from the
actual work situation towards a stronger focus on the internal relation-
ships between the different organizational units and/or the organization
and its environment.
● External expertise: contrary to the information found in scientific literature,
interviews with this group can focus on the concrete topic of research and
thus bring otherwise disjointed background information into the equa-
tion. Experts of this kind are not particularly difficult to interview and,
because it is centred on explicable, explicit knowledge, the information
they provide is not difficult to interpret either. A good tactic here is to
let their expertise come to the fore of its own accord in the course of
the interview and then use targeted questions to obtain greater depth of
information and establish any limits. In the analysis phase, the specific
issues addressed are then condensed, elaborated and put into a systematic
context using appropriate condensing techniques (such as the qualitative
content analysis found in Mayring 2003).

To gain access to latent structures of meaning, rigorous interpretive

organizational analysis must always consider the way people differentiate
the situated practices of action collectives within the actual organization.
Researchers cannot rely here on external experts, because their knowledge
has already been homogenized in line with generalizable principles and,
thus, cannot be used to shed appropriate light on specific individual situ-
ations. Similarly, the information that can be drawn from an organization’s
 Expert Interviews in Organizational Research 225

formal structure is also of limited use, since it only represents the surface
beneath which different ways of regulating action and cooperation can
establish themselves.
To uncover this differentiation in the course of the organizational ana-
lysis, research must be carried out in cycles, with each cycle integrating both
survey and interpretation phases (Lueger, 2001, p. 363ff., Froschauer and
Lueger, 2006, p. 254ff.). In this sense, the progressive insights obtained indi-
cate the path for further surveys. The basic strategy behind this approach
echoes two of the principles of theoretical sampling (Glaser and Strauss,
1967, p. 45ff.): maximum structural variation and minimization of differences.
Maximum structural variation seeks out extreme cases and possible anom-
alies to demarcate the internal setting and identify general characteristics
pertaining to the research field (scope and generalizability of conclusions),
while a minimization of differences compares prior interpretations with
similar cases to identify unclarity in the argumentation. Such sampling
strategies continue with survey and interpretation activities until the “the-
oretical saturation” point is reached and the inclusion of additional new
data can no longer be expected to contribute further to theory development
(Glaser and Strauss, 1967, p. 61f.).
In general, expert interviews of this kind generally require a high degree
of flexibility to provoke a level of structuring that takes into consideration a
range of different aspects. This flexibility extends across the entire research
process: from the selection of content central to the general area of interest
(for example to explicate the full breadth of the research area with regard
to the specific knowledge interests in the case of external experts) to the
choice of interview candidates (for example to uncover everyday working
relationships in the case of subject matter experts), the way interviews are
started (for example with a clarification of the research topic in the case of
external experts, by focusing on activating organizational knowledge and
establishing links to concrete lifeworld experience in the case of subject
matter experts and by the introduction of an organizational context rele-
vant to all participants in the case of group interviews), the adopting of an
open approach to individual interviews (as a general rule) and the sparing
use of moderation in group interviews (to reveal relationship dynamics, in
particular in the case of subject matter experts). By incorporating interven-
tions into the composition of such expert interviews, organizations reveal
their way of dealing with such “disruptions” or demands and thus divulge
specific internal modus operandi and management principles that are diffi-
cult to tap into by other means.
Open interviews encourage interviewees (experts) to talk to outsiders
(researchers) about their organizational lifeworld. Researchers define them-
selves as the “learners” in such settings, explicitly allocate the role of the
expert to the interviewee(s) and later orient themselves on the information
provided. For the analysis, the interview process embodies the representation
226 Ulrike Froschauer and Manfred Lueger

strategies used in the organization, strategies which can, however, only be

disclosed in an intensive analysis of the dynamics of interaction. These
dynamics are more prevalent in group interviews, where expectations
and established patterns of behaviour gain relevance and depict (at least
in part) the communication structures found in actual organizational
relationships.
In all cases, it is always useful to determine and clarify any areas that
were not mentioned in the open interviews yet, given the assumed expert-
ise, would also appear to play a role. Subsequent interviews can then focus
on such aspects and explore such open issues in greater depth. However, in
particular in the case of interviews with experts with subject matter know-
how, priority should not be given to the use of guided interviews (thereby
pushing the content dimension to the fore): the actual expertise in question
is not reflective knowledge but rather practical know-how which can only
be explicated to a limited extent. A further risk with guided interviews (in
particular in combination with an assessment of the manifest content) is
that the information provided will offer a retrospective, rationalized view
of organizational phenomena which bears no relation to the interviewee’s
relevance system and provides only limited insight into the social dynamics
of the field of research.
Differences can also be determined with regard to interpretation: the ana-
lysis of interviews with external experts focuses primarily on manifest con-
tent (confirming the need to cover as many areas of reference and factors of
influences as possible), while with subject matter experts it centres on latent
meaning content. The expertise attributed to experts with organizational
know-how can be seen as a particularly good expression of organizational
sensemaking processes.
Since hermeneutic and reconstructive interpretation methods are suscep-
tible to subsumptive interpretation, quality assurance strategies that fit both
the requirements of the interpretation framework and the actual interpret-
ation method are also required (cf. Lueger, 2001).

● Interpretation framework requirements: to avoid researcher know-how affect-

ing results, it is a good idea to keep the interpretation and survey proc-
esses separate; team-based interpretation offers a good way of raising the
level of the argumentation and wealth of possible readings; the interpret-
ation should not be subject to time or results constraints to give it depth
and avoid premature conclusions.
● Interpretation method: appropriate strategies here include deconstruction
to avoid any premature use of available knowledge, sequential analysis to
explicate the generative structure of meaning, comparative constitution
of meaning to contrast alternative assumptions, an emphasis on verifying
the analytical viability of the interpretation as well as a careful examin-
ation of any anomalies.
 Expert Interviews in Organizational Research 227

10.3 Expert interviews and organisational

analysis in practice: a case study

The following section demonstrates the practical use of expert interviews

in interpretive organizational analysis. The actual case study described
(cf. Lueger, 1997) sits at the crossroads between autonomous and contracted
research and was deliberately chosen for its ability to illustrate the appropri-
ateness of a research-based approach in interpretive organizational research.
The case study is described in accordance with the key research phases
applied in expert interviews.

10.3.1 Negotiation of research framework

The knowledge interests behind the study were both theoretical and prac-
tical. Firstly, and from a research perspective, it sought to examine the basic
theoretical question of the conditions and possibilities of social order in
organizations in the context of alternative ways of seeing and doing things.
Consequently, it aimed to analyse communication relationships, examine
the specific social constructions of reality in an organizational context and
study the communication control mechanisms, which create, maintain or
change social order in an organization.
During the search for a suitable target (that is organization), meetings
were held with the Director of a major manufacturing company with his
own very specific interest: his company was – as he saw it – struggling with
management-workforce communication problems. The workforce no longer
trusted its management, and this situation had led to severe difficulties in
communication as well as conflicts and friction. Any attempts to organize
information in a better manner and provide more information had only
served to aggravate the situation and make the relationship between the
management and the workforce even worse. At that time, the company was
planning to develop a new communication strategy that would be accept-
able to the workforce and could be used to provide information that would
be accepted and considered reliable.
This immediately opened up one of the basic problems of interpretive
organizational research: accepting the Director’s viewpoint would, in turn,
mean (a) accepting the problem as defined by management, (b) assuming
the “communication problem” to be purely technical, and (c) condoning
the proposed solution (an improved communication strategy). This is an
unproductive approach from a social sciences perspective because it focuses
entirely on external expertise and reveals nothing about the relevance of
the “communication problem,” the views of the workforce, the reasons why
the problem emerged in the first place or indeed why it continued to be a
problem. This would be tantamount to a one-sided instrumentalisation of
the research by the management (as the client contracting the research).
Such an approach would also be of limited benefit to the company as a
228 Ulrike Froschauer and Manfred Lueger

whole as it contributes nothing towards a long-term understanding or reso-

lution of the already identified problem.
It thus became clear that a joint definition of the commissioned research
was required – one which redefined the starting point for the “problem”
and granted the research team sufficient leeway to study all aspects of the
phenomenon. The joint definition that was subsequently agreed upon rede-
fined the problem in accordance with the interests of both sides (research
and organization): the project would seek to uncover the organizational
dynamics behind the actual specifics of internal communication. In this way, the
research was relieved of the political function of delivering and legitimiz-
ing a particular solution to a predefined problem. It would “simply” seek to
understand why the problem was indeed a problem, who actually saw it as a
problem, where the possible causes lay and, thus, identify potential starting
points for resolving the situation properly.
Consequently, the study had to have access to those people with know-
ledge of the subject matter in hand. To do so, the researchers had to be able
to interview all members of staff (with their prior agreement) without inter-
vention on the part of management in the way the interviews were organ-
ized. From a research perspective, all members of staff are considered equal
interview partners regardless of their hierarchical position in the organiza-
tion and differ only in their specific area(s) of expertise.
Interpretive social research assumes the research results to be dependent
not only on the analysed object, but also on the approach taken by the
research (cf. Lueger, 2001, p. 272ff.). The positioning of the research in the
organization is a particularly delicate point, because it influences the way
people see the researchers (for example as emissaries of the Board) and, as a
result, can have lasting effects on the research work itself (for example will-
ingness to cooperate, atmosphere in interviews). Consequently, and given
the conflict situation between the management and the workforce, the fol-
lowing three positioning elements were used in this particular study.

● A member of the organization was sought to act as coordinator between

the research team and the organization. The person chosen should fulfill
three requirements: he or she should (a) occupy a position in the organ-
ization that had limited links to management (decoupling of research from
management), (b) have at least a rough overview of the organization and
relatively good contacts to as many divisions as possible (reflective subject
matter expertise), and (c) not be involved in the current lines of conflict
(internal organizational acceptance). The coordinator chosen acted as a cen-
tral point of contact for the research team, negotiated research decisions
and organized all research activities at an internal level.
● Everyone who participated in the study was given a guarantee that the
information they provided would be treated in absolute confidence and
that the results of the analysis would be based solely on organizational
 Expert Interviews in Organizational Research 229

structures. This guarantee of anonymity was also extended to the organ-

ization as a whole (for reasons of competition).
● To increase trust, it was also agreed that all members of staff would be
informed of the results (feedback) to both prevent one-sided access and
generate transparency about the project within the organization.

10.3.2 Selection of experts and

stimulation of structuring processes
The negotiations regarding the scope of the research served as an initial source of
information, because the view of the problem and ideas about possible results
or problem-solving strategies provided offered important insights into
internal organizational processes. These meetings centred on the relation-
ships between the different operations around the world and the internal
divisions, with a focus on the development of the company as a whole, its
business activities, corporate strategies and the specific problem in hand
(reflective subject matter expertise from a management perspective).
The meetings with the coordinator (as subject matter expert) served as the
information basis for the implementation of the project. The first of these meet-
ings fulfilled three functions: (1) to discuss the proposed project in detail,
(2) to provide initial orientation regarding the organizational structure,
internal differentiations and peculiarities, and (3) to agree on how the actual
work would be carried out (selection of possible interview candidates). In the
process, a number of other topics were also addressed, such as the issues to
look out for and who best to contact to obtain a better understanding of the
way things worked in the organization. Reflective subject matter knowledge
also played a key role in subsequent meetings with the coordinator although
the focus from then on lay on the relationships between the organizational
units. The analysis of these meetings served to establish an initial level of
understanding and paved the way for the interviews that were to follow.
Based on these analysis results, a procedure and the specific criteria (sub-
ject matter know-how and reflective expertise) for selecting participants for
the individual interview groups were negotiated with the coordinator in line
with the current status of analysis. Examples of know-how criteria included
contacts in a given field or direct cooperation relationships. To activate
reflective subject matter expertise, individual and group interviews were
held with people who acted as points of contact either within the organiza-
tion or to its relevant environments. The intention here was to incorporate
internal organizational relationship structures in the make-up of the inter-
view groups. Consequently, and to reconstruct communication relation-
ships in the organization, interviews began with a description of how the
individual participants had been chosen for that group.
Pursuant to the theoretical sampling method (Glaser and Strauss, 1967,
p. 45ff.), the interviews conducted in the course of the analysis sought to
activate the kind of sensitive subject matter knowledge required to provide
230 Ulrike Froschauer and Manfred Lueger

an understanding of actual organizational dynamics. For example, in the

initial research cycle, interviews were held either with members of staff in
operations which had only recently joined the group (outsiders), with mem-
bers of staff in established core corporate areas (company traditions), with
members of staff in executive positions (to determine the mood) or with
particularly exposed members of the management team (conflict positions).
In accordance with the insights gained from reflection on the status of the
analysis following the first interview and interpretation phase, a second
research cycle was introduced in which interview groups were put together
with the primary purpose of expanding the initial differentiation and val-
idating the reliability of the views already collected. Since the interpretation
had revealed the central role played by two conflicting corporate philoso-
phies, the interview groups in this cycle all contained representatives of
both corporate philosophies. They also included some members of staff who
were directly affected by the conflict and others who had kept a neutral
distance. As a result of the internal communication barriers between some
divisions, all the interviewees had access to highly specialized know-how
and reflective knowledge.
Since the analysis had uncovered two areas of context of particular rele-
vance to understanding this case, external expertise was also sought: a meet-
ing with an industry expert (to find out more about industry developments)
and a meeting with a finance expert (since financial issues also played an
enormous part in the conflict).

10.4 Expected results of this procedure

The above discussion underlines the central requirements placed on the car-
rying out and analysis of expert interviews in an interpretive research-based
approach to organizational analysis. These can be summarized as follows:

● If the analysis seriously considers the requirements of interpretive social

research, reliance on external expertise is completely inadequate.
● The assumption of a disproportionate distribution of knowledge forms the
starting point for differentiating between different types of expertise.
● These types of expertise differ in the knowledge structures they represent,
with practical subject matter experience providing key input for the
research.
● Specific data collection and interpretation procedures and techniques are
required to gain access to the different types of expertise.
● The performance capabilities of such expert interviews only develop in
the context of an appropriately organized research project.

Expert interviews can be used to address two important areas of know-

ledge. First and foremost, they provide communicative access to the field
 Expert Interviews in Organizational Research 231

of research, whereby descriptions and reasons serve as an expression of the

underlying selection mechanisms in the organization. In addition, in their
role as subject matter experts, the members of the organization find them-
selves confronted with specific research settings in which they themselves
have to take the initiative. By assuming at least partial responsibility for the
organization of the research process, they demonstrate their specific compe-
tences as a special form of expertise.
The final remaining step is to clarify which types of knowledge can be
accessed by a methodological procedure of this kind. The answer to this
question is illustrated below on three levels using the case study described
above:

● This study revealed that the company’s original intention of developing an

elaborate information strategy would have started out from a completely
wrong point. Since the workforce no longer had any trust in company
management, any initiatives automatically took place in an organiza-
tional setting which interpreted all attempts by management to improve
the flow of information as negative. However, the study was able to show
how the lack of trust, which formed the background to the problem had
arisen and identify the events and actions that maintained and stabilized
this situation (starting points for solving the problem).
● A number of important events occurred on a scientific level. For example,
the analysis showed that the everyday practical interpretation of events in
the company had gathered such a momentum of its own that the people
involved could almost no longer control it. The polarization of two dif-
ferent views of the company that had formed in the course of its develop-
ment and created two contextually different horizons of meaning meant
reaching an understanding was almost impossible in some areas. That
explained why entire divisions effectively blocked each other in their
daily work, although they were each dedicated to contributing to the
company’s future success. Furthermore, the study clearly demonstrated
the blocking effect of long-term success on development. In this context,
important insights were gained into organizational learning, revealing a
blind spot in similar studies, namely the building of defence strategies as
an important protection mechanism on the part of the workforce.
● Methodological insights regarding the implementation of such analyses
were also gained. For example, the dynamics of the research processes
showed that the conditions for access in areas of conflict area changed as
the study progressed. At the start of this study, they were characterized
by defensive attitudes and reproduced the exclusion strategy common
throughout the company. With the development of the first partly trust-
ing relationship, this mistrust suddenly changed into great openness.
Two aspects were instrumental in this change: researchers basically work
across all borders in the organization and quickly come to know more
232 Ulrike Froschauer and Manfred Lueger

(or other things) about certain (predominantly ‘cross-border’) events in

the company than the interviewees themselves. Agreeing to participate
in an interview created pressure on the interviewees to be frank, because
although they could not know what the researchers already knew or didn’t
know, they could at least assume that they already had access to a wealth
of information.

In general, a key element in the classification of different expertise is the

variation in perspectives: contrasting various horizons of meaning for organ-
izational knowledge and constructing images of normality for different col-
lectives reveals the interaction between very different strategies of action.
A systematic differentiation into various areas of subject matter know-how
and reflective expertise provides understanding of the dynamics of mutual
coordination. Doing so requires a move away from the traditional view of
expertise (and its focus on special cognitive knowledge) towards a stronger
focus on perceptive competences and know-how in form of practical
expertise. This requires data collection and interpretation methods designed
to provide a systematic reconstruction of latent structures of meaning.

