2

DOCUMENTS AND
DOCUMENTARY
RESEARCH
CONTENTS

Introduction8

What is a Document? 8

Varieties of Document 10

The Possibilities and Advantages of Documentary Research 13

Documents in, or as the Focus of, Research 15

The Purpose and Usage of Documents 18

Conclusion19

03_TIGHT_DRSS_CH_02.indd 7 18/03/2019 4:22:53 PM

 8 STARTING DOCUMENTARY RESEARCH

Introduction
This chapter considers what documents are, how they vary in nature, and the possibili-
ties and advantages associated with undertaking documentary research. It then discusses
the use of documentary research as the main approach or as part (i.e. alongside other
approaches or techniques) of a research project, and the purpose and usage of documents.
Understanding something of the nature and variety of documents is key to being able to
appreciate and undertake documentary research, as well as – given the ubiquitous nature
of the documentary approach – research in general. Hopefully, such an understanding
may encourage you to make more, and more varied, usage of documents in your research.

What is a Document?
The answer to this question may appear, on first thought, to be obvious, but a little more
consideration reveals that it is not so. Indeed, a good deal of effort has been devoted
to trying to provide a convincing answer to this question, but without yet achieving
universal accord. Thus, as matters stand, the definition of a document is not fixed and/
or generally agreed; alternative sources offer their own definitions, and these both vary
and change or develop over time (Buckland 1997). Or, in other words, what constitutes
or characterises a document is contested.
This should not be a particular cause for concern, however, as this is a common state of
affairs throughout the social sciences and beyond, where progress in our understanding
occurs through debate and disagreement. Indeed, it might be said that anything worthy
of serious consideration in the social sciences is a matter of contestation and debate.
This may be readily illustrated by the following four definitions of a document or
documents, which have been ordered by ascending date to show some of the develop-
ment that has taken place in our understanding over the last three decades. More and
different definitions could, of course, have been employed, but these four usefully illus-
trate something of the variation and change apparent:

[A] document in its most general sense is a written text. (Scott 1990, p. 12)

Documents are texts that can be published or unpublished, written, oral and virtual and may
reside in either the public, private or virtual domains. (Fitzgerald 2007, p. 281)

A document is simply any written, printed, photographed, painted or recorded material that can
be used to provide information or evidence. (Dolowitz, Buckler and Sweeney 2008, p. 39)

A document may be defined briefly as a record of an event or process. (McCulloch 2011, p. 249)

Thus, Scott offers a relatively early and common-sense definition: documents are writ-
ten texts – books, papers, scripts, diaries, committee minutes, etc. – things which have
or can have (when they are printed) a physical form. Both Fitzgerald and Dolowitz et al.

03_TIGHT_DRSS_CH_02.indd 8 18/03/2019 4:22:53 PM

 DOCUMENTS AND DOCUMENTARY RESEARCH 9

add to or extend this definition, with Fitzgerald adding oral and virtual documents, and
Dolowitz et al. extending the definition further to include photographs, paintings and
recorded material. Writing at the beginning of the current decade, McCulloch reverts
to a simpler but inclusive definition, portraying documents as records of something,
without specifying, or limiting, the nature of the record.
We may identify both background and foreground modifications lying behind these
changes in definition. In the background, affecting everything not just in academic life
but globally, there have been many, and some quite profound, changes in the nearly
30 years since Scott was writing. Probably the most obvious of these is the impact of the
internet, with the growing number of online sources available (including sources that
were originally only available as written texts) forcing a broadening of the definition of
a document away from the physical, written text.
Meanwhile, in the foreground, practices and fashions in research methodology have
also evolved, with a growing interest in the visual and aural seeking to complement
the focus on the written. Nevertheless, most of the interest in documentary research
remains on the written text, whether online or in physical form. Most social researchers
remain fairly conventional in their outlook and in the research techniques they employ.
Other authors, in seeking to define what a document is, have sought to distinguish it
from other terms. After all, a common strategy in defining something is to state what it
is not as well as what it is. Thus, Lincoln (1980) makes an interesting – but now perhaps
rather outdated – distinction between documents and records:

we shall define a record as a written statement prepared by an individual or an agency for the
purpose of attesting to an event or providing an accounting, and a document as any written
(or filmed) material other than a record and which was not prepared specifically in response to
some request from or some task set by the investigator. (p. 4, emphasis in original)