Further readings
Holstein, J. A. and Gubrium, J. F. (2003) Inside Interviewing. New Lenses, New Concerns
(Thousand Oaks: Sage).
Kimberley, D. E. (2005) Qualitative Organizational Research (Grennwich: Information
Age Publisher).
Riessman, C. K. (2008) Narrative Methods für the Human Sciences (Los Angeles: Sage).

References
Amann, K. (1994) “Guck mal, Du Experte. Wissenschaftliche Expertise unter
ethnographischer Beobachtung und wissenssoziologischer Rekonstruktion” in
Hitzler, R., Honer, A. and Maeder, C. (eds) Expertenwissen. Die institutionalisierte
Kompetenz zur Konstruktion von Wirklichkeit (Opladen: Westdeutscher Verlag),
pp. 32–43.
Baecker, J., Borg-Laufs, M., Duda, L. and Matthies, E. (1992) “Sozialer Konstruktivisimus –
eine neue Perspektive in der Psychologie” in Schmidt, S. J. (ed.) Kognition und
Gesellschaft. Der Diskurs des Radikalen Konstruktivismus 2 (Frankfurt am Main:
Suhrkamp), pp. 116–45.
Berger, P. L. and Luckmann, T. (1981) The Social Construction of Reality. A Treatise in the
Sociology of Knowledge (Harmondsworth: Penguin Books).
Bourdieu, P. (2005) Outline of a Theory of Practice, 19th printing (Cambridge:
Cambridge University Press).
Bourdieu, P. (1995) The Logic of Practice (Stanford: Stanford University Press).
Deleuze, G. (1992) Woran erkennt man den Strukturalismus? (Berlin: Merve-Verlag).
De Sombre, S. and Mieg, H. (2005) “Professionelles Handeln aus der Perspektive einer
Kognitiven Professionssoziologie” in Pfadenhauer, M. (ed.) Professionelles Handeln,
1st edn (Wiesbaden: VS Verlag für Sozialwissenschaften), pp. 55–66.
 Expert Interviews in Organizational Research 233

Dreyfus, H. L. and Dreyfus, S. E. (2005) “Expertise in Real World Contexts” in

Organization Studies 26(5), 779–92.
Fleck, L. (1981) Genesis and Development of a Scientific Fact (Chicago: University of
Chicago Press).
Frindte, W. (1998) Soziale Konstruktionen. Sozialpsychologische Vorlesungen (Opladen:
Westdeutscher Verlag).
Froschauer, U. (2006) “Veränderungsdynamik in Organisationen” in Tänzler, D.,
Knoblauch, H. and Soeffner, H. G. (eds) Zur Kritik der Wissensgesellschaft (Konstanz:
UVK Verlagsgesellschaft), pp. 189–215.
Froschauer, U. and Lueger, M. (2003) Das qualitative Interview. Zur Praxis interpretativer
Analyse sozialer Systeme. (Wien: WUV-Universitätsverlag).
Froschauer, U. and Lueger, M. (2006) “Qualitative Organisationsdiagnose als
Grundlage für Interventionen und als Intervention” in Frank, H. (ed.) Corporate
Entrepreneurship (Wien: WUV), pp. 233–87.
Gergen, K. J. (1994) Realities and Relationships: Soundings in Social Construction
(Cambrigde: Harvard University Press).
Gergen, K. J. (2000) An Invitation to Social Construction (London: Sage).
Glaser, B. G. and Strauss, A. L. (1967) The Discovery of Grounded Theory (New York: de Gruyter).
Glinka, H. -J. (1998) Das narrative Interview. Eine Einführung für Sozialpädagogen
(Weinheim: Juventa-Verlag).
Goffman, E. (1975) Frame Analysis: An Essay on the Organization of Experience, 2nd
printing (Cambridge: Harvard University Press).
Hitzler, R. (1994) “Wissen und Wesen des Experten. Ein Annäherungsversuch – zur
Einleitung” in Hitzler, R., Honer, A. and Maeder, C. (eds) Expertenwissen. Die insti-
tutionalisierte Kompetenz zur Konstruktion von Wirklichkeit (Opladen: Westdeutscher
Verlag), pp. 13–30.
Hitzler, R. and Honer, A. (eds) (1997) Sozialwissenschaftliche Hermeneutik. Eine
Einführung (Opladen: Leske & Budrich).
Holstein, J. A. and Gubrium, J. F. (2003) “Active Interviewing” in Holstein, J. A. and
Gubrium, J. F. (eds) Postmodern Interviewing (London: Sage), pp. 67–80.
Honer, A. (1994) “Die Produktion von Geduld und Vertrauen. Zur audiovisuellen
Selbstdarstellung des Fortpflanzungsexperten” in Hitzler, R., Honer, A. and
Maeder, C. (eds) Expertenwissen. Die institutionalisierte Kompetenz zur Konstruktion
von Wirklichkeit (Opladen: Westdeutscher Verlag), pp. 44–61.
Hörning, K. H. (2001) Experten des Alltags. Die Wiederentdeckung des praktischen
Wissens, 1st edn (Weilerswist: Velbrück Wissenschaft).
Knoblauch, H. (2005) Wissenssoziologie (Konstanz: UVK Verlagsgesellschaft).
Küsters, I. (2006) Narrative Interviews. Grundlagen und Anwendungen, 1st edn
(Wiesbaden: VS Verlag für Sozialwissenschaften).
Luckmann, T. (2006) “Die kommunikative Konstruktion der Wirklichkeit”
in Tänzler, D., Knoblauch, H and Soeffner, H. G. (eds) Neue Perspektiven der
Wissenssoziologie (Konstanz: UVK Verlagsgesellschaft), pp. 15–26.
Lueger, M. (1997) “Unternehmenserfolg als Hindernis. Zur sozialen Logik von
Entwicklungsbarrieren” in Journal für Betriebswirtschaft 47(5), 248–65.
Lueger, M. (2000) Grundlagen qualitativer Feldforschung. Methodologie – Organisierung –
Materialanalyse (Wien: WUV Universitätsverlag).
Lueger, M. (2001) Auf den Spuren der sozialen Welt. Methodologie und Organisierung inter-
pretativer Sozialforschung (Frankfurt am Main: Lang).
Luhmann, N. (1982) “Sinn als Grundbegriff der Soziologie” in Habermas, J. and
Luhmann, N. (eds) Theorie der Gesellschaft oder Sozialtechnologie. Was leistet die
Systemforschung? (Frankfurt am Main: Suhrkamp), pp. 25–100.
234 Ulrike Froschauer and Manfred Lueger

Luhmann, N. (1995) Social Systems (Stanford: Stanford University Press).

Mayring, P. (2003) Qualitative Inhaltsanalyse. Grundlagen und Techniken, 8th edn
(Weinheim: Beltz).
Meuser, M. and Nagel, U. (1991) “ExpertInneninterviews – vielfach erprobt, wenig
bedacht. Ein Beitrag zur qualitativen Methodendiskussion” in Garz, D. and
Kraimer, K. (eds) Qualitativ-empirische Sozialforschung. Konzepte, Methoden, Analysen
(Opladen: Westdeutscher Verlag), pp. 441–71.
Oevermann, U. (1984) “Zur Sache. Die Bedeutung von Adornos methodologischem
Selbstverständnis für die Begründung einer materialen soziologischen
Strukturanalyse” in Friedeburg, L. von and Habermas, J. (eds) Adorno-Konferenz
1983 (Frankfurt am Main: Suhrkamp), pp. 234–89.
Oevermann, U. (1993) “Die objektive Hermeneutik als unverzichtbare Größe für
die Analyse von Subjektivität. Zugleich eine Kritik der Tiefenhermeneutik” in
Jung, T. and Müller-Doohm, S. (eds) Wirklichkeit” im Deutungsprozess. Verstehen
und Methoden in den Kultur- und Sozialwissenschaften, 1st edn (Frankfurt am Main:
Suhrkamp), pp. 106–89.
Oevermann, U. (2001) “Die Struktur sozialer Deutungsmuster” in sozialersinn.
Zeitschrift für hermeneutische Sozialforschung 1, 35–81.
Oevermann, U., Allert, T., Konau, E. and Krambeck, J. (1979) “Die Methodologie einer
‘objektiven Hermeneutik’ und ihre allgemeine forschungslogische Bedeutung in
den Sozialwissenschaften” in Soeffner, H. G. (ed.) Interpretative Verfahren in den
Sozial- und Textwissenschaften (Stuttgart: Metzler), pp. 352–434.
Polanyi, M. (1983) The Tacit Dimension (Gloucester: Peter Smith).
Schütz, A. (1972) The Phenomenology of the Social World (Evanston: Northwestern
University Press).
Schütz, A. and Luckmann, T. (1973) The Structures of the Life-World (Evanston:
Northwestern University Press).
Schütz, A. and Luckmann, T. (1984) Strukturen der Lebenswelt, 2nd volume (Frankfurt
am Main: Suhrkamp).
Schütze, F. (1977) Die Technik des narrativen Interviews in Interaktionsfeldstudien –
dargestellt an einem Projekt zur Erforschung kommunaler Machtstrukturen (Bielefeld:
manuscript unpublished).
Weber, M. (1978) Economy and Society: An outline of Interpretative Sociology (Berkeley:
University of California Press).
Weick, K. E. (1995) Sensemaking in Organizations (London: Sage).
Weick, K. E. (2003) “Enacting an Environment. The Infrastructure of Organizing” in
Westwood, R. I. and Clegg, S. R. (eds) Debating Organization: Point-Counterpoint in
Organization Studies (Malden: Blackwell), pp. 184–94.
11
Between Scientific Standards
and Claims to Efficiency: Expert
Interviews in Programme Evaluation
Angela Wroblewski and Andrea Leitner

In the past 20 years evaluation has been established in the German-speaking

area as an integral part of social policy, particularly in connection with the
development and implementation of social policy programmes. This devel-
opment was strongly influenced by EU policy, specifically by the “open
method of coordination” and the precept to evaluate all programmes co-
financed by EU money. In most European countries this resulted in a boost
to professionalization, which had already set in decades earlier in the USA1
In the meantime, in the field of evaluation of social policy programmes a
state-of-the-art has developed in the German-speaking area that does receive
Anglo-American research but which simultaneously foregrounds participa-
tory and responsive practices as antithesis to quantitatively oriented effect
analysis.
Especially in this context expert interviews play an important role for
generating information and validating hypotheses. This has several reasons:
first, due to a lack of other data sources or for reasons of efficiency certain
questions can only be answered by involving experts. Second, interviews
with different players make it possible to take differing points of view and
perspectives into account and to reveal possible areas of conflict. Third,
access to information in the evaluation process is made easier and accept-
ance of the results is increased. Finally, expert interviews offer the possibil-
ity to reflect on experiences with implementation, thus contributing to the
further development of programmes.
Even though practically no evaluation can make do without expert inter-
views, the question arises to what extent generally valid rules for their per-
formance and analysis can be drawn up for this method (Flick, 2006, Deeke,
1995). These doubts in a “methodology” of expert interviews in the practice
of programme evaluation result, among other things, from the variety of
different approaches to and methods of evaluation. This variety developed
due to the increased use of evaluations in the private as well as the public

235
236 Angela Wroblewski and Andrea Leitner

domain, in which expert interviews carry different respective weight. It

would therefore be amiss to present one particular approach or a limited
spectre of approaches as the “ideal way” in evaluation practice in general
and thus for expert interviews in evaluation research. At the same time,
abiding by certain standards of acquiring information seems essential to
put evaluations, which are usually located in the vicinity of political con-
sulting, on a methodological basis that makes scientific practice intersub-
jectively verifiable and thus easier to control.
Despite the variety in approaches to evaluation, expert interviews in
evaluations generally display special characteristics which are of lesser sig-
nificance in other research contexts: the experts usually are also partici-
pants in the intervention to be evaluated, so-called stakeholders2; from this
they have a special interest in the results of the investigation. This causes
a specific interaction process in which the evaluators are often perceived
as controlling bodies. This may not only lead to distortions in response
behaviour – certain aspects are made out to be more positive or negative
than they actually are – but may in particular compromise the willingness
to provide information. On the other hand, evaluations mostly are com-
missioned projects, in which the clients not only (co)determine the subject
matter of the evaluation content-wise by formulating the questions and the
objective of the evaluation. Clients also influence the choice of method(s)
through their interest in fast results and the resources that are made avail-
able for the evaluation.
Based on our experiences with evaluation studies in the areas of labour
market and education3, this paper discusses the challenges and applicabil-
ity of expert interviews that result from the “stakeholder” problem and the
claim to efficiency. The first part illustrates the spectre of different scientific
approaches in evaluation; the approach we use, which is mostly oriented at
the model of responsive evaluation, is embedded. Subsequently we show
which value expert interviews have for certain evaluation steps and ques-
tions. The second part of the paper presents our approach to expert inter-
views and how we try to cope with the stakeholder-question.

11.1 Methodological variety in evaluation practice and

significance of expert interviews

11.1.1 The development of approaches to evaluation

Evaluation research has emerged in various disciplines (Rossi and others,
2004, Beywl, 1988, Shadish and others, 1991), constantly keeps developing,
has adopted more or less uniform and standardized procedures depending
on topics and organizational frame conditions, but is still far from a gener-
ally accepted methodology and terminology (Stufflebeam and others, 2000,
Lassnigg and others, 2000). This is, among other things, due to the fact
that evaluation as applied field of research has developed from a variety of
 Expert Interviews in Programme Evaluation 237

scientific areas, which have all introduced “their” methods. Williams (1989)
attempts to reduce this theoretical variety of conceptions on the basis of
meta-theoretical analyses of the “classics” of evaluation research. He ultim-
ately distinguishes between two dimensions:

● the methodological dimension alternates between qualitative and quantita-

tive approaches with the extreme positions interpretative and descriptive
versus scaled and causal.
● depending on orientation and inclusion of user or client interests the
application-oriented dimension alternates between the extreme positions of
orientation on general audience versus specific end users.

Based on these two dimensions, a map of evaluation practice may be drawn

which also renders the basic developmental tendency of evaluation research
visible. According to Lassnigg (1997), the approaches to evaluation by no
means develop in a linear manner. Rather, he illustrates multiplication and
diversification with the image of a balloon which has air blown into it and
which slowly expands without the previous approaches losing in importance.
For a long time the highly quantitatively-oriented experimental method, in
which the use of resources and the achieved outputs are compared while the
processes of the programme remain in the “blackbox.”4 This was considered
the ideal approach to effect analysis. Over time, however, not only effects
were of interest also questions regarding the underlying effect mechanisms
and conditions became more relevant. This development led to an increas-
ing application of qualitative methods in evaluation research. In addition,
the perception increasingly spread that evaluation should support the fur-
ther development of programmes. Hence, special emphasis was placed on
the clients’ interests and evaluation became more like political consulting.
In general, however, the “balloon effect” described above suggests to
(potential) users of evaluations that the evaluation methods are arbitrary
and barely comprehensible. The impression is created that it is possible to
“prove anything” at will; that evaluation is primarily used to legitimize
unpleasant decisions or as a substitute to appease those who are dissatisfied
or to postpone decisions.5 This, however, is where evaluation practice begins
its efforts to develop uniform and binding evaluation standards. Most devel-
oped are the Standards of the Joint Committee in the USA, which – based on
evaluation in the education sector – define terms of reference for programme
evaluations (Sanders 1999) and which have also strongly influenced the
development of evaluation standards in the German-speaking area.6 These
standards explicitly attempt to take the competing demands on evaluation
from clients and science into account. Clients expect to be provided with
evaluation results that have a useful format, meet the requirements and
are on schedule and inexpensive, whereas for science adhering to criteria
of scientific quality and high methodological standards are foregrounded.
238 Angela Wroblewski and Andrea Leitner

The Standards of the Joint Committee, however, focus on practical aspects of

research (applicability and preparation of the results, performance of the
evaluation and propriety of the approach), whereas methodological ques-
tions are only touched upon briefly and the prevailing standards in the
various disciplines referred to.7
One approach in evaluation that explicitly takes user and client interests
into consideration, sets clear standards for evaluation and is characterized
by increased application of qualitative methods is the model of “respon-
sive evaluation” (Beywl, 1988). Beywl defines a number of central principles
of responsive evaluation which show a rather close affinity to the general
principles of qualitative methodology (for example Lamnek, 1995, p. 21ff).
Initially the evaluators are to react flexibly to the concrete events of the
object of investigation, that is to resolve preconceptions, prejudices and
implicit paradigms to the extent possible (principle of stimulus reception
from the field of enquiry). Neither the object of investigation itself nor its
context is to be regarded as fixed or limited from the start. Either area may
change in the course of the evaluation; here the problems perceived by the
participants play a central role (principle of openness). Depending on the
encountered conditions, flexibility and openness are also required with
regard to the methods used. The object is to be approached from all angles,
with several methods, ideally by different researchers and from different
perspectives (triangulation8). Even in case of a clear preference for qualita-
tive methods, this principle includes the use of quantitative methods.

11.1.2 Relevance of expert interviews for

certain steps in the evaluation and sets of questions
Using expert interviews is not equally suitable for all evaluation models. To
show the significance and function of expert interviews for specific sets of
questions we first describe an evaluation scheme oriented at the model of
responsive evaluation originally developed for programmes of active labour
market policy (Lassnigg and others, 2000). The individual evaluation steps
and related sets of questions are described in broad strokes and the relevance
of expert interviews is discussed for each.
In the responsive evaluation model expert interviews play an important
role. Various possibilities and approaches are utilized in order to acquire expert
opinions in the evaluation: classical one-on-one interviews and repeated con-
versations with the same experts9 or group discussions (possibly to supple-
ment one-on-one interviews). In addition written methods may also be used
(for example the Delphi method) or oral and written interviews combined.
In principle we distinguish between four evaluation steps: policy analysis,
analysis implementation, effect analysis and cost-benefit analysis. Which
steps are actually performed in a particular evaluation depends on the
respective question(s) and objective(s) of the evaluation. In a comprehensive
evaluation of a programme these steps build on and refer to one another.
 Expert Interviews in Programme Evaluation 239

In a policy evaluation the objective(s) of the programme, its labour mar-

ket context and the target group’s characteristics are put in relation to one
another. Concretely, the main concern is to grasp the programme in its
essential parameters and to subject it to a general examination of its internal
consistency and its appropriateness regarding the context (economic devel-
opment, labour market situation) and the target group. Usually this evalu-
ation step is based on document analysis (planning document, funding
applications, and so on). Against the background of the labour market situ-
ation, the concept is analysed in general as well as for the respective target
group; for this labour market statistics are usually used. The key questions
of the policy evaluation, such as for example those about the motivation,
expectations and assumed effects the programme is based on, can only be
answered on the basis of interviews with decision-makers and the people
involved in programme development, for they are hardly ever stated expli-
citly in concepts or planning documents. In addition, experts are import-
ant when a programme that has been targeted at a particular group is to
be assessed with regard to its appropriateness for the group’s needs This
especially when the programme is targeted at problematic groups about
whose life situation or needs the evaluators are not sufficiently informed
(for example handicapped, former addicts, homeless people or migrants).
The implementation evaluation focuses on the implementation and real-
ization of the programme. Organizational aspects are taken into account,
as are contents and the composition of participants. The realization of the
programme is analysed against the background of the respective object-
ive, that is feedback is given to the conception of the programme in order
to be able to identify problem areas (for example difficulties with reach-
ing the target group, acceptance problems, or regional disparities) and to
develop suggestions for improvement. Methodologically the analysis is – to
the extent available – based on monitoring data (information on number
of participants and their characteristics, length of participation, premature
dropouts and so on). Depending on the main question of the evaluation
and available resources, field research (such as for example participant sur-
veys) may also be performed. For complex programmes, for example if sev-
eral institutions are involved in the implementation, the analysis of the
intersection points between the players or institutions becomes a key ques-
tion. The questions are the design of the points of intersection, the type of
cooperation between the involved parties, information flows and possible
problem areas (for example ill-defined assignment of competences, dupli-
cation of work or differing ideas about objective and implementation of
the programme). In this regard personal interviews often are the only feas-
ible way of acquiring information, since the mechanisms forming the basis
of certain developments can hardly be uncovered in any other way. If it
becomes apparent in the implementation evaluation, for example, that the
target group could not be reached or certain subgroups among participants
240 Angela Wroblewski and Andrea Leitner

are underrepresented (such as for example older women or people with only
compulsory schooling), the access paths to the programme should be ana-
lysed. Typically several players are involved in the selection process, such as
for example the advisors of the Public Employment Service (AMS) branch
offices, who suggest a programme to the unemployed, the realizing insti-
tutions, which select particular participants, and finally the unemployed
themselves, who decide whether or not they consider participating. Each
of these players has a certain range of freedom in decision-making and acts
according to individual values, norms and motives, which ultimately deter-
mine specific participant selection. The interplay of these factors can only
be shown through expert interviews.
Through effect analysis all effects caused by the programme (intended
as well as unintended) are to be assessed. The analysis aims at filtering
out those effects that can be traced to the intervention.10 The methods
selected for effect analysis are primarily quantitative; in recent years sig-
nificant progress has been made in this respect (for example Schmid and
others, 1996, Shadish and others, 2001). Apart from determining particu-
lar effects (output), however, the underlying effect mechanisms are to be
analysed as well. These can usually not be measured quantitatively, but are
for example based on experts’ subjective assessment of macroeconomic or
social changes.
Expert opinions also play a role in the cost-benefit analysis, which may
seem surprising at first glance since this is the monetary appraisal of a pro-
gramme’s costs and benefits. Quantifying benefit components, however, is
often connected with information problems, if for example factors such as
transfer payment or administrative expense savings or cost efficiency due to
reduced alcohol or drug consumption are to be included on the benefit side.
In the cost-benefit analysis, assumptions about these factors are to be made
and explicated, for which – in addition to experiences from other studies –
expert opinions are frequently taken into account as well.

11.2 Practical aspects of preparing,

conducting and analysing expert interviews

The key objective of expert interviews in the context of evaluations is to

efficiently supplement the information derived from documents and data
with the various perspectives of experts in order to obtain a well-rounded
picture of the programme to be evaluated. At the core of the following
remarks is the “classical” conversation with experts. We particularly address
the characteristics resulting from the evaluation situation, especially the
“stakeholder problem” and the related interaction process.
The “stakeholder problem” emerges already when selecting the inter-
view partners, since the experts mainly come from those institutions that
are substantially involved in the development or implementation of the
 Expert Interviews in Programme Evaluation 241

programme. The “stakeholders” frequently associate an evaluation with

controlling functions, even if the evaluation primarily is to fulfil insight-
generating functions and changes or questions regarding the continuation
of programmes and programmes are not directly linked to the results. The
social situation at the interview can thus be regarded as an oral examin-
ation, in which arguments justifying one’s behaviour are provided to and
the real problems are concealed from the evaluators. Such self-portrayal
and information control strategies are of course not only found in expert
interviews in the context of evaluations, but also affect other forms of data
collection and social interactions in general (Leitner, 1994, Steinert, 1984,
Goffmann, 1967). Based on the interviewees’ occupational position, how-
ever, it can be assumed that they present themselves more frequently and
in front of different audiences with this topic and consider certain precau-
tionary measures or strategies already before the interview in order to per-
form as well as possible in the “exam situation.” Due to the co-occurrence
of these two factors (evaluation situation and the specific competences of
the experts), distortion effects such as refusal to answer, interviewer effects,
position effects, presence effects or client effects are to be expected at higher
levels than with classical interviews or expert interviews with lower interest
in the results.11
In the following, a possible approach to expert interviews is described
which has resulted from practice and being confronted with problems and
which continues to be developed further. Here the specific interaction pro-
cess in evaluations is explicitly taken into account, and the involved par-
ties are specifically included in phrasing the questions and designing the
enquiry. In this way response distortions are reduced. As a consequence
acceptance of the evaluation increases and thus the quality of the results
improves. A number of key points arising in connection with preparing,
conducting and analysing expert interviews in evaluation research are
addressed.

11.2.1 Preparing the interviews

Before conducting the interviews it must be clarified who is to be included
as an expert and which topic areas are to be focused on in the conversations
with the experts. When choosing the experts special consideration must be
given to including all levels of players to the extent possible. In practice it
can be difficult when choosing the interview partners to reach all relevant
players (due to lack of information, but also for reasons of availability, such
as for example former “key players” changing jobs, scheduling problems)
or to induce the important people to participate in the evaluation. It has
proven useful initially to make only a preliminary selection since additional
key players, who are for example mentioned in several interviews and who
were originally not known to the evaluators, may still emerge during the
project.
242 Angela Wroblewski and Andrea Leitner

As stated above, the experts are part of the programme in various forms
or functions. They are involved in conception and planning, implementa-
tion and realization as well as in changes to and further developments of
the programme. In addition it is often helpful to confront their perspectives
with those of users or players from the environment of the programmes
(similar or complementary programmes). Expert status is therefore based on
function and includes both decision-makers as well as implementers with-
out decision-making authority. Experts with decision-making authority are
mostly selected on the basis of the position they occupy in the hierarchy
of the institution involved in the programme. By contrast, for the selection
of experts from the implementation level, evaluators frequently depend on
additional information (which they for example receive from other experts
or gather from documents).
In a further step, the intended topic areas of the interviews are estab-
lished on the basis of the evaluation question. In an implementation evalu-
ation, for example, which is to give information on the implementation of
a programme, show problem areas and starting points for changes, the key
topic areas will be the practical experiences gained when realizing the pro-
gramme, the perception of problem areas, opportunities for improvement
in the experts’ view, the appraisal of individual aspects of the programme as
well as the assessment of its effects.
From these topic areas a “rough guideline” (catalogue of topics) is devel-
oped which defines the framework for the expert interviews. The design of
the guideline is based on theoretical assumptions about influencing fac-
tors, problem areas or effects resulting from already available information
about the programme (for example documents, media coverage, or previ-
ous studies) and experiences with the evaluation of similar programmes.
Starting from this catalogue of topics, however, the focus of each individual
conversation varies depending on the position of the experts. Interviews
with trainers, for example, will focus on the immediate contact to the par-
ticipants, whereas with Public Employment Service representatives the exe-
cution of the programme and the cooperation within other institutions
involved in the programme will be of central interest.
Individual preparation is therefore necessary for each interview. The spe-
cific questions for the individual topics are developed, which are oriented
at the function or position of the respective experts and which take the
current state of information in the project into account. In the process of
the evaluation the underlying guideline is continuously revised, if individ-
ual topic areas turn out to be irrelevant or if no new information arises for
certain topics (that is the topic is considered “saturated”).12 Each individ-
ual expert interview is prepared taking the respective expert’s occupational
positions into consideration by formulating hypotheses that are presented
to the experts during the interview. This facilitates thematizing and dis-
cussing diverging interests between the groups of players or resulting role
 Expert Interviews in Programme Evaluation 243

conflicts without calling the expert status of the interview partners into
question. Simultaneously, the interviewers demonstrate their knowledge of
correlations and effects, thus strengthening their competence and position
in the interview.
It is tried and tested practice not to conduct the interviews en bloc, but to
distribute them over the duration of the project. At the start of the evalu-
ation, interviews with the heads of the institutions involved prove helpful
since these frequently have a good overview of the programme, its organ-
ization, content and changes that may already have been made as well as
special characteristics compared to other programmes. In combination
with the analysis of programme-related documents (for example concept,
annual reports), this information provides a solid basis for the further evalu-
ation steps. Since frequently several people from the participating institu-
tions are interviewed as experts (for example in addition to senior staff also
employees from the planning department), it has proven useful to conduct
one interview each in the initial phase and towards the end of the project,
because often additional questions appear during the duration of the pro-
ject or surprising results (positive as well as negative) emerge which can be
discussed with the experts. For this reason, the expert interviews should be
held parallel to the other steps of the enquiry, such as for example analysing
the administrative data or conducting a participant survey, in order to be
able to react flexibly to the results of the respective work stages.

11.2.2 Conducting the interviews

Due to the specific interaction process in the expert interview and the evalu-
ation situation it is frequently not easy in the interviews to persevere with
the topic of the interview or gain access to experts and identify such distor-
tions in the analysis. Vogel (1995, p. 78ff.) paraphrases the problems occur-
ring in expert interviews with iceberg, paternalism, feedback and catharsis
effects. “Iceberg effect” refers to an interview situation that is character-
ized by obvious mistrust or disinterest on the expert’s part. “Paternalisation
effects” emerge when experts attempt to take charge of the interview and
challenge the interviewers’ competence. The phenomenon of interviewees
repeatedly attempting to reverse the question-answer situation in the inter-
view and to obtain information from the interviewers is referred to as “feed-
back effects.” Another situation, in which the topic of the interview can
hardly be maintained is constituted by the experts using the interview as
an opportunity to vent their day-to-day professional frustration and annoy-
ance or to primarily showcase themselves (“catharsis effects”). To avoid such
problems it is important to regard the interview as a social interaction that is
characterized by the roles of the conversation partners and the expectations
of these roles as well as a stimulus-response pattern.
A brief outline of the evaluation design (key questions and methodological
approach) and the current project status provides a suitable introduction to
244 Angela Wroblewski and Andrea Leitner

the interview. In the process the interviewers’ status is also clarified, the
respective expert interview is placed in the context of the overall project,
and the topics and sets of questions to be prioritized with a particular expert
are defined. This approach facilitates eliminating possible objections from
the interviewees – they have nothing to say to the programme or have not
been involved in the programme for a long time already – in advance.13
Simultaneously, also the distribution of roles in the interview is established
and possible assumptions about it are corrected. It is attempted from the
start to avoid potential iceberg effects by emphasizing the cooperative
nature of the interview and underlining the significance of the expert’s
knowledge for the evaluation. In this context it should, as a matter of prin-
ciple, also be made clear how information obtained in the interview will
be handled (for example anonymity of the interview partners’ statements,
release of interview transcripts). This may also contribute to increasing
the interviewee’s willingness to cooperate since fears of undesirable conse-
quences are reduced.
In practice the introductory phase proves essential for the success of the
interview and the amount of work involved in its utilization. This pertains
not only to gaining access to the interview partners by creating a pleasant
environment for the conversation, but also to setting the course regarding
content. Here it is important to demonstrate the interviewers’ knowledge
and competences by inserting quotes from key documents for example
about official positions of the institution on the topic. With regard to topics
such as for example gender mainstreaming, for which almost all institu-
tions have written commitments by now – which are however heavily based
on unchallenged hollow phrases – this can facilitate that the time of the
interview is not devoted to clarifying “official” definitions, but that the con-
tents of specific interest can be reached quickly.
The “stakeholder problem” was also apparent in conversations with
trainers of training programmes, since they are in a control situation and
self-presentation and information control form a part of their teaching con-
tents and competences. In a programme to support women returning to
their careers after a longer absence from work (back-to-work programme),
for example, the selected participants differed significantly from the tar-
get group. This “creaming – effect” (that is selecting the best among the
applicants) was linked to trainer behaviour in the programme: the train-
ers are assessed on the basis of achieving certain targets, especially their
placement rate (Leitner and Wroblewski, 2001). In the interviews with the
trainers it was nearly impossible to obtain specific information on the prob-
lem of participant selection, because the interview partners – experts in
rhetoric, communication and self-presentation – were not willing to discuss
this sensitive topic. Consequently, hypotheses about participant selection
were introduced in the interviews, and on this basis the effects of the cri-
teria by which the participating institutions are assessed were addressed and
 Expert Interviews in Programme Evaluation 245

their usefulness discussed. In addition, further sources of information (for

example administrative data, interviews with other players) were included
to analyse the selection processes.
An essential aid for sticking to the point and returning to questions
that have remained open in the course of the conversation is the use of
the guideline to monitor the questions that have been answered. Two
interviewers who cooperate can more easily avoid an “exam situation” and
create a “relaxed” conversational atmosphere, for it is easier to follow the
“natural” course of the conversation without constantly returning to the
guideline – the second interviewer is mindful of still open questions and
can introduce these into the conversation. Using two interviewers makes
conducting the interviews easier because they support each other, avoid
“lows” in the conversation and are able to react more quickly to unexpected
information. For this it is necessary however to “distribute the roles”: one
interviewer primarily takes over the “question part,” the other is respon-
sible for follow-up or additional questions. The interviews are recorded on
tape – with the interview partners’ consent. If such a consent is refused or
if important information emerges only after the tape has been switched
off, which may for example be the case when evaluating a problematic pro-
gramme (in the sense of conflict-laden or controversial in public or within
the participating institutions), the second interviewer is also responsible for
keeping a written record.
Conducting the interviews is also made significantly easier if the pertin-
ent terminology is used. For one thing, the experts are more likely to accept
the interview partners as equal conversation partners, because they have
certain prior knowledge and can therefore be expected to be able to grasp
also the respective relevant context. Furthermore, the interviews may thus
become much shorter, since considerably fewer additional questions are
necessary. Simultaneously – as is the case also in other research areas – the
interviewers must take care not to accept unchallenged certain assumptions
that have turned out to be relevant in similar contexts.
For the end of the interview it has proven useful to ask an open question
about possibly unaddressed points that are in the expert’s opinion import-
ant with regard to the programme. Thus the interview partners are given
the opportunity to introduce points that are important to them (or also to
vent their frustration; see “catharsis effects” above) in a rather informal con-
text (since this is after the end of the interview).