By contrast, the authors previously quoted would all include such records as documents;
indeed, McCulloch (2011) clearly sees the terms as pseudonyms.
Lincoln’s (1980) definition is also of interest, however, for what it says – in the sec-
ond part of the quotation – about what documents are not. In excluding material that
was specifically prepared in response to a request or task set by the investigator (i.e. the
researcher), she is drawing a line between documents and, for example, interview tran-
scripts and survey responses. That this is a blurred line is, though, apparent, because it
would allow – as most researchers would probably now accept – interview and survey
data collected by someone else to be considered as documents by another researcher.
Does this mean, though, that almost anything can be considered as a document? If
documents are not just written or printed texts, but may be oral or virtual, or filmed
or painted, what then cannot (at least potentially) be considered as a document? The
addition of the qualifier ‘at least potentially’ suggests that it depends very much on the
researcher and their perspective. If they are treating the material – written, oral, virtual,
visual – as a source for their research, and they have not recently generated the material
themselves, then it may be considered as a document.

03_TIGHT_DRSS_CH_02.indd 9 18/03/2019 4:22:53 PM

 10 STARTING DOCUMENTARY RESEARCH

Rather than prescribe my own definition, and possibly add further to the potential
confusion, or adopt one of the available definitions, I will proceed bearing all of these
definitions and points in mind. In doing so, I will aim simultaneously to both keep the
range of interest broad and acknowledge where the primary focus lies.

Varieties of Document
As the discussion in the previous section will have indicated, documents as a class may
be very varied; but just how varied? We may, again, usefully compare what different
documentary researchers have written at different times. Here are two sample categori-
sations or indicative lists:

Documentary sources of information, of all kinds, figure centrally in the research of

sociologists. Official statistics on crime, income distribution, health and illness, censuses of
population, newspaper reports, diaries, reference books, government publications, and similar
sources are the basis of much social research by academics and their students. Yet these
materials have rarely been given the attention that they deserve in accounts of sociological
research methods. (Scott 1990, p. ix)

Documents that may be used for systematic evaluation as part of a study take a variety of
forms. They include advertisements; agendas, attendance registers, and minutes of meetings;
manuals; background papers; books and brochures; diaries and journals; event programs (i.e.,
printed outlines); letters and memoranda; maps and charts; newspapers (clippings/articles);
press releases; program proposals, application forms, and summaries; radio and television
program scripts; organisational or institutional reports; survey data; and various public records.
Scrapbooks and photo albums can also furnish documentary material for research purposes.
These types of documents are found in libraries, newspaper archives, historical society offices,
and organisational or institutional files. (Bowen 2009, pp. 27–28)

The earlier of the two authors, Scott – in line with his definition quoted in the previous
section – offers a broad, but by no means comprehensive, categorisation of different
kinds of written texts that may be used as the basis for documentary research. Tellingly,
he also argues (as I did in Chapter 1) that these sources have been relatively neglected;
and, though he makes his case from the perspective of sociology, it might be extended
to apply to all social research.
Bowen (2009), more recently, goes somewhat beyond Scott’s (1990) purview – though
much of his description relates to written texts – to add items such as scrapbooks and
photo albums. He also usefully indicates where these kinds of items are likely to be
found. We might, in the light of some of the later definitions quoted in the previous
section, go even further. Documents may be texts or data sets, printed or hand-written,
quantitative and/or qualitative, physical or online, personal or official, closed or open,
visual or representational.