11.2.3 Analysing the results and feedback

In the analysis of expert interviews the “stakeholder problem” must expli-
citly be taken into account. Besides, here the tension between the demanded
adherence to scientific quality criteria and efficiency criteria becomes
most obvious. How is it possible within the specified time and with scarce
resources to meet the standards demanded by science? This becomes even
246 Angela Wroblewski and Andrea Leitner

more of a challenge as expert interviews are frequently characterized by

subjective interpretation or strategic disclosure of information, but due to
limited resources a more detailed textual analysis is hardly possible. Our
approach – and this should have become clear from the previous chapters –
consists in preparing and conducting the interviews with a clear eye to
their analysis. Due to the topic-centred questions and the iterative process
of developing the guideline and obtaining the information, which also
includes other sources apart from the expert interviews, the process of ana-
lysis is in part anticipated already in the previous steps. Intensive prepar-
ation and well-attuned teamwork in conducting the interviews make the
analysis significantly easier. The analysis will be focused on the main topics
mentioned in the interviews. The transcripts of the interviews form the
basis for the analysis. For resource reasons frequently only key passages are
transcribed verbatim; the remaining parts of the interview are paraphrased
and summarized if necessary. In practice additional parts of the interview
may also turn out to be vital at a later project stage and are transcribed ver-
batim afterwards.
Gathering the interview partners’ different opinions and positions regard-
ing the topic or problem areas specified in the guideline is emphasized in
the analysis. Here the respective interests are to be pointed out and jux-
taposed. Contrasting the various positions and interests also facilitates a
plausibility check of the results. All information is to be summarized, cat-
egorized and interpreted thus that ultimately the overall structure of the
programme becomes graspable. Conflict-free as well as current or potential
problem areas are to be taken into account. In the already mentioned evalu-
ation of the back-to-work programme, for example, it became apparent that
the involved parties were hardly aware of any conflict areas within the pro-
gramme. Among other reasons, however, this is also due to the fact that
the programme was implemented highly successfully, that is different ideas
about the objectives and target group of the programme were not relevant.
Nevertheless those were possible areas of conflict.
The most significant step in the analysis consists in disengaging from
the interview level and coming to the programme level. This is achieved
by juxtaposing the individual interviews against the background of further
information on the programme and its context. Thus diverging interests,
contradictory requirements for individual players, inefficient processes
but also “best practices” can be pointed out and included in the analysis.
Through this approach the expert interviews attain central significance
within the evaluation project, since precise knowledge of the frame condi-
tions as well as the processes in the programme is a prerequisite for inter-
preting the results.
The involved parties (stakeholder) are informed of the results either
in feedback rounds, which are an integral part of responsive evaluation,
or through the final presentation and reporting. In the report particular
 Expert Interviews in Programme Evaluation 247

attention is paid to possible deviations of the implementation from the ori-

ginal conception as well as to possibilities of further developing the pro-
gramme. These aspects generally obtain greater acceptance if the main
stakeholders, who are in most cases also the addressees of the evaluation
results, have contributed to the evaluation as experts. Through this feed-
back of results the experts have the possibility to see their situation in
the context of the overall programme “in a new light” so to speak, which
increases understanding for the situation of the other players and improves
the basis for cooperation. This does not only grant greater importance to
the principle of objectivity, it is also evident that expert interviews – and in
particular methods which require the different players’ immediate reaction
to one another – have yet another function: they increase the players’ self-
reflection and stimulate the communication process between the different
players.

11.3 Conclusions

As has been shown in this contribution, expert interviews are an import-

ant but methodologically little noticed information-gathering instrument
in evaluation research, which is characterized by a specific type of social
interaction resulting from the “stakeholder problem.” These characteristics
increase the need for methodological reflection in practice and for the fur-
ther development of standards for the use of expert interviews in evaluation
research.
It seems necessary, however, not only to continue to develop the meth-
odology but also to raise the significance of expert interviews in relation to
quantitative approaches. This is a matter of putting the results of quantita-
tive and qualitative approaches side by side at equal weight and to regard
them as mutual complements. Quantitative results, so-called “hard facts,”
are frequently emphasized in the evaluations, because they are easier to
exploit politically or as effective publicity and enable comparison with
other programmes. On the basis of quantitative results alone, however, the
really exciting questions of evaluation, such as for example “How do these
results come about?” or “How can undesirable effects be avoided?,” cannot
be answered. To determine connections between effects or factors influen-
cing the results, expert interviews are a scientifically suitable method which
additionally also meets the claims to efficiency demanded in commissioned
research.

Notes
1. For example, in the USA the first textbooks and professional journals were pub-
lished in the 1970s, the American Evaluation Association was founded in the early
1980s. In Europe a comparable professionalization process set in about 20 years
248 Angela Wroblewski and Andrea Leitner

later (for example the European Evaluation Society was founded in 1994, the
German Evaluation Society – Deutsche Gesellschaft für Evaluation – in 1997). See
also Steiner and others 2008.
2. According to Weiss (1983), “stakeholders” are “those whose lives are affected by
the program and its evaluation” (Weiss, C. H. 1983: “The stakeholder approach
to evaluation” in Bryk, A. S. (ed.) Stakeholder-based evaluation, San Francisco; cited
from Shadish and others 1991: 179ff.).
3. For example Leitner and Wagner (2008); Leitner (2007); Wroblewski and others
(2007a, b); Wroblewski and Vogtenhuber (2006); Lutz and others (2005); Leitner
and Wroblewski (2001); Lassnigg and others (2000); Lassnigg and others (1999).
4. The roots of evaluation research are located in the USA in the 1960s. It primarily
aimed at ascertaining the effects of governmental programmes in order to be
able to assess and improve the interventions. Experimental methods are com-
parative group approaches in which participants in a programme are compared
to an as-identical-as-possible group of non-participants in order thus to be able
to filter out the effects of participation. Ideally participants are assigned to the
two groups randomly (Shadish and others 1991, 2001).
5. This problem was clearly expressed by two very well-known evaluation
researchers: “...if you advocate a particular policy reform or innovation, do not
press to have it tested” (Burtless and Haveman [1984]: Policy Lessons From Three
Labor Market Experiments, Ottawa; cited from OECD 1991: 49), or more astutely
yet by Wilensky’s law: “the more evaluation, the less program development, the
more demonstration projects, the less follow-through” (Wilensky 1985: 9).
6. See for this the standards of the German Evaluation Society (Gesellschaft für
Evaluation, or DGEval; www.degeval.de) or those of the Swiss Evaluation Society
(Schweizerische Evaluationgsgesellschaft, or SEVAL; www.seval.ch).
7. The Joint Committee (Sanders 1999) divides the total of 30 standards into four
groups (utility, feasibility, propriety, accuracy), which are called “standards of
excellence” by Patton (1997, p. 15ff). Utility standards are to ensure that an
evaluation is oriented at the information requirements of the designated evalu-
ation users. Feasibility standards are to ensure that an evaluation is performed
realistically, well thought out, diplomatically and cost-consciously. Propriety
standards are to ensure that an evaluation is legally and ethically proper and
the rights of those included in the evaluation and/or affected by its results are
preserved. Accuracy standards are to ensure that an evaluation concerning the
quality and/or the applicability of the evaluated programme generates and con-
veys adequate subject-specific information.
8. Method triangulation is referred to by Vogel (1995, p. 74) as “cross-examination.”
To us this term seems too narrow, since the concern here is not only to check
individual enquiries through other methods. Rather, an enquiry into different
aspects of the evaluation through a division of labour also seems useful.
9. Vogel (1995, p. 77) refers to expert interviews “en passant” or to experts as
“sources.”
10. It is customary to work with quasi-experimental approaches (comparative group
designs) to answer this question (Heckman and Smith, 1996, or Shadish and
others, 2001).
11. Distortion effect overviews can be found for example in Steinert (1984) or
Reinecke (1991). Systematic response distortions, such as socially desirable
response behaviour or agreement tendencies, which have comparatively lower
significance in expert interviews, were analysed for example by Reinecke (1991)
or Esser (1986).
 Expert Interviews in Programme Evaluation 249

12. Voelzkow (1995) refers to interviews that are based on a similar approach as
“iterative expert interviews.”
13. These objections may arise if evaluation is primarily understood as the gath-
ering and analysis of the effects of a programme, that is if a wide-spread but
nevertheless too narrow notion of evaluation is used.

Further readings
Gubrium, J. and Holstein J. (eds) (2002) Handbook of Interview Research: Context and
Methodology (Thousand Oaks, CA: Sage Publications).
Patton, M. Q. (2002) Qualitative Research & Evaluation Methods, 3rd edn (Thousand
Oaks, CA: Sage Publications).
Yin, R. K. (2003) Case Study Research. Design and Methods, 3rd edn (Thousand Oaks,
CA: Sage Publications).

References
Beywl, W. (1988) Zur Weiterentwicklung der Evaluationsmethodologie. Grundlegung,
Konzeption und Anwendung eines Modells der responsiven Evaluation (Frankfurt am
Main, Bern and New York: Verlag Peter Lang).
Brinkmann, C., Deeke, A. and Völkel, B. (eds) (1995) “ExpertInneninterviews in
der Arbeitsmarktforschung” in Beiträge zur Arbeitsmarkt- und Berufsforschung 191
(Nürnberg: Institut für Arbeitsmarkt- und Berufsforschung).
Deeke, A. (1995) “Experteninterviews – ein methodologisches und forschungsprak-
tisches Problem” in Brinkmann, C., Deeke, A. and Völkel, B. (eds) Experteninterviews
in der Arbeitsmarktforschung. Diskussionsbeiträge zu methodischen Fragen und prak-
tischen Erfahrungen. Beiträge zur Arbeitsmarkt- und Berufsforschung Nr. 191 (Nürnberg:
Bundesanstalt für Arbeit), pp. 7–22.
Esser, H. (1986) “Können Befragte lügen? Zum Konzept des ‘wahren Wertes’ im
Rahmen der handlungstheoretischen Erklärung von Situationseinflüssen bei der
Befragung” in Kölner Zeitschrift für Soziologie und Sozialpsychologie 38/2, 314–36.
Flick, U. (2006) “Interviews in der qualitativen Evaluationsforschung” in Flick, U.
(ed.) Qualitative Evaluationsforschung. Konzepte, Methoden, Umsetzungen (Reinbek bei
Hamburg: Rowohlt Taschenbuch Verlag), pp. 214–32.
Goffman, E. (1967) Interaction Ritual (Chicago: Aldine Publishing Company).
Heckman, J. J. and Smith, J. A. (1996) “Experimental and Nonexperimental
Evaluation” in Schmid, G., O’Reilly, J. and Schömann, K. (eds) International
Handbook of Labour Market Policy and Evaluation (Cheltenham: Edward Elgar),
pp. 37–88.
Lamnek, S. (1995) Qualitative Sozialforschung, Bd. 1: Methodologie, 3rd edn (Weinheim:
Beltz).
Lassnigg, L. (1997) “Evaluation: Aufdecken, Zudecken, oder was sonst ...” in
Zilian, H. G. and Flecker, J. (eds) Pathologien und Paradoxien der Arbeitswelt (Vienna:
Forum Sozialforschung), 227–62.
Lassnigg, L., Leitner, A., Steiner, P. M. and Wroblewski, A. (1999) Unterstützung beim
Wiedereinstieg, Möglichkeiten und Wirkungen frauenspezifischer Maßnahmen (Vienna:
AMS Österreich).
Lassnigg, L., Leitner, A., Unger, M. and Wroblewski, A. (2000) Zukunftsmodell
Qualifizierungsverbund – Evaluierung des Qualifizierungsverbundes Triestingtal (Vienna:
AMS Österreich).
250 Angela Wroblewski and Andrea Leitner

Lassnigg, L., Leitner, A., Wroblewski, A., Steiner, M., Steiner, P. M., Mayer, K.,
Schmid, G. and Schömann, K. (2000) Evaluationsschema für Maßnahmen der aktiven
Arbeitsmarktpolitik in Wien (Vienna: Institute for Advanced Studies).
Leitner, A. (1994) “Rationalität im Alltagshandeln. Über den Erklärungswert der
Rational Choice-Theorie für systematische Antwortverzerrungen im Interview” in
Sociological Series No. 5 (Vienna: Institute for Advanced Studies).
Leitner, A. (2007) Frauenförderung im Wandel. Gender Mainstreaming in der
österreichischen Arbeitsmarktpolitik (Frankfurt am Main: Campus Verlag).
Leitner, A. and Wagner, E. (2008) Evaluation der Implacementstiftung Pflegeberufe
(Vienna: Institute for Advanced Studies).
Leitner, A. and Wroblewski, A. (2001) Aktive Arbeitsmarktpolitik im Brennpunkt V:
Chancen für Wiedereinsteigerinnen mit längeren Berufsunterbrechungen (Vienna: AMS
Österreich).
Lutz, H., Mahringer, H. and Pöschl, A. (2005) Evaluierung Europäischer Sozialfonds
2000–2006. Ziel 3. Österreich (Vienna: WIFO).
OECD (1991) Evaluating Labour Market and Social Programmes: The State of a Complex
Art (Paris: OECD).
Patton, M. Q. (1997) Utilization-Focused Evaluation, 3rd edn (Thousand Oaks, London
and New Delhi: SAGE Publications).
Reinecke, J. (1991) Interviewer- und Befragtenverhalten: Theoretische Ansätze und method-
ische Konzepte (Opladen: Westdeutscher Verlag).
Rossi, P. H., Freeman, H. E. and Lipsey, M. W. (2004) Evaluation: A Systematic Approach,
6th edn (Thousand Oaks, London and New Delhi: SAGE Publications).
Sanders, J. R. (ed.) (1999) Handbuch der Evaluationsstandards. Die Standards des
“Joint Committee on Standards for Educational Evaluation” (Wiesbaden: Verlag für
Sozialwissenschaften).
Schmid, G., O’Reilly, J. and Schömann, K. (eds) (1996) International Handbook of
Labour Market Policy and Evaluation (Cheltenham and Brookfield: Edward Elgar
Publishing Company).
Shadish, W. R., Cook, T. D. and Campbell, D. T. (2001) Experimental and Quasi-
Experimental Designs for Generalized Causal Interference (Boston and New York:
Houghton Mifflin Company).
Shadish, W. R., Cook, T. D. and Levition, L. (1991) Foundations of Program Evaluation.
Theories of Practice (Newbury Park, London and New Delhi: SAGE Publications).
Steiner, P. M., Wroblewski, A. and Cook, T. D. (2009) “Randomized Experiments and
Quasi-Experimental Designs in Educational Research” in Ryan, K. and Cousins, B.
(eds) International Handbook of Educational Evaluation (Thousand Oaks, London and
New Delhi: SAGE Publications), pp. 75–95.
Steinert, H. (1984) “Das Interview als soziale Interaktion” in Meulemann, H. and
Reuband, K. H. (eds) Soziale Realität im Interview. Empirische Analysen methodischer
Probleme (Frankfurt am Main: Campus Verlag), pp. 17–59.
Stufflebeam, D. L., Madaus, G. F. and Kellaghan, T. (2000) Evaluation models. Viewpoints
on Educational and Human Services Evaluation, 2nd edn (Bosten, Dordrecht and
London: Kluwer Academic Publishers).
Voelzkow, H. (1995) “ ‘Iterative Experteninterviews’: Forschungspraktische
Erfahrungen mit einem Erhebungsinstrument” in Brinkmann, C., Deeke, A. and
Völkel, B. (eds) Experteninterviews in der Arbeitsmarktforschung. Diskussionsbeiträge
zu methodischen Fragen und praktischen Erfahrungen. Beiträge zur Arbeitsmarkt- und
Berufsforschung Nr. 191 (Nürnberg: Bundesanstalt für Arbeit), pp. 51–8.
Vogel, B. (1995) “ ‘Wenn der Eisberg zu schmelzen beginnt ...’ – Einige Reflexionen
über den Stellenwert und die Probleme des Experteninterviews in der Praxis der
 Expert Interviews in Programme Evaluation 251

empirischen Sozialforschung” in Brinkmann, C., Deeke, A. and Völkel, B. (eds)

Experteninterviews in der Arbeitsmarktforschung. Diskussionsbeiträge zu methodischen
Fragen und praktischen Erfahrungen. Beiträge zur Arbeitsmarkt- und Berufsforschung
Nr. 191 (Nürnberg: Bundesanstalt für Arbeit), pp. 73–84.
Wilensky, H. L. (1985) “Nothing Fails Like Success: the Evaluation-Research Industry
and Labour Market Policy” in Industrial Relations 24/1, 1–19.
Williams, J. (1989) “A numerically developed taxonomy of evaluation theory and
practice” in Evaluation Review 13/1, 18–31.
Wroblewski, A., Leitner, A., Gindl, M., Pellert, A. and Woitech, B. (2007) “Wirkungsanalyse
frauenfördernder Maßnahmen des bm:bwk” in BMBWK (ed.) Materialien zur Förderung
von Frauen in der Wissenschaft, 21 (Vienna: Verlag Österreich).
Wroblewski, A., Leitner, A. and Osterhaus, I. (2007) IBEA – Integrative Berufsorientierung:
Evaluation der Maßnahmen und Ergebnisse des Moduls 3 (Vienna: Karl-Kummer-
Institut).
Wroblewski, A., Vogtenhuber, S. (2006) Studienbedingungen an Wiener Universitäten
(Vienna: Austrian Chamber for Employees), http://www.equi.at/dateien/
studienbedingungen.pdf, date accessed 8 June 2008.
12
The Delphi Method:
Eliciting Experts’ Knowledge in
Technology Foresight
Georg Aichholzer

12.1 Abstract

The Delphi method is basically a highly structured group process through

which experts assess issues on which knowledge is uncertain and imperfect
by nature. The Expert Delphi has in the meantime acquired a fixed position
as part of the technology foresight projects that are becoming of increas-
ing importance around the world. This paper presents the methodological
principles, sets out specific advantages and critical aspects and illustrates
the evolution and application in the context of foresight with examples
at international level. Design decisions, innovative elements and exploit-
ation context are presented in the light of the Austrian Technology Delphi
(including the problem-oriented approach, the application in the form of a
“Decision Delphi,” the broader definition of the expert and special meas-
ures to motivate participation). This is followed by an investigation of more
recent methodological modifications: the combination with a cross-impact
analysis on future European transport systems; a Finnish approach based
on a balance between consensus and diversity objectives; and the use of the
Expert Delphi with internet support.

12.2 The Delphi method

12.2.1 Basic features

As a rule, the Delphi method is carried out as an anonymous enquiry put
to a group of experts in two, rarely more, rounds, the questionnaire for the
second round feeding back the results of the first round mostly in the form
of values for the median or arithmetic mean and the distribution param-
eters. Basic characteristics are anonymity, iteration, controlled feedback and
statistical aggregation to create a group response (Rowe and Wright, 1999,
p. 354). Anonymity and the feedback of the results are intended to allow

252
 The Delphi Method 253

a virtual debate permitting an approximation to consensus uninfluenced

by distorting influences (status, group pressure, rhetoric etc). In this sense,
the Delphi method is seen as an efficient and effective group process that
allows many of the psychological distracting factors in conventional group
discussions to be eliminated. The use of the pragmatic knowledge of experts
organized by a monitor group ultimately leads to future-related statements
on issues on which knowledge is only uncertain or incomplete. The method
promises “to obtain the most reliable consensus of opinion of a group of
experts ... by a series of intensive questionnaires interspersed with controlled
feedback” (Dalkey and Helmer, 1963, p. 458).
A major justification for the validity of the Delphi method derives from
the “theory of errors,” according to which the aggregated group responses
can be expected to represent a statement that is superior to the majority of
the individual experts’ ones (Parentè and Anderson-Parentè, 1987, p. 140ff.).
A further argument for its use in the field of technology foresight is the fun-
damental possibility of identifying new development tendencies and trend
shifts and of making the circle of persons any size whatsoever. The latter
is of importance particularly for the desired stimuli for longer term think-
ing, knowledge transfer and indirect coordination amongst the actors in
national innovation systems.
At the same time, many aspects of the Delphi method are fundamentally
regarded as problematic, such as a tendency towards conformity instead of
genuine consensus, feedback at the risk of manipulation, negative responses
on the basis of anonymity, formulation of suitable statements as objects of
assessment, unclear criteria in the selection of participants, expert biases,
low response rates in final rounds, proof of reliability and validity, time
and effort required (cf. Landeta 2006, Keeney and others, 2001). Strategies
for improving internal and external validity concentrate inter alia on the
consideration of multiple perspectives in the research process, and attempt
to increase the credibility and applicability of the results by a combin-
ation of the Delphi method with other investigations (cf. Engels and Powell
Kennedy, 2007).

12.2.2 Evaluation results

The at times controversial discussion about the scientific value of the Delphi
method has in the meantime led to a number of efforts to evaluate the
Delphi approach. Overall, the positive aspects prevail. Although the assess-
ment uses a wide variety of criteria, the focus is mostly on validity, reliabil-
ity, accuracy of prediction and achievement of consensus.
A systematic meta-analysis conducted by Rowe and Wright (1999) in the
form of the analysis of 27 papers on the evaluation of the Delphi method
showed the following: overall, accuracy proved to be superior as against
statistical groups and as against groups interacting normally. In compari-
son with other structured group methods such as the Nominal Group
254 Georg Aichholzer

Technique, however, the results were not consistently better. The phenom-
enon of consensus increase is to be found practically everywhere, while
the precise cause remains unsolved. In conclusion, the authors argue for a
shift from method comparative studies to process studies on Delphi, and an
increased consideration of the phenomenon of assessment change and the
role of feedback.1
In Germany, Michael and Sabine Häder in particular have contributed
further analyses on the essential design aspects and the evaluation of the
Delphi method. Their conclusion, like that of Rowe and Wright, is that the
“possibilities and limits have still not been investigated satisfactorily,” and
that an important task is to “describe more accurately the preconditions for
the successful use of the method” (Häder and Häder, 2000, p. 27ff., own
translation from German), with cognitive psychological contributions being
of considerable importance in the theoretical justification of the effective-
ness of Delphi.
Landeta (2006) sums up his evaluation by emphasizing that despite the
manifold weaknesses and points of criticism the Delphi method is super-
ior to using expert knowledge through statistical groups and classic direct
interaction groups. A growing application of this technique over decades
together with an enormous extension of the subject domains is clearly evi-
denced by research and published articles. An important field for the use of
the Delphi method is Technology Foresight. As one of the best-known prog-
nostic methods, the Delphi method is particularly appreciated in this field
both because of the advantages that result from its basic characteristics and
thanks to the fact that it has proved its value relatively well.

12.3 Technology foresight as a field of application

12.3.1 Objectives and development trends

Technology foresight has developed into an important instrument worldwide, 2
and recently for instance was also tried in China (Rongping and others,
2008). In technology policies today, it is regarded as a “source of strategic
intelligence” and a “strategic core competence” (Major and others, 2001).
Foresight studies aim at a methodologically controlled assessment of scien-
tific and technical development trends promising high benefits for society.
A common definition characterises foresight as a “systematic attempt
to look into the long-term future of science, technology, economy and
society” ... with the aim “to identify the areas of strategic research and the
emergence of generic technologies likely to yield the greatest economic and
social benefits” (Irvine and Martin, 1984).
Foresight is thus the attempt to apply suitable methods to use the know-
ledge of experts to reduce uncertainty and as it were to rationalize the future,
although still loaded with tensions and certain restrictions (Rappert, 1999).
 The Delphi Method 255

At the same time, foresight projects are also attributed an important func-
tion in communication and mediation between social interests.3 Martin
and Johnson (1999) emphasize above all the role of the networking of activ-
ities in different social subsystems and see this as a central contribution to
strengthening the national innovation systems.
Since the 1990s, there has been a huge growth in foresight activities and
national foresight programmes (cf. Grupp and Linstone, 1999, Blind and
others 1999, Cagnin and others, 2008). The more recent generation of tech-
nology foresight distinguishes itself from the term forecasting, not regard-
ing itself as a deterministic prognosis but rather emphasizing the probability
character of its statements, attempting to take account of the interdepend-
ence of technological and social factors and seeing its function not restricted
to mere content-related results. At least equally important is the foresight
process by which the central elements of the “five Cs” – “communication,
concentration on the longer term, co-ordination, consensus and commit-
ment” (Martin, 1995, p. 144) all have an effect above all in promoting net-
working, coordination and implementation. In this way, the focus is on the
communication processes, the consultative components and the feedback
to technology policy actors as elements of the desired self-learning system.
Over the last 15 years, a considerable differentiation of the field in both
conceptual and thematic respects can be observed, as reflected in the var-
iety of alternatives such as “Adaptive Foresight” (Eriksson and Weber, 2008),
“Regulatory Foresight” (Blind, 2008), or “Sustainability Foresight” (Truffer
and others, 2008).

12.3.2 The role of Delphi

All foresight activities rely on the estimates of experts. In the light of the
open definition of the term expert, a variety of degrees of factual knowledge
is mostly distinguished, operationalized correspondingly in various grades
and investigated empirically. The methodological approaches to obtaining
informative statements about long-term technology trends and the oppor-
tunities they can be expected to offer as a result at national level vary. The
most important methods include Delphi surveys, scenario analyses, commit-
tees of experts, trend analyses, relevance tree analyses and morphological
approaches (Johnston, 2001, p. 718). Mention should also be made of lists
of critical technologies and workshops. Frequent application is made firstly
of mostly broadly designed Delphi surveys amongst experts and secondly
analyses by expert committees. More recent studies combine elements of
both approaches or apply Delphi together with other methods such as Cross
Impact Analysis (Scapolo and Miles, 2006) or Technology Roadmapping
(Kanama and others, 2008). Other attempts such as using the potential of
social online networks for the fundamental objectives of foresight – the
opening up of creativity, expertise and collective intelligence – are still at a
rather explorative stage (Cachia and others, 2007).
256 Georg Aichholzer

In a number of applied studies in the field of technology foresight, Delphi

methods have hitherto achieved an acceptable accuracy level in the assess-
ment of technological developments (Grupp, 1995, p. 53, Ono and Wedemeyer,
1994). Delphi appears to be particularly appropriate as a result of the fact that
the variety of important factors and corresponding knowledge requirements
(technical, commercial, political, social and so on) in this field can only be
partially covered by individual experts and the latter therefore profit from
the simulated reciprocal communication. The field of application of technol-
ogy foresight itself is moreover, irrespective of the methods used and despite
the observable advances, not free from criticism: objections are raised inter
alia to a fixation on the supply of technologies, the continued influence of
technological determinism and the ignoring of progress in more recent social
science technology research (Knie, 1997, p. 227). Other objections concern
the questionable transfer of the approach and issues of Japanese technology
foresight, the domination of the centralized instead of decentralized deter-
mination of the points of focus and the technical “expertocracy” that results,
as well as the neglect of steps to ensure a greater focus on demand and needs
(Tichy, 1997, p. 198f.). Accordingly, attempts are continually being made to
achieve methodological and conceptual adaptations and improvements.
The following intends to show the design and implementation of an
Expert Delphi in the light of an example. Technology Delphi Austria rep-
resents an innovative autonomous alternative form of technology foresight
that attempts to take account of more recent progress in theory and to com-
bine the need situation specific to the country with political relevance.

12.4 Technology Delphi Austria: an application example

12.4.1 Objectives and project design

Technology Delphi Austria was a central element of the first technology
foresight project carried out at national level in Austria. The main objective
was to explore global technology trends with regard to relevance and oppor-
tunities for Austria and to determine niches with future potentials. This
meant identifying innovation opportunities through which Austria could
achieve topic leadership in the long-run, meet with economic demand and
respond to the need to solve societal problems. At the same time the fore-
sight process should stimulate the actors of and those affected by innov-
ation projects to deal with different future developments the realization of
which would be essentially shaped by their own decisions. This should con-
tribute to the self organization of the national innovation system. Hence it
was imperative to develop an approach tailored to Austria’s situation which
concentrated on some key areas, taking account of demand side aspects,
including a parallel exploration of societal trends and providing a decided
bottom-up generated, implementation-relevant input to a more long-term
oriented technology policy.
 The Delphi Method 257

The Institute of Technology Assessment (ITA) of the Austrian Academy of

Sciences was entrusted with the execution of the Technology Delphi and
developed a foresight design with some innovative elements. These were
in particular a decided problem-, demand- and implementation-oriented
approach, a conception as “decision Delphi,” an emphasis on bottom-up
generated input and the combination with a Society and Culture Delphi.
The approach also included a combination of methods and, being a pilot
project for Austria, it deliberately concentrated on a number of selected areas.
Thus the Technology Delphi consisted of a whole set of Delphi projects in
seven subject fields involving a broad range of experts. The Delphi method
was employed as a form of interviewing these experts in a basic round and
two main rounds: the basic round consisted of expert panels of two dozen
persons at most per subject area, whereas the two main rounds comprised
a considerably larger circle of experts of up to 300 participants per subject
area.4 The overall organization of the expert Delphi is shown in Table 12.1.

Table 12.1 The Austrian Technology Delphi: organization, roles and tasks

Sponsor Organizer/Monitor group Experts

Ministry of Science Institute of Technology Expert panels in seven

and Transport Assessment (ITA) subject fields:
commissions 1. Sustainable buildings
foresight project 2. Lifelong learning
“Delphi Austria” 3. Medical technology/
self-help
4. Sustainable production
5. Organic food
6. Physical mobility
7. Quality defined materials
Establishment of Project preparation: preparatory
steering studies and basic project
committee design (for example review
of major technology Delphi
studies, scenarios of global
technology trends, strength/
weakness analysis of the
Austrian R&D system,
exploratory expert survey for
identification of priorities and
co-nomination of experts,
citizen survey on attitudes
towards technology); selection
of seven subject fields;
establishment and briefing of
expert panels for each field;
concept for Delphi surveys

Continued
258 Georg Aichholzer

Table 12.1 Continued

Sponsor Organizer/Monitor group Experts

Organization of expert panels Basic round of expert

for Delphi basic round; Delphi:
development of Delphi Expert panels in each
questionnaires for each subject field develop
subject field; organization of theses for subsequent
pre-tests Delphi main rounds
Organization of Organization of expert Delphi Two main rounds of
regular meetings; in two main rounds (expert expert Delphi survey
discussion of data base, technical design, among extended expert
interim reports by printing, mailing, collection circle (in each subject
ITA; co-ordination of completed files) field)
of two combined Statistical analysis for feedback
Delphi processes in second main round
(Technology
Delphi together
with Society and
Culture Delphi)
Statistical analysis of final Discussion of draft reports
results and elaboration of among expert panels and
draft reports comments on results in
each subject field
Use of results as Final reports with results Individual follow-up
decision basis of expert Delphi; public activities in some expert
for (technology) presentation; publication and panels
policy dissemination activities
programmes;
implementation
measures

The Ministry of Science and Transport as the sponsor of Delphi Austria

established a steering committee, which also cared for the integration of the
overall project carried out by three independent teams, ITA being responsible
for the Technology Delphi. Strictly speaking, this was not one Technology
Delphi but a set of seven Delphi projects carried out in seven subject areas
in parallel. (In addition, one other research team was responsible for a simi-
lar structure of Delphi projects as part of the Society and Culture Delphi).
The methodological decision in favour of broadly conceived Delphi surveys
was taken very early mainly for two reasons: strengthening the cooperation
between business, government and science was seen as imperative and
expert committees alone were considered to be less suitable because of the
danger of dominance by single experts (Tichy, 1997, p. 201). The determin-
ation of subject areas deliberately aimed at a problem-oriented rather than a
pure technology-oriented conception and built on a number of preparatory
 The Delphi Method 259

studies. Based on these and a set of priority criteria seven problem-oriented

subject areas were finally selected for the Technology Delphi (ITA, 1998,
vol 1, p. 30ff.).

12.4.2 Organization as decision Delphi

The three generic types of Delphi introduced by Rauch (1979) – classical,
policy, and decision Delphi – suggested the design of the Austrian Technology
Delphi towards a decision Delphi (Tichy, 2001). The classical Delphi is a
means for achieving a group opinion based on anonymous, multiple-level
group interaction on a principally factual basis. A precondition for allowing
assessments according to the pattern of a conditional scientific prognosis is
that these can essentially build on some underlying laws and regularities.
A policy Delphi, in contrast, is a tool for revealing the views of lobbyists on
political issues. It is a matter of ideas and concepts, and not of future data
and facts or of decision mechanisms.
A decision Delphi, finally, is an instrument which allows the preparation
and influencing of decisions and hence processes in society where develop-
ments are not governed by whatever kind of regularity or law but evolve out
of countless small, decentralized and uncoordinated actions: “In a decision
Delphi reality is not predicted or described; it is made. ... Decision Delphi,
however, has to be seen as much more than a simple self-fulfilling prophecy.
Its main social function could be to coordinate and structure the general
lines of thinking in a diffuse and unexplored field of social relations, and to
transfer the future development of such an area from mere accident to care-
fully considered decisions. ... The Delphi feedback serves as a major source of
information in this process. Some elements of brainstorming are in this way
taken over by Delphi” (Rauch, 1979, p. 163ff.).
Given these characteristics of the decision Delphi it seemed to be predes-
tined for the Austrian Technology Delphi: after all this foresight project was
a matter of developments which are shaped by a multitude of individual
decisions by actors of the innovation system in a wide variety of segments
of society. What was of key interest was the foresight process, particularly
its contribution to facilitating coordination and consensus finding among
decision-makers.
This was supported in several ways: on the one hand, through opinion
forming among the respondents of the two main rounds of the expert
Delphi survey (1,638 in round 1 and 1,127 in round 2), many of the partici-
pants being decision-makers at various levels. Here the main mechanism
allowing for virtual exchange and individual adjustment of assessments was
the feedback of group response on each innovation statement and suggested
measure from round 1 in round 2 of the Delphi survey. On the other hand,
exchange and coordination of views occurred through the (only partially
anonymous)5 interaction in the much smaller expert panels of the basic
round (14 – 23 persons per subject area, in total 128). These panels consisted
260 Georg Aichholzer

primarily of decision-makers who not only defined the innovation state-

ments for the Delphi questionnaires but also later on commented on the
final results and drafted recommendations. Debating the replies of the
extended group of experts on questions formulated by the smaller panels
in the basic round meant a further cause for exchange and coordination of
opinions. This set-up was deliberately aimed so that the experts who nor-
mally take numerous and largely uncoordinated small decisions in their
professional roles were each confronted with the different views of their
peers. All this should help the Technology Delphi acquire a dynamic char-
acter in the sense of a feedback loop on the decisions of the participants in
the expert panels and Delphi surveys and contribute “to create the future in
reality rather than just predicting it” (Rauch, 1979, p. 159).