03_TIGHT_DRSS_CH_02.indd 10 18/03/2019 4:22:53 PM

 DOCUMENTS AND DOCUMENTARY RESEARCH 11

This is not to say, to pick up on the argument in the previous section, that every-
thing is a document, or may be treated as a document, but rather – to turn it around
slightly – that everything we can comprehend may be documented in some way. The
key distinction between documentary and other forms of research is that documents
already exist – even, perhaps, if we have previously created them ourselves – before we
research them.
In experimental, interview-based, observational, survey and other forms of research,
the first part of the research process involves the collection, creation or co-creation of
new empirical data, which is then analysed. In documentary research, the data already
exists. While it will still need to be accessed and/or collected, this should not take any-
thing like the time involved in new empirical research. Documentary research does not
set out to create new data, but to undertake analyses of existing data, whether that data
has previously been analysed or not.
We can better illustrate the variety included within documentary research if we
consider some examples. Box 2.1 summarises ten recent examples of published docu-
mentary research across a range of disciplines and from a variety of countries.
The selection of examples in Box 2.1 could not be said to be representative, but
is, rather, meant to be illustrative. Many alternative examples could have been used
instead. The selection of these examples (and this is the case throughout the book)
should not be taken to imply that they are necessarily the highest quality research –
though they may well be, having been published in refereed international journals – but
rather that they are useful and interesting.

BOX 2.1

Some Examples of Documentary Research

1. Awang, Jindal-Snape and Barber (2013) analysed the Malaysian government’s strategies for
promoting ‘positive behaviour’ among the young by carrying out a content analysis of 91 official
circulars issued to schools between 1969 and 2011. The circulars were read and analysed by the
researchers to identify all mentions and examples of positive and negative behaviour, and of
the strategies proposed for encouraging positive behaviour.
2. Boon et al. (2009), concerned about the role of pharmacists in North America, carried out a
systematic search in selected health databases for articles relating to pharmacy, natural health
products and dietary supplements. An in-depth, qualitative content analysis of 259 articles
was carried out. Articles were coded in terms of the roles, responsibilities and behaviours of
pharmacists.
3. Clark, Parker and Davey (2014), interested in the role of nurse practitioners in caring for
the aged in Australia, carried out a documentary analysis of 32 project proposals awarded
funding by the government department concerned. A content analysis was carried out

(Continued)

03_TIGHT_DRSS_CH_02.indd 11 18/03/2019 4:22:54 PM

 12 STARTING DOCUMENTARY RESEARCH

(Continued)

to identify the key features and characteristics of the models of aged care that were
successfully proposed.
4. Davis (2012) considers the use of leadership and change theory in changing practice in early
years settings by undertaking an analysis of the scripts produced by 10 professionals writing
about this for an MA course in the UK. In her analysis, she noted the cultural and pedagogic
changes proposed and the theories used to justify them.
5. EzzelArab (2009) reports on a documentary and contextual analysis of ‘al-La iha al-Wataniyya’
(‘The National Programme’), ‘a proposal adopted in April 1879 by Egypt’s economic and political
elites as an expression of resistance to the fiscal program of European control’ (p. 301). He
examines both the Arabic and French versions of the document in terms of its signatories and
contents.
6. Grazioli and Jarvenpaa (2003) focus on consumer and business deception on the internet,
examining documentary evidence published between 1995 and 2000 relating to 201 cases
of internet deception. They searched newspapers, journals and legal documents included in
certain databases, as well as the internet sites of monitoring agencies, to identify the tactics
adopted and what might be done to respond to them.
7. Himmelsbach et al. (2015) examine documentary data to reconstruct flood events on the
Upper Rhine River and its tributaries in Europe since AD 1480. They employ critical source
analysis to derive data from over 4,000 references – ‘flood marks, drawings, flood maps,
newspapers, gauge data and contemporaneous administrative reports and chronicles’ (p. 4151) –
relating to around 2,800 flood events.
8. Lampe et al. (2017) report on a systematic literature review carried out to assess the reliability
of observational data on aggression. Two search engines (PsycINFO and PubMED) were used to
identify relevant articles published between 1994 and 2014. Identifying 37 articles for analysis,
they found that the inter-rater reliability of observing aggression was fair to excellent; though
they noted that most observations took place in laboratory settings, with only a limited number
in naturalistic settings.
9. Viswambharan and Priya (2016) carried out a documentary analysis of a documentary (i.e. a
film) on the post-Godhra communal riots in India in 2002 to explore disaster mental health.
They use constructionist grounded theory to examine the documentary as a whole and the
different categories of survivors’ suffering identified.
10. Williams (2015) analyses documentary data from company reports, news articles, industry
analyst reports and industry magazines to explore how the Coca Cola Company successfully
re-entered the Indian and Chinese markets following the liberalisation of their economies.
Relevant documents over the period 1975–2012 were identified using keyword searches, and then
triangulated across sources to create a narrative of events.