12.4.3 Selection of experts

The selection of the “right” experts is regarded as one of the main meth-
odological problems (Häder and Häder, 2000, p. 18f.). This revolves above
all around the appropriate subject knowledge, motivation and influence on
the practical implementation of the results and the composition and size of
the group. Technology Delphi thus involves a twofold selection: firstly the
experts in the basic round who develop the content for the Delphi survey
carried out using standardized questionnaires in two further rounds, and
secondly the much larger group of experts included as respondents in these
two main rounds.
The aim was to achieve a high level of competence for the assessment of the
prospects, opportunities and consequences of a wide variety of (technical
and organizational) innovations and at the same time to eliminate as far as
possible the problems of interest-related points of view and the narrowing
of perspectives that results from specialisation. The risk of faulty progno-
ses resulting from “déformation professionelle” has long been identified:
“We find a curious ahistoricity in the outlook of most scientists and tech-
nologists, together with a tendency for inbreeding” (Linstone, 1978, p. 298).
There is also evidence of a tendency towards an optimism bias amongst
experts (Tichy, 2004).
For this reason, the Technology Delphi by no means only involved scien-
tific and technical experts but also experts from business, the state admin-
istration, the social and economic sciences, professional interest groups,
social movements, NGOs, and user representatives in the broadest sense.
The assessment of the chances of implementation and the social desirability
of individual innovation projects were also investigated and taken appro-
priately into account in the overall assessment. It was precisely because the
objective of Delphi Austria was to regard subject areas under investigation
more as problem fields than as purely technology fields that account was
taken of correspondingly more broad-ranging competencies in the defin-
ition of the expert group. In addition to expertise in the core technological
 The Delphi Method 261

field, the aim was to include the knowledge necessary to assess problem-
solving capacities, and to explore opportunities for topic leadership as well
as the institutional preconditions.
Given this objective, the circle of experts to be surveyed in every special-
ist field extends into a large number of different areas of competence that
are applied in the individual stages of the innovation cycle – from familiar-
ity with area-specific problems and needs through scientific and technical
expertise to market-side and socio-economic know-how. The questionnaires
were adjusted to the involvement of a deliberately heterogeneous skills base
inter alia by allowing each question to be answered or omitted according to
the degree of the individual person’s subject knowledge. For the subsequent
analysis, only responses with at least average subject knowledge were taken
into account.
The main selection principle was to specifically include and take account
of the diversity of the fields of competence of relevance for the identifica-
tion, development and use or exploitation of technical and organizational
innovations. This principle was operationally implemented by determining
corresponding target areas or institutional contexts as the basis for the choice
of experts. These were grouped into the following three basic categories:

● scientific research (universities and non-university research institutions,

broken down according to natural science / technical and social science
disciplines);
● business (broken down according to industry, production and services);
● administration and federations (public administration and research pro-
motion institutions, interest, consumer and user representatives, and
NGOs).

Persons with subject knowledge were selected for the expert survey for the
seven Delphi fields in a way that as far as possible achieved a balanced dis-
tribution over the three basic categories. Within the fields, the concrete
competence areas and institutions attributable to the general categories of
technology/science, social science, industry, services, public administration
and user representation needed to be identified: in the field of (technolo-
gies for) lifelong learning, for instance, these were in the category business
(producers and service providers), further education and distance learning
institutions, personnel consulting companies, persons responsible for in-
company training in major enterprises, the Austrian employment service,
evening schools, publishing houses, the producers of educational software,
postal and telecommunications facilities, network operators and providers
of internet services, computer companies, the ORF broadcasting company,
educational journalists and similar units. The field-specific characteristics
were identified analogously in the other categories and taken as the basis
for the selection.
262 Georg Aichholzer

Table 12.2 Size of the two main Expert Delphi rounds

Round 1 Round 2

Return Return
Field Experts rate % Experts rate %

Sustainable buildings 219 43,2 142 66,7

Lifelong learning 309 52,2 219 73,0
Medical technology / self-help 191 41,2 139 73,5
Sustainable production 313 41,6 211 70,6
Organic food 183 43,0 126 71,6
Physical mobility 300 50,3 200 69,9
Quality defined materials 123 50,0 90 75,0
Total 1638 45,7 1127 71,2

Source: ITA (1998, vol. 1), pp. 72–73.

With respect to the optimum size of Delphi groups, recommendations vary

considerably (cf. Häder and Häder, 2000, p. 18ff.). In the opinion of Parentè
and Anderson-Parentè (1987, p. 149ff.), there is in fact no fixed upper limit;
however, at least ten is recommended as the lower limit, after deduction of
possible dropouts. For this reason, a size suitable for group discussions was
aimed at for the working groups in the (less strongly structured) basic round.
In expectation of a drop-out rate of up to 50 per cent for each field, around two
dozen experts were invited; the effective group size was ultimately between 14
and 23 per field. For the considerably larger circle of experts in the two main
Delphi rounds, the target size was based more on the number of cases that
would be sufficient for statistical analysis, the average expected return rates
and the dropout rate, which depends inter alia on the degree of specialization
or the heterogeneity of the area. The operational selection criteria at individ-
ual level were a number of indicators for maximum subject competence and
high professional reputation such as nominations by peers, relevant manage-
ment functions or a professorial chair, membership in leading bodies, receipt
of subsidies and subject-related awards. The effective participation in the two
main rounds of the Expert Delphi is shown in Table 12.2.

12.4.4 Development of Delphi statements and questionnaires

In the same way as the selection of the experts, the selection of meaningful
and intelligible scenarios – in the foresight context the terms statements
and theses are also used – are also among the key questions of a Delphi
application. Both creativity and methodological experience are required.
According to the recommendations by Parentè and Anderson-Parentè (1987,
p. 149f.), a Delphi statement should not be longer than 20 words, should be
easy to verify or specific and should not exceed a total of 25 statements for
one expert group.
 The Delphi Method 263

In the Technology Delphi, the theses describe specific innovations and

their future development states. They were developed by the expert panels
in the basic round. For each of the seven fields of investigation, this was
done in the form of chaired workshops led by two members of the ITA team
and assisted by communication via email. Following an introduction to the
objectives and methods, the participants were given information mater-
ial and secondary analysis results on international technology trends and
familiarized with their task. This consisted essentially in identifying the
technical and organizational innovations that are most likely to achieve
Austrian topic leadership over the next 15 years in each of the seven areas.
These high-opportunity innovations were to be formulated, in the form of
theses, as a description of the development stage to be expected in 15 years,
the hypothetical state being expressed as a gradation from development
up to general dissemination. An example of such an innovation thesis from
the field of medical technology and self-help for older people read for
instance as follows:

Biosensors (antibody sensors) are being developed that detect specific

allergens in the surroundings (air, water) and for instance permit asthma
warnings.

On the basis of the theses on innovations and measures developed by the

expert working groups in the basic round, the ITA team designed a separate
questionnaire with the following basic structure for each of the seven fields.
This was posted to a much broader circle of experts for an assessment in the
above-mentioned two main Delphi rounds for each field.
The following questions were put for each thesis:

● level of subject knowledge of the respondent,

● degree of innovation,
● importance,
● probability of implementation in Austria within the next 15 years,
● the chances of topic leadership for Austria with respect to (a) research
and development, (b) organizational and social implementation and
(c) commercial exploitation,
● desirability of the development.

The first four questions were assessed using a five stage assessment scale based
on the Austrian school grading system (1 = very high or positive, 5 = very
low or negative), the others concerned agreement/disagreement; multiple
responses were permitted for the “chances.”
In addition, a list of measures proposed by the experts in the basic round
was checked for its ability to increase the Austrian chances of success for
promising innovations. However, this was not done for each individual
264 Georg Aichholzer

innovation thesis but for groups. These were concrete individual measures
in the following seven categories: research-related, technological, commer-
cial, regulatory, cooperation-related, training and further training related
and society-related. A proposed measure in the technical category in the
field of “lifelong learning” was for instance:

Offering information about certification issues (requirements, certifica-

tion offices, organisational process) using information and communica-
tion technology.

Measures proposed were also subjected to the five-level assessment scale.

For each thesis and list of measures there was in addition a corresponding
space for comments. Since the range of possibilities for innovations and
support measures suggested by the experts in the basic round was not to be
regarded as absolutely complete, the respondents were also asked in an open
question to propose any further innovations as well as measures with which
Austria might have even better chances.
Finally, the questionnaires on all the subject fields ended with a stand-
ard list of 17 statements on “megatrends” enquiring firstly about basic ten-
dencies in the field in question and secondly the assessment of trends in
social, commercial, political and ecological conditions. In this way, the
innovations and developments raised in individual fields were placed in the
greater context of the surrounding influences and at the same time it was
possible to investigate the range of interpretation patterns, perspectives and
subjective world images amongst the Delphi experts. However, this framing
with assessing megatrends was only implemented in the first main round
(for more details see Aichholzer, 2000, p. 81f.).
Overall, the Technology Delphi thus comprised 30 to 42 innovation the-
ses to be assessed in each of the seven fields of investigation (hence a total
of 271), as well as 10 to 24 measures theses and questions on 17 megatrends
in each theme block.

12.4.5 Encouragement of the motivation to participate

Achieving the maximum participation across all the rounds of the Delphi
process and in each of the thematic fields constitutes a particular challenge
for a national Technology Delphi. In order to secure a sufficient number
of respondents with a high subject knowledge including at the level of the
individual theses (or innovations) in the light of the wide variety of special
expertises required, the Delphi groups must be relatively large and in par-
ticular efforts must be made to secure a maximum return rate. Finally, there
is also the problem of motivating experts to participate in the Delphi. In the
foresight sector, a number of incentives can be identified whose applica-
tion can be differentiated depending on the methods used and the stake-
holder category (Salo, 2001, p. 698f.): exercising influence, using learning
 The Delphi Method 265

opportunities, developing contacts, demonstrating loyalty, receiving com-

pensation. With the Delphi method, the main stimuli are seen to be in the
learning and influence possibilities. Both were appropriately raised in the
invitation to the experts, above all by emphasizing the intention of imple-
menting the results at the highest level of technology policy, namely that
of the minister.
In securing motivation, the Technology Delphi largely relied on a pro-
cedure that had been successfully carried out in postal survey research, the
Total Design Method (TDM) introduced by Dillman (1970). The basic idea is to
design each element of the survey and to coordinate the individual aspects
in such a way that maximizes the quality and quantity of the responses
(Hippler, 1988, p. 246). Two of the basic principles applied in implementa-
tion are:

● minimizing the costs of participation and the answering of the ques-

tions for the respondent, while at the same time maximizing the visible
advantages;
● creating a relationship of trust between the researcher and the
respondent.

Accordingly the aim was to design each stage of the survey in such a way
that it satisfied these conditions as far as possible:
One essential measure was that the design of the questionnaires involved
a reduction of the often extremely large scope in other foresight studies.
While for instance the German (Cuhls and others 1998) and British Delphi
studies (Loveridge and others, 1995) involved the assessment of over 100
theses according to more than a dozen criteria in a large number of areas,
the scope per field of the Austrian Technology Delphi was limited to around
30 to 40 theses that were only to be assessed according to six criteria (plus
a list of measures with up to two dozen items per list but which were only
assessed according to one criterion, the degree of suitability).
This type of proposed measures is, as compared with the forms previ-
ously used in technology forecasting, a further innovation, characterized
by a considerably larger extent of specificity, that was also intended to
have a positive effect on motivation. Instead of “commercial and political
measures,” a typical proposed measure in the field of “environmental pro-
duction and sustainability” was for instance “granting specific subsidies to
small and medium-sized enterprises.”
Other specific TDM recommendations implemented in the design of the
questionnaires concerned the technically uniform design and structure, the
ease of completion by means of a single scale (based on the Austrian school
grading system), the avoidance of filters and response coding, the invitation
to personal comments, an attractive graphic design, printed in brochure
form and on good quality paper.
266 Georg Aichholzer

The implementation of the survey as a central design item of the TDM

concept followed a series of special preparatory measures: a personal let-
ter to the potential expert respondents by the director of the institute
organizing the Technology Delphi (personal address in the letterhead;
accompanying text with the explanation of the importance, the expected
benefits, reward through early receipt of the results; personal signature);
official letter from the Minister for Science with a request for participation
in the Delphi survey; guarantee of confidentiality through the possibility
of anonymous responses; enclosed envelope for post-paid return; mid-week
mailing; a polite reminder card after three weeks; telephone follow-up in
the fifth week (in the second round, two waves of telephone follow-ups
in the fourth and fifth weeks), if necessary with a repeat mailing of the
questionnaire; joint letter of thanks by the institute carrying out the study
and the Ministry for Science.
This approach made a major contribution to the higher degree of accept-
ance and willingness to co-operate amongst the experts addressed. A num-
ber of indicators can also be cited as criteria for quality:

● The good response level was expressed not only in the return ratio of
46 per cent in the first and 71 per cent in the second round, which were
above average for the field of a Technology Delphi.
● The high level of willingness to comment and the quality of the contents
of the comments on the individual innovations and proposed measures
(resulting for each field in several dozen pages of comments) is a further
indication.
● The rate of item non-response, even taking into account the responses
not considered for lack of sufficient subject knowledge, is relatively low
at an average of less than 25 per cent (even the lowest number of valid
responses, to a highly specialized innovation in the field of materials, still
amounted to 33).
● The response rates across the questionnaires are relatively constant in all
fields and do not show any clear decline towards the end.
● Explicit refusals, finally, only amounted to a marginal percentage of the
dropout rate.

12.4.6 Feedback and analysis

If account is taken of the basic round with its lower degree of structure,
its face-to-face approach and considerably smaller size, but nevertheless of
central importance for the generation of the Delphi content (cf. Rowe and
Wright, 1999, p. 354), the Austrian Technology Delphi was conducted in a
total of three stages. The first stage, referred to as the basic round, is clearly
distinguishable from the following two main rounds in terms of function
and form. The Delphi survey proper with standardized questionnaires was
 The Delphi Method 267

carried out with a considerably larger number of participants in the two

main rounds over a period of roughly four months. The first round was
carried out in June and July, and the statistical interim analysis for the feed-
back for the second round was made in August, the second round extend-
ing over September into October. Together with the questionnaire for the
second round, the respondents were given feedback in the form of the mean
values (arithmetic mean) from the individual questions of the first round,
with the result that these group responses could be taken into consideration
by the experts in their second assessments. With an average return rate of
71.2 per cent and a total of 1,127 remaining participants in the second and
final round, panel dropout remained very much within limits. The restric-
tion to two rounds took account firstly of the experience that major changes
to the experts’ decisions are limited to the initial rounds (Rowe and Wright,
1999, p. 372), and secondly of the expectation that motivation would decline
considerably with each further round.
The final results were analysed in two stages: an initial overall ana-
lysis resulted in a Technology Delphi Volume 1 (objectives, investigation
approach, and overall results). In the second stage, the results were analysed
in detail according to the seven investigation areas. The results, summa-
rized in a draft report, were discussed with the experts of the basic round in
the concluding workshop and then set out in Technology Delphi Volume 2.
A further technology Delphi Volume 3, finally, contains the tables with stat-
istical data in the form of frequency counts on all questions.
These results of the Technology Delphi summarized in the three report
volumes (ITA 1998) took account of the variable “subject knowledge” in the
following form: the analysis only included the responses of experts who
stated that they had an “average,” “rather high” or “very high” subject know-
ledge of the question concerned. Conversely, this meant that the responses
by respondents who stated that their subject knowledge was only “rather
small” or “very small” was excluded. This approach appeared appropriate,
since it has been shown that a restriction to the highest expertise level does
not necessarily lead to more valid results (Parentè and Anderson-Parentè,
1987, p. 137), while at the same time a minimum level of information about
the topic in question is essential.

12.4.7 Results and implementation

The results of this example application are only set out briefly as illustra-
tions here: alongside the wealth of concrete innovation opportunities (for
example high-tech steels and lightweight materials or the low noise railway
project), a number of broader insights can be highlighted, such as the good
opportunities for Austria’s topic leadership in particular in the application
of high – although not always highest – technology in fields of medium
technology penetration and on Austria’s lead markets (for example environ-
ment technology, organic food). At the same time, deficits became apparent,
268 Georg Aichholzer

such as too short innovation horizons and the ambivalent attitude to organ-
izational innovations.
The objective of providing implementation-relevant results for technology
policy at national level was achieved. This is demonstrated for instance by
the implementation measures observable after completion of the Technology
Delphi. These included systematic focus and stimulus programmes in sub-
sidy policies such as an “Austrian Program on Technologies for Sustainable
Development”, with the subprograms “Building of Tomorrow,” “Energy
Systems of Tomorrow” and “Factory of Tomorrow.”
Three results can be ascribed to the use of the Delphi method in the
Austrian Technology Delphi: the experience of a social process that is of
value, acts as a focus, promotes information transfer and contributes to the
networking of the national innovation system, the generation of relevant
results and the achievement of practical political effects.

12.5 Delphi and method innovation

The following shall first illustrate the use of the Expert Delphi as described in
combination with other methods in the light of two international examples.
The second part examines fundamental methodological innovations based
on the internet.

12.5.1 Combinations of methods

An interesting combination of Expert Delphi and Cross-Impact Analysis (SMIC,
a French acronym for interactive systems and matrices) that also includes
a comparison of methods is provided by a study analysed by Scapolo and
Miles (2006) about future transport systems and the role of advanced
transport telematics in European cities. This project used both methods as
quantitative approaches to identify the knowledge of the same expert base,
namely to verify and improve the validity of the Delphi results. In par-
ticular, the SMIC method, by identifying interdependency, was intended to
compensate for a weakness in the Delphi technique, namely the possibility
of events that mutually amplify or exclude each other, or the generation of
an “artificial” consensus. SMIC served not as an instrument to refine prob-
ability assessments but rather to clarify complex causal relationships that
ultimately led to a fundamental progression of possible scenarios, thereby
offering additional information to decision makers in the assessment of a
problem.
The cornerstones of the study design were as follows: from a pool of 300
international experts, three subgroups were formed with the aim of having
one available for the two-round Delphi, one for the SMIC method and one
to reply to both instruments, and, after conclusion of the investigations, to
complete an evaluation questionnaire. The assessment criteria for the indi-
vidual Delphi statements focused on the extent of use of transport telematics
 The Delphi Method 269

over the next 20 years, their effects on transport volume and environment
and the influence of economic, social, legal and political factors. The num-
ber of individual questions for the relatively flexible Delphi method was
ultimately set as a list of 19 statements, while the SMIC approach is gener-
ally limited to a maximum of six events. A typical SMIC statement reads:

“51–70 per cent of European medium size cities use automated sys-
tems to monitor traffic in real-time which are able to provide real-
time on-board journey information, as well as congestion level and
incidents warnings.” (Scapolo and Miles, 2006, p. 694)

While in the first stage the probability of an event was to be assessed

individually, in the second stage the task was to consider them in pairs in a
matrix and to assess conditional probabilities. Further details of the opera-
tionalization of the Delphi and SMIC methods are summarized in the art-
icle together with reflections on design decisions.
The main result of a combination of methods and their evaluation is as
follows: as far as concerns usefulness, the assessment is overall positive, and
the experts regard the process as a valuable means of communication for
the exchange of opinion, in particular the Delphi method with the inform-
ative feedback of group opinion. However, not all the respondents appreci-
ated one of the characteristics of the Delphi method, specifically the several
rounds of the survey, and regard this as a pressure towards the conver-
gence of opinions. The failure to take account of possible interdependen-
cies between the individual statement contents is also seen as a deficiency.
The SMIC method is generally experienced as being more difficult; its main
advantage, the assessment of interdependencies, is regarded as complex and
time-consuming. On the other hand, its accuracy is seen as a considerable
advantage. Overall, the implementation of the two methods raises many
similar questions to which as yet insufficient attention has been paid and
which emphasize the need for practical methodological guidelines.
The consensus pressure in the Delphi method also gave rise to a search for
a complementary methodological reinforcement of the aspect of diversity.
A Finnish research group developed a foresight method that combines consen-
sual elements for the setting of priorities, networking and the development of
visions with complementary elements such as controversial ideas to increase
diversity and to encourage innovation. The RPM Screening6 method is based
on an invitation to experts to generate innovation ideas and was used in a
foresight project in Finland by the Minister for Trade and Industry (Könnölä
and others, 2007). In the pilot project, around 50 experts from each of five
stakeholder groups (industry, government, research, trade and NGOs, tech-
nology enterprises and investors) were invited to jointly generate, revise and
evaluate specific innovation ideas on three topic areas – nutrigenomics, health
and social services, and services for personal experiences. This was done via
270 Georg Aichholzer

internet-assisted instruments based on RPM screening in conjunction with a

workshop meeting. Sixty postgraduate students on a decision analysis course
were also included. The selection decision was based so as to ensure diversity
in three carefully differentiated dimensions: variety, balance and difference.
The RPM screening method relates above all to the analysis of the innovation
ideas collected, the screening being based on the systematic linking and math-
ematical modelling of a consensus-based and a dissent-based approach (for
further details see 8ff.). The pilot project generated 166 innovation ideas, some
of which were used in the Delphi process of a regional foresight project.

12.5.2 Online and Real-Time Delphi

The first approaches using the Delphi method with computer systems and
networked communication can, according to Gordon (2008, p. 160), be
dated back to 1970, when Murray Turoff organized a computer conference
on the basis of the ARPA network and the term “collaborative intelligence” was
already being used in conjunction with Delphi-like methods. Efforts to use
the new possibilities of the internet for a modified and more efficient Delphi
technique finally led to a form that in certain respects can be regarded as
method innovation, above all because the feature of consecutive/iterative
survey rounds is eliminated.
This internet supported “round-free” approach developed by Theodore
Gordon and Adam Pease is called Real Time or RT Delphi (Gordon and Pease
2006). “Real time” refers to the fact that the state of responses to the Delphi
questions is updated and accessible online practically in real time. The RT
Delphi is regarded as being particularly suitable for synchronous participa-
tion, a small number of participants and the need for rapid completion; it can,
however, as a matter of principle also be used for asynchronous participation
if the number of participants is larger and more time is available. If a partici-
pant joins an ongoing RT Delphi with numeric response codes, the following
information is shown on the screen for each question: the average or median
of the group answers so far plus any distribution parameters; the number of
previous responses; a button that opens a window to show the justifications
for the other participants’ responses; another button that opens a window
for entering the respondent’s own justification for his responses; and finally
a space for the respondent’s own assessments on the individual questions.
This alternative does not involve an explicit second round, and instead the
process is restricted to one round. The organizer of the process can select the
time limit and invite participants to make repeated visits. Participants can
return to their current responses at any time and amend or retain them in
the light of information about the other participants’ responses. Anonymity
and feedback are guaranteed. The type of questions can relate both to indi-
vidual features of future developments and to matrix-like question formats
such as for benefits matrices or cross-impact matrices. Basic alternatives of
the RT Delphi are firstly its use with persons present simultaneously in a
 The Delphi Method 271

conference room setting and secondly a group of participants linked via an

internet platform and distributed at random geographically. The authors also
briefly present a number of available systems for an online Delphi.
The principal strengths of the RT Delphi are apparently its efficiency and
suitability for standard question formats and for matrix formats, its main
weakness is that it has only been subject to rudimentary trials and hence
there are still outstanding questions concerning quality assurance. The use
of an RT Delphi on global energy scenarios (Gordon, 2007) is regarded as
having succeeded in that the number of responses it produced was compar-
able to that from conventional alternatives.
Further light is shed by a similar comparison between two online versions
of the Delphi method (Zipfinger, 2007), one with a conventional series of
rounds and an RT Delphi with a single round. This produced a lower panel
dropout rate in the round-based Delphi, while the number of responses and
comments was greater in the RT Delphi. Furthermore, there was a greater
tendency to amend responses in the round-based Delphi, while the degree
of consensus in the RT Delphi was higher. The future will show whether
the author of this comparative test is right to predict that the round-based
Delphi will prove to be more workable in practice.

Notes
1. See also the comments by Ayton and others (1999) addressing additional aspects
and inter alia emphasizing the urgent need to include insights from social psych-
ology and cognitive psychology.
2. For an overview, see for instance Georghiou and others (2008) or the International
Journal of Technology Management (2001) Vol. 21, issue 7/8, Special Issue on
Technology Foresight.
3. “Foresight is part of the ever-present need to establish a ‘social contract’ between
researchers, government, and the public” (Rappert, 1999, p. 544). Similarly, Grupp
and Linstone: “Foresight ... (brings in) elements to moderate or negotiate between
the social interest groups. Foresight results provide the code to communicate
between social actors in science, technology, and society” (1999, p. 89).
4. Detailed documentation on the Technology Delphi is provided in a research report
comprising three volumes (ITA 1998). On the overall design of Delphi Austria and
the combination of the Technology Delphi with a Society and Culture Delphi see
Aichholzer (2001).
5. Proposals for Delphi statements were developed in expert workshops moderated
by ITA’s team in writing on small cards without visible individual origin and par-
tially amended via electronic communication.
6. Based on Robust Portfolio Modelling.

Further readings
Gupta, U. G. and Clarke, R. E. (1996) “Theory and Applications of the Delphi
Technique: A Bibliography (19751994)” in Technological Forecasting and Social
Change 53(2), 185–211.
272 Georg Aichholzer

Rask, M. (2008) “Foresight – balancing between increasing variety and productive

convergence” in Technological Forecasting and Social Change 75(8), 1157–75.
Rowe, G. and Wright, G. (2001) “Expert Opinions in Forecasting: The role of the
Delphi Technique” in Armstrong, J. S. (ed.) Principles of Forecasting: A Handbook for
Researchers and Practitioners (Berlin, New York: Springer), pp. 125–44.

References
Aichholzer, G. (2000) “Innovative Elemente des österreichischen Technologie-Delphi”
in Häder, M. and Häder, S. (eds) Die Delphi-Technik in den Sozialwissenschaften.
Methodische Forschungen und innovative Anwendungen (Wiesbaden: Westdeutscher
Verlag), pp. 67–93.
Aichholzer, G. (2001) “The Austrian foresight program: organization and expert
profile” in International Journal of Technology Management 21(7/8), 739–55.
Ayton, P., Ferrell, W. R and Stewart, T. R. (1999) Commentaries on “The Delphi
technique as a forecasting tool: issues ad analysis” by Rowe and Wright in
International Journal of Forecasting 15(4), 377–79.
Blind, K. (2008) “Regulatory foresight: Methodologies and selected applications” in
Technological Forecasting and Social Change 75(4), 496–516.
Blind, K., Cuhls, K. and Grupp, H. (1999) “Current foresight activities in Central
Europe” in Technological Forecasting and Social Change 60(1), 15–35.
Cachia, R., Compañó, R. and Da Costa, O. (2007) “Grasping the potential of online
social networks for foresight” in Technological Forecasting and Social Change 74(8),
1179–203.
Cagnin, C., Keenan, M., Johnston, R., Scapolo, F. and Barré, R. (eds) (2008)
Future-Oriented Technology Analysis: Strategic Intelligence for an Innovative Economy
(Berlin, Heidelberg, New York: Springer).
Cuhls, K., Blind, K. And Grupp, H. (1998) Delphi ’98 Umfrage. Studie zur globalen
Entwicklung von Wissenschaft und Technik (Karlsruhe: Fraunhofer Institut für
Systemtechnik und Innovationsforschung, unpubl).
Dalkey, N. C. and Helmer, O. (1963) “An experimental application of the Delphi
method to the use of experts” in Management Science 9(3), 458–67.
Dillman, D. (1978) Mail and Telephone Surveys: The Total Design Method (New York:
Wiley).
Engels, T. C. E. and Powell Kennedy, H. (2007) “Enhancing a Delphi study on family-
focused prevention” in Technological Forecasting and Social Change 74(4), 433–51.
Eriksson, E. A. and Weber, K. M. (2008) “Adaptive Foresight: Navigating the complex
landscape of policy strategies” in Technological Forecasting and Social Change 75(4),
462–82.
Georghiou, L., Harper, J. C., Keenan, M., Miles, I. and Popper, R. (eds) (2008) The
Handbook of Technology Foresight: Concepts and Practice (Cheltenham, UK: Edward
Elgar Publishing).
Gordon, T. J. (2007) “Energy forecasts using a ‘Roundless’ approach to running a
Delphi study” in Foresight 9(2), 27–35.
Gordon, T. (2008) “Book Review: Sabine Zipfinger, Computer-aided Delphi: An
experimental study of comparing round-based with real-time implementation
of the method. Johannes Kepler University, Linz, Austria (2007)” in Technological
Forecasting and Social Change 75(1), 160–64.
Gordon, T. and Pease, A. (2006) “RT Delphi: An efficient, ‘round-less’ almost real
time Delphi method” in Technological Forecasting and Social Change 73(4), 321–33.
 The Delphi Method 273

Grupp, H. (1995) Der Delphi-Report – Innovationen für unsere Zukunft (Stuttgart:

Deutsche Verlags-Anstalt).
Grupp, H. and Linstone, H. A. (1999) “National technology foresight activities around
the globe: resurrection and new paradigms” in Technological Forecasting and Social
Change 60(1), 85–94.
Häder, M. and Häder, S. (2000) “Die Delphi-Methode als Gegenstand methodischer
Forschungen” in Häder, M. and Häder, S. (eds) Die Delphi-Technik in den
Sozialwissenschaften. Methodische Forschungen und innovative Anwendungen (Wiesbaden:
Westdeutscher Verlag), pp. 11–31.
Hippler, H. J. (1988) “Methodische Aspekte schriftlicher Befragungen. Probleme und
Forschungsperspektiven” in planung und analyse 6, 244–48.
Irvine, J. and Martin, B. (1984) Foresight in Science: Picking the Winners (London:
Pinter).
ITA Institut für Technikfolgen-Abschätzung (1998) “Technologie Delphi” (3 Bände).
In Bundesministerium für Wissenschaft und Verkehr (ed.) Delphi Report Austria 1–3
(Wien: Bundesministerium für Verkehr, Innovation und Technologie).
Johnston, R. (2001) “Foresight – refining the process” in International Journal of
Technology Management 21(7/8), 711–25.
Kanama, D., Kondo, A. and Yokoo, Y. (2008) “Development of technology foresight:
integration of technology roadmapping and the Delphi method” in International
Journal of Technology Intelligence and Planning 4(2), 184–200.
Keeney, S., Hasson, F. and Mc Kenna, H. P. (2001) “A critical review of the Delphi
technique as a research methodology for nursing” in International Journal of Nursing
Studies 38, 195–200.
Knie, A. (1997) “Technik als gesellschaftliche Konstruktion. Institutionen als
soziale Maschinen” in Dierkes, M. (ed.) Technikgenese (Berlin: edition sigma),
pp. 225–43.
Könnölä, T., Brummer, V. and Salo, A. (2007) “Diversity in foresight: Insights from
the fostering of innovation ideas” in Technological Forecasting and Social Change
74(5), 608–26.
Landeta, J. (2006) “Current validity of the Delphi method in social sciences” in
Technological Forecasting and Social Change 73(5), 467–82.
Linstone, H. A. (1978) “The Delphi Technique” in Fowles, J. (ed.) Handbook of Futures
Research (Westport: Greenwood Press).
Loveridge, D., Georghiou, L. and Nedeva, M. (1995) United Kingdom Technology
Foresight Programme: Delphi Survey (Manchester: Policy Research in Engineering,
Science and Technology, PREST, The University of Manchester, unpubl.).
Major, E., Asch, D. and Cordey-Hayes, M. (2001) “Foresight as a core competence” in
Futures 33(2), 91–107.
Martin, B. R. (1995) “Foresight in Science and Technology” in Technology Analysis &
Strategic Management, 7(2), 139–68.
Martin, B. and Johnston, R. (1999) “Technology foresight for wiring up the national
innovation system” in Technological Forecasting and Social Change 60(1), 37–54.
Ono, R. and Wedemeyer, D. J. (1994) “Assessing the validity of the Delphi Technique”
in Futures, April, 289–304.
Parentè, F.J. and Anderson-Parentè, J.K. (1987) “Delphi inquiry systems” in Wright, G.
and Ayton, P. (eds) Judgemental forecasting (Chichester: John Wiley), pp. 129–56.
Rappert, B. (1999) “Rationalising the future? Foresight in science and technology
policy co-ordination” in Futures 31(6), 527–45.
Rauch, W. (1979) “The Decision Delphi” in Technological Forecasting and Social Change,
15(2), 159–69.
274 Georg Aichholzer

Rongping, M., Zhongbao, R., Sida, Y. and Yan, Q. (2008) “Technology foresight
towards 2020 in China: the practice and its impacts” in Technology Analysis &
Strategic Management 20(3), 287–307.
Rowe, G. and Wright, G. (1999) “The Delphi technique as a forecasting tool: issues
and analysis” in International Journal of Forecasting 15(4), 353–75.
Salo, A. A. (2001) “Incentives in technology foresight” in International Journal of
Technology Management 21(7/8), 694–710.
Scapolo, F. and Miles, I. (2006) “Eliciting experts’ knowledge: A comparison of two
methods” in Technological Forecasting and Social Change 73(6), 679–704.
Tichy, G. (1997) “Technologieprognosen und Technologiepolitik” in Wirtschaft und
Gesellschaft 23(2), 193–209.
Tichy, G. (2001) “The decision Delphi as a tool of technology policy – the Austrian
experience” in International Journal of Technology Management 21(7/8), 756–66.
Tichy, G. (2004) “The over-optimism among experts in assessment and foresight” in
Technological Forecasting and Social Change 71(4), 341–63.
Truffer, B., Voß, J. P. and Konrad, K. (2008) “Mapping expectations for system trans-
formations: Lessons from Sustainability Foresight in German utility sectors” in
Technological Forecasting and Social Change 75(9), 1360–72.
Zipfinger, S. (2007) Computer Aided-Delphi. An Experimental Study of Comparing
Round-Based with Real-Time Implementation of the Method (Linz: Trauner).
Index