The range of disciplines covered includes business, climatology, education, health

care, history, pharmacy and politics; in other words, it extends throughout and well
beyond the social sciences that are the main focus of this book (though, in all cases,
the researchers are, in essence, using elements of social science research methodology:

03_TIGHT_DRSS_CH_02.indd 12 18/03/2019 4:22:54 PM

 DOCUMENTS AND DOCUMENTARY RESEARCH 13

Chapter 4 contains a fuller discussion of documentary research in the disciplines). The

countries discussed include not only (in articles published in the English language) the
obvious English-speaking ones – the USA, the UK and Australia – but also developing
countries such as China, Egypt, India and Malaysia.
The examples given also vary significantly in other ways. Thus, in terms of scale,
while Himmelsbach et al. (2015) based their analysis on over 4,000 references, both
Viswambharan and Priya (2016) and EzzelArab (2009) analyse a single document, in the
former case a film documentary and in the latter case two different versions of a text.
While most of the analyses were based on written texts – such as government circulars,
articles, project proposals, historical chronicles and student assignments – others made
use of a range of other kinds of ‘documents’. These included drawings, film, flood marks
and maps, and a variety of internet sources. Clearly, then, documents may not only be
very varied, but also very useful for a wide range of research interests and projects.

The Possibilities and Advantages of Documentary Research

The discussion so far in this chapter should already have suggested some of the
possibilities and advantages of documentary research. In a relatively early account,
Lincoln – maintaining a firm distinction between documents and records which,
as we have already suggested, many would disagree with – identifies six major
advantages:

First, documents and records are a stable, rich and rewarding resource … Second, records
(as opposed to documents, although occasionally the same will hold true for those writings
also) constitute a legally unassailable base from which to defend oneself against allegations,
interpretations and libel … Third, both documents and records represent a ‘natural’ source of
information – a delight to the naturalistic inquirer. Fourth, they are available on a low-cost or
no cost basis, requiring often only the investigator’s time and energy … Fifth, documents and
records both are non-reactive … Sixth, whether or not the inquirer finally decides to interact with
his [sic] subjects, content analysis and other forms of documentary and record analysis enable
supplementary and contextual data to be gathered. (1980, pp. 10–12, emphasis in original)

However, Lincoln either overlooks or underplays two other advantages which – at least
nowadays – might be seen as being of even greater significance: the accessibility and
scale of documents. As documentary research does not create new data but analyses
existing data, it offers considerable opportunities for social researchers in terms of both
the accessibility and the scale of the research they can undertake within given time and
resource constraints.
After all, most social researchers are operating under considerable constraints,
particularly those who are expected to complete a project within a given, relatively
short timescale. This is the case for many students completing dissertation projects

03_TIGHT_DRSS_CH_02.indd 13 18/03/2019 4:22:54 PM

 14 STARTING DOCUMENTARY RESEARCH

at undergraduate or masters level, and even for research students undertaking PhDs,
as well as for other researchers working inside or outside academe. The time and
financial resources available will be limited, so analysing existing data where possi-
ble, rather than first collecting new data and then analysing that, offers considerable
potential. Not only is the time taken in collecting new data saved, but existing data
sets may be much larger, more representative and more detailed than an individual
researcher, or a small team, could hope to collect in the time they have available.
Naturally enough, given the time at which she was writing, Lincoln also did not
mention the online availability of documents. A huge and increasing amount of data –
both quantitative and qualitative – is available for social research, much of it online
and freely accessible, at least to bona fide researchers. It would be foolish not to at least
explore the possibilities, whether as part of a research project or as the whole project,
before committing to further data collection. Collecting new data poses much greater
access and ethical challenges (ethical issues are discussed further in Chapter 3), and
many of the potential targets for social research – i.e. the general public and various
sub-groups of it – are now fatigued by and resistant to requests to complete yet another
survey or be interviewed again.
One further advantage of documentary research is that there are kinds or topics of
research which it is difficult or impossible to do in any other way. This is self-evidently
the case with historical research that reaches further back than present-day lifetimes, as
documents of various kinds will be the only sources of data available. More contempo-
raneously, it is similarly the case when the subject of the research concerns the internet
and the material found on it, much of which is simply not available elsewhere or in
other formats.
Documentary research may also be the only option when other forms of access are
not possible. This was the case, for example, with the research carried out by Grazioli
and Jarvenpaa (2003) on internet deception, outlined in Box 2.1. Since people engaged
in deception would be difficult to contact and highly unlikely to be willing to cooper-
ate with the research, they decided to search newspapers, legal documents and other
sources for the data they required:

Gathering real-world data on deviant behavior is usually difficult because perpetrators actively
attempt to hide evidence that the behavior occurred … Under these conditions, one viable
methodology consists of identifying cases of Internet deception available in public records and
performing content analysis on them. (p. 103)

Unless, therefore, you have a strong methodological or personal preference for engaging
in other forms of research, you would be well advised to consider documentary research
as an alternative or complementary approach. You can do documentary research with-
out ever leaving the house or office, and you don’t need to talk to anyone. You will likely
complete your research much quicker this way than you would otherwise.
Indeed, I even considered using the expression ‘social research for the anti-social’ as
the subtitle for this book – recognising that many who are involved in social research do

03_TIGHT_DRSS_CH_02.indd 14 18/03/2019 4:22:54 PM

 DOCUMENTS AND DOCUMENTARY RESEARCH 15

not (like me) relish the close engagement with others necessitated in some forms of data
collection – but did not want this to be taken in the wrong way.

Documents in, or as the Focus of, Research

If you do decide to carry out a piece of documentary research – or perhaps have this
decided for you – then this book should, hopefully, be of considerable use. Even if you
decide to take a different approach to your research, however, involving the collection of
new empirical data, your research will almost certainly (as was pointed out in Chapter 1)
involve an element of documentary research – for example, reading and analysing some
of the relevant academic and/or policy literature. So this book should still be of some use.
Reviews of the academic and policy literatures are both classic genres of documentary
research. Thus, literature reviews are considered in detail in Chapter 6; while their ‘big
brother’, systematic reviews, form the subject of Chapter 7, along with meta-analyses;
and policy research is the focus of Chapter 10. Each requires careful planning and reflec-
tion, and the approach taken should be made clear and justified when they are written
up. Documentary research, even when it only forms part of a research project, demands
just as much rigour and care as other forms of research.
Box 2.2 summarises ten recent examples of research that has made use of documen-
tary analysis alongside other research methods. As with Box 2.1, it gives some idea of the
diversity of mixed methodological approaches (see also Chapter 14) that are possible.

BOX 2.2

Some Examples of Documentary Research Being Used in Conjunction

with Other Methods

1. Avby, Nilsen and Ellström (2017) explored knowledge use and learning among social workers
engaged in everyday child investigation work in Sweden. They employed interviews, participant
observation and reflective dialogues, as well as the documentary analysis of case files, to
collect data. They characterise their overall research design as ethnography, and based their
research in two children’s services departments employing 40 social workers between them.
The study focused on five cases and seven of the social workers.
2. Brunton and Mackintosh (2017) consider the purpose of university sport and its relation to
national sport policy in the UK. Adopting an interpretivist approach, they carried out eight
in-depth, semi-structured interviews with those with overall responsibility for sport at eight
universities in the north of England. ‘In addition, documentary analysis was carried out to
consider the evidence of how universities link to government and national sports policy. The
strategic plan of each university was analysed alongside key strategies for sport and higher

(Continued)

03_TIGHT_DRSS_CH_02.indd 15 18/03/2019 4:22:54 PM

 16 STARTING DOCUMENTARY RESEARCH

(Continued)

education sport in England’ (p. 382). Qualitative content analysis and thematic analysis were
used.
3. Canonico et al. (2017) studied knowledge integration mechanisms in an interdisciplinary
research project. They describe their research as a case study, focusing on a single university/
industry research project in Italy. ‘The authors used three data collection techniques: internal
document analysis, observation/site visits and semi-structured interviews. Documentary
analysis was used to understand the organizational structure and to identify knowledge
integration issues. Observation and site visits at university research laboratories were used
to increase understanding on particular issues. Staff interviewed included managers and
academic researchers’ (p. 604).
4. Del Río, Peñascoa and Mir-Artigues (2018) were interested in the drivers for, and barriers to,
the deployment of concentrated solar power as an energy source in the European Union by
2030. They researched this by means of a ‘thorough literature review and interviews with key
stakeholders in the sector’ (p. 1019). The literature review focused on material published in the
most recent five-year period, 2011–15: 91 references are provided at the end of the article.
5. Foley et al. (2017) researched the impact of the reconfiguration of the emergency care system
in Ireland. They adopted a comparative case study approach, examining the experience in three
regions of the country. ‘Documentary analysis of reconfiguration planning reports was used to
identify planned public engagement activities. Semi-structured interviews with 74 purposively-
sampled stakeholders explored their perspectives on reconfiguration, engagement activities
and public responses to reconfiguration’ (p. 800).
6. Giovannini (2016) discusses the new regional political voices developing in England, focusing on
the example of Yorkshire First. She analyses the party’s manifesto and other policy documents,
and supplements this with the results of an online questionnaire survey sent to all of the
party’s members. This enables her to identify ‘the emergence of a nascent grass-roots form of
regionalism in Yorkshire that breaks away from past experiences and seeks to challenge the
dominant narrative of regionalisation’ (p. 598).
7. McGrattan and Williams (2017) examine the impact of devolution in the UK on national
identification in Wales and Northern Ireland. They undertook an ‘original qualitative research
including data from documentary analysis, focus groups and structured interviews’ (p. 465). In
the Welsh case, the documentary evidence consists of ‘selected quantitative data’ in the form of
responses to surveys on the desire for independence, while, in the Northern Irish case, it comes
‘mainly from political party documents and the Northern Ireland Life and Times Surveys’ (p. 471).
8. Rickinson et al. (2017) look at the relationship between evidence-informed practice and
evidence-informed policy in education in Australia. Data collection involved: ‘(i) in-depth
semi-structured interviews with 25 policy-makers who were involved in the development
of the selected policies; (ii) documentary analysis of policy documents, background research
reports and other relevant papers relating to the selected policies; (iii) unstructured
observation (where possible) of meetings and events connected with the development of the
selected policies; and (iv) feedback from 40 wider policy staff who took part in a verification
workshop to discuss the project’s emerging findings’ (p. 