Abels, Garbiele, 9, 138 communication

access problems, 104–5 everyday, 205
accomplice, interviewer as, 67, 68–9 practices, 27–8
accuracy standards, 248n7 rules of, 11
active participants, 24–5, 38n12 as sensemaking process, 218–19
Aichholzer, Georg, 12, 252 communicative validity, 186
analysis, 35–6 competence, 83–5, 195–7
data, 35–6, 132–5, 245–7 competence ascriptions, 58–69
effect, 240 Complex Societies, 23
secondary, 143–9 conditions of work, 127
anonymity, 110n9, 252–3 confidentiality, 110n9
archaeological model, 55–8, 73, 75n10 constructionism, 217
argumentative/discursive interview constructionist discourse, 21
approach, 11, 210–12 constructivist definition, 49–50
assessment, of experts, 129–31 contemporary society, 20–1
asynchronous interaction, 166 content of work, 127
Austrian Technology Delphi, 12, 252, contextual knowledge, 46
256–68 conversational form, 87
authority, interviewer as, 64–5, 68–9 conversational interactions, 142–50
cost-benefit analysis, 240
balloon effect, 237 counter-expert systems, 21, 38n11
bargaining, process of, 187–91, 199 critic, interviewer as potential,
Bauman, Zygmunt, 88 65–7, 68–9
Behrens, Maria, 9, 138 Cross-Impact Analysis, 255, 268–70
benefit motivation, 192–4
Berger, Peter L., 88 data, dependence of, on performance
biographical interviews, 32 levels of researchers, 121–9
biographical structures, 26–7 data analysis, 35–6, 245–7
Bogner, Alexander, 1, 7, 43 quality of experts and, 132–5
Bourdieu, P., 221 secondary, 143–4, 145–9
Burke, L. A., 165 data collection, 31–5, 204
Busse, G., 163–4 decision Delphi, 252, 257, 259–60
decision-making, questions on, 33–4
canonical documents, 90 Delphi method, 157, 238, 252–71
case studies, 54 basic features of, 252–3
catharsis effect, 144, 146, 209, 243 combination of methods, 268–70
certification, 88, 89 development of statements and
Christmann, Gabriela B., 9–10, 157 questionnaires, 262–4
Cicourel, A.V., 204 evaluation results, 253–4
classical Delphi, 259 Expert Delphi, 12, 252, 256, 268–70
closed questions, 165–6 motivation and, 264–6
coding, 36 online, 270–1
co-experts, 58–60, 68–9, 162 Real Time, 270–1
common ground, 76n21 selection of experts, 260–2

275
276 Index

Delphi method – continued analysis of, 35–6, 245–7

Technology Delphi Austria, 256–68 application of, 37n2
technology foresight and, 254–6 archaeological model of, 55–8, 73,
democratic theory, 3–4, 51 75n10
democratic research, 185–91, 198–9 concepts of, 117–19, 139–42
Dexter, L.A., 139, 140 conducting, 243–5
diachronic secondary analysis, 143 data collection, 31–5
disciplinarity, 22 debate over, 1, 43–5, 98–9
discursive consciousness, 29–30, 31, vs. elite interviews, 99–103, 107–9
38n16 epistemological interest of, 81–4
discursive interview structure, 11, ethical considerations, 184–99
210–12 ethnographic embedding of, 87–91
distortion effects, 241, 248n11 exploratory, 7, 46, 101, 160–2
doctors, 87–8, 118 gender-specific interaction effects in,
dual career couples, 27 9, 142, 145–6, 149–52
guided, 203–7, 226
effect analysis, 240 interaction effects in, 7, 45, 56, 57,
Eliade, Mircea, 88 141–2, 144–50, 243
elite interviews, 98–103, 105–9 interaction model of, 55–8, 73, 105–6
elites interaction structures in, 55–72
access problems with, 104–5 in interpretive organizational
definition of, 99, 110n3 research, 217–32
vs. experts, 107–9 introduction to, 17–18
functional, 34, 50, 83 with managers, 203–13
global, 84 methodological ambiguity of, 43–8
local, 83–4 methodological pluralism in, 72–3
sampling, 103–4 methodology in, 5–12
empirical research, 37n2, 203–13 neutrality in, 70–2
enactment, 218 open, 225–6
epistemic cultures, 23, 27–8 in political science, 138–52
ethics, research, 10, 184–99 practical aspects of, 240–7
ethnographic research, 28 pre-knowledge for, 162
evaluation research, 11–12 preparing, 241–3
application-oriented dimension, 237 procedural consequences of, 223–6
development of approaches to, 236–8 in programme evaluation, 235–47
expert interviews in, 235–47 quality differences and, 129–35
methodological dimension, 237 systematizing, 46–7, 101, 160–2
methodological variation in, 236–40 techniques for, 84–7, 105–7
evaluator, interviewer as, 64–5 telephone-based, 9–10, 157–78
everyday communication, 205 theory-generating, 7, 47–8, 50, 53–5,
everyday knowledge, 220 74n9, 101
exam situation, 241, 245 types of, 7, 46–8, 101
executives, private lives of, 205 uses of, 1, 230–1
expectations, 57–8, 206 expertise
expert/counter-expert distinction, 5, 21, external, 222, 223, 224, 230
38n11 reflective subject matter, 222, 224
Expert Delphi method, 12, 252, 256, research, 223
268–70 social science research on, 3–5
expert interviews technical, 3
advantages of, 2–3 types of, 220–3
 Index 277

expert knowledge, 3, 4–5, 82–3, 220–3 face-to-face interviews, 166–7, 169

as analytic construction, 28–9, 52–3, feasibility standards, 248n7
72–3, 102 feedback, 252–3
defined, 7, 18–19 feedback effect, 144, 148–9, 243
explicitness of, 29–30 female researchers, 34–5
negotiating, 27–8 see also gender-specific interaction
relevances of, 23–31, 53–5 effects
societal change and, 21–3 focused interviews, 74n5
socio-cultural conditions of the follow-up questions, 34
production of, 25–7 foresight activities, 255
experts see also technology foresight
access problems with, 104–5 formative power, 107–8
as active participants, 24–5, 38n12 Foucault, M., 28
assessment of, 129–31 Froschauer, Ulrike, 11, 217
bargaining with, 187–91, 199 functional elites, 34, 50, 83
competence of, 83–5, 195–7
constructivist definition of, 49–50 gender relations, 34–5, 140–1
definition of, 7, 18–19, 24–5, 48–52, gender-specific interaction effects, 9,
54–5, 99–100, 107, 220–1 142, 145–6, 149–52
vs. elites, 107–9 Gläser, Jochen, 8–9, 117
“good” and “bad”, 119–21 global elites, 84
identification of, 87–90 Global Risk Society, 23
laicization of, 5 Greenpeace, 38n10
vs. laypersons, 19, 20 guided expert interviews, 203–7, 226
male, 34–5 guidelines, 87
motivations of, 191–8
perceptions of, about interviewer, habitus, 221
57–69, 106 Häder, Michael, 254
post-traditional, 4 Häder, Sabine, 254
quality of, 8–9, 117–19, 121–9, 132–6 harm motivation, 194–8
reformulation of concept of, 48–55 harm reduction, 185
relational concept of, 44 Honer, Anne, 84, 87
relationship management with, 191–8
relevances of, 26 iceberg effect, 76n25, 144, 146–7, 243
researchers and, 185–7 implementation evaluation, 239–40
selection of, 229–30, 260–2 implicit knowledge, 51, 81–2
sensitive, 194–5 indexical language, 85
social embeddedness of, 197–8 informational conversations, 139
social science research, 3–5 informed consent, 186
as stakeholders, 11–12, 236, 240–1, insider knowledge, 38n12
244–5 Institute of Technology Assessment
status of, 18, 23, 139–40 (ITA), 257
trained, 118 institutionalization, 38n10
voluntaristic concept of, 49 institutional-organizational context, 35
expert systems, 21 institutional reflexivity, 4
explicitness, 51 institutions, life-worlds of, 28
exploratory expert interview, 7, 46, 101, intellectuals, 88
160–2 interaction effects, 7, 45, 56, 57, 141–2,
external expertise, 222, 223, 224, 230 150–2
eye contact, 173 catharsis effect, 144, 146, 209, 243
278 Index

interaction effects – continued everyday, 220

in expert interviews, 144–50 implicit, 51, 81–2
feedback effect, 144, 148–9, 243 insider, 38n12
gender-specific, 9, 142, 145–6, 149–52 interpretive, 8, 38n17, 52–4, 72,
iceberg effect, 144, 146–7, 243 99, 107
paternalism, 63, 66, 86, 144, 145–6, 243 local, 25
profile effect, 144, 149–50 non-explicit, 30
interaction model, 55–8, 73, 105–6 operational, 86
interaction structures, 55–72 outsider, 38n12
Intergovernmental Panel on Climate procedural, 8, 99, 107
Change, 28–9 process, 38n17, 52
interpretation method, 226 scientific, 4, 21, 223
interpretive knowledge, 8, 38n17, 52–3, social construction of, 120
54, 72, 99, 107 social relevance of, 52
interpretive organizational research, sociology of, 18–19, 23–4, 30, 45, 49,
217–32 50–1, 53, 151
interpretive paradigm, 45 specialist, 51, 52, 222
interpretive power, 107–8 specialized, 5, 47, 82–3, 88, 223
interviewee subject matter, 221–4
see experts technical, 38n17, 52
interviewer theoretical, 90
as accomplice, 67, 68–9 types of, 19
as authority, 64–5, 68–9 knowledge-power relationship, 28
as co-expert, 58–60, 68–9, 162 knowledge production, 7, 25
as expert from different knowledge new forms of, 20–3, 30–1
culture, 60–2 as open-ended process, 28–9
female, 34–5 quality of researchers and, 121–3
as layperson, 62–4, 68–9 socio-cultural conditions of, 25–7
neutrality of, 56, 70–2, 141–2 transdiciplinary, 29
perceptions about, 57–69, 106 Knowledge Societies, 23, 27–8
personal qualities of, 34 knowledge workers, 108
as potential critic, 65–9
preparation by, 31–2, 241–3 latent content, 6
qualifications of, 90–1 Laudel, Grit, 8–9, 117
as quasi expert, 8, 90 lay persons, 5, 19, 20, 51
role of, 141 interviewer as, 62–4, 68–9
status of, 86–7 lay specialists, 107
interview guidelines, 161–2, 164 Leitner, Andrea, 11–12, 235
interview memos, 143 linguistic irregularities, 172–3
interview protocols, 143 Littig, Beate, 1, 8, 98
interview setting, 84–6 local elites, 83–4
interview techniques, 84–7, 105–6, local knowledge, 25
164, 165 Luckmann, Thomas, 88
interview topics, preparation of, 31–2 Lueger, Manfred, 11, 217
IT industry, 89
male-dominated fields, 34–5
Kern, B., 209 managers, interviews with, 203–13
knowledge maximum structural variation, 225
contextual, 46 meaning, 218–19
see also expert knowledge meaning collectives, 218
 Index 279

Menz, Wolfgang, 1, 7, 43 operational knowledge, 86

methodical-systematic contributions, 163 organizational dynamics, 228
methodology organizational research, expert
analysis, 35–6 interviews in, 217–32
data collection, 31–5 outsider knowledge, 38n12
in evaluation research, 236–40
in expert interviews, 5–12 paraphrase, 35
lack of standardization in, 43–4 participatory observation, 2
for manager interviews, 203–13 paternalism effect, 63, 66, 86,
pluralism in, 72–3 144–6, 243
qualitative, 204–5 Pfadenhauer, Michaela, 7–8, 81
sampling, 103–4 phenomena
standardized, 204 different descriptions of identical,
for telephone-based interviews, 164–6 125–7
method-relational approach, 49–50 identical descriptions of different,
Meuser, Michael, 6–7, 17 123–5
Miller, M. K., 165 plurality, 29
minimization of differences, 225 policy Delphi, 259
Mode 1 knowledge production, 20, 21, 22 policy evaluation, 239
Mode 2 knowledge production, 20, 22, 25 political science interviews, 138–52
modernity political science theory, 3–4
ambivalences of, 20–1 postmodern discourse, 21, 29, 38n12
expertocrization of, 21 post-traditional experts, 4
second, 54 power, 107–8
modernization theory, 4–5, 20 power dynamics, 186–7, 188
motivated research subjects, 191–8 practical consciousness, 38n16
motivation, 191–8 pre-knowledge, 162
benefit, 192–4 preparation, 241–3
Delphi method and, 264–6 private persons, 26–7, 52, 53
harm, 194–8 problem-centered interviews, 74n5
problem framing, 5
Nagel, Ulrike, 6–7, 17 procedural knowledge, 8, 99, 107
narrative-based interview structure, 11 process knowledge, 38n17, 52
narrative interviews, 203 professional organizations, 89
narratives, 32–3 professionals, 87–8
negotiation, 27–8, 227–9 see also experts
networking, 27–8 professional terminology, 85, 245
neutrality, 70–2, 141–2 professions, 118
NGO representatives, 7 profile effect, 144, 149–50
non-elites, 110n10 programme evaluation, expert
non-explicit knowledge, 30 interviews in, 235–47
non-verbal signals, 167–8, 173–74 propriety standards, 248n7

Obelene, Vaida, 10, 184 qualitative analysis, 138

objective hermeneutics, 203 qualitative interviews
Oblené, Vaida, 10 methodical aspects, 203–13
observation, levels of, 11 openness of, 204–5
Opdenakker, R., 166–7 telephone-based, 162–8
open-ended questions, 165–6 qualitative social research, 6, 86–7
open interviews, 31, 225–6 qualitative theory, 48
280 Index

quality, of experts, 8–9, 117–19, 121–9, scientific knowledge, 4, 21, 223

132–6 secondary analysis, 143–9
quality assurance strategies, 226 second modernity, 54
quasi-experts, 86, 90 second opinions, 5
questionnaires, 31, 160, 263, 265 sensemaking, 218–19
questions, 33–4 sensitive experts, 194–5
closed, 165–6 shamans, 88–9
open-ended, 165–6 shared agenda, 192–3
standardized, 204 SMIC methods, 268–70
social constructivism, 3, 45, 102, 120,
Rammert, W., 23 217–18
rationality, 4 social cues, 166–7
realist approach, to expertise, 3 social embeddedness, 197–8
reality, construction of social, 217–18 social movements, 38n10, 51
Real-Time Delphi (RT Delphi), 270–1 social relevance, 52, 53–5
reflective subject matter expertise, social-representational approach, 49–50
222, 224 social science research, trends in, 3–5
reflexivity, 51 societal change, 19–23
relational space, 140 society, functional elites of, 34
relationship management, 191–8 socio-cultural conditions, 25–7
research sociological conceptualization, 36
democratic, 185–91, 198–9 sociology of knowledge, 18–19, 23–4,
empirical, 37n2, 203–13 30, 45, 49–51, 53, 151
ethnographic, 28 specialist knowledge, 222
evaluation, 11–12, 235–47 specialists, 82–3, 107
qualitative, 6, 86–7 specialization, 5, 223
social science, 3–5 specialized knowledge, 5, 47, 82–3, 88
researchers special knowledge, 51, 52
assessment of, 129–31 Sprondel, W. M., 19, 51
dependence of data on performance stakeholder problem, 240–1, 244–5
levels of, 121–9 stakeholders, 11–12, 236
female, 34–6 standardized interviews, 168, 204
“good” and “bad”, 119–29, 134–5 standardized surveys, 47
as information sources of science Standards of the Joint Committee, 237–8,
studies, 119–21 248n7
interpretive organizational, 217–32 status relation, 34
research ethics, 10, 184–99 structuring, 219
research expertise, 223 studying down, 186
responsive evaluation, 238 studying up, 186
rhetorical interviews, 71 subjective interpretation, 204, 246
risk controversies, 5 subject matter know-how, 221–4
role expectations, 57–72, 76n23, 106, surveys, 47
206, 207 synchronic secondary analysis, 143
routine activities, breaches of, 33 synchronous interaction, 166
RPM Screening, 269–70 systematic quantitative surveys, 2
systematizing expert interview, 7, 46–7,
sampling, 103–4, 225, 229–30 101, 160–2
Schütz, Alfred, 19, 20, 51
Science Citation Index, 120–1 technical expertise, 3
science studies, 119–21, 122 technical knowledge, 38n17, 52
scientific criticism, 4 technocratic debate, 4
 Index 281

technological determinism, 256 theoretical generalization, 36

Technology Delphi Austria, 256–68 theoretical knowledge, 90
development of Delphi statements theoretical sampling, 225, 229–30
and questionnaires, 262–4 theory-generating expert interviews, 7,
feedback and analysis, 266–7 47–8, 50, 53–5, 74n9, 101
motivation to participate in, 264–6 theory of errors, 253
objectives and project design, 256–9 theory of society, 4
organization as decision Delphi, topic guide, 31–2, 47, 87, 242
259–60 Total Design Method (TDM), 265–6
results and implementation, 267–8 trade secrets, 86
selection of experts, 260–2 trained experts, 118
technology foresight, 12, 253, 254–6 transcription, 35
Technology Roadmapping, 255 transdiciplinarity, 22
telephone-based interviews, 9–10, 157–78 Trinczek, Rainer, 11, 203
compared with face-to-face Turner, Stephen, 4
interviews, 166–7, 169
disadvantages of, 164 universities, 22
duration of, 166, 169 university planning, 158–9
interview guidelines for, 161–2, 164 utility standards, 248n7
problems with, 169–77
qualitative-oriented, 162–8 validation criteria, 9
standardized, 168 voluntaristic concept of expert, 49
techniques for, 164, 165–6
terminology, 85, 245 Weber, Max, 50
thematic comparison, 36 workplace studies, 94n23
thematic competence, 86 Wroblewski, Angela, 11–12, 235
thematic focusing, 85
thematic units, 35 Zuckerman, H., 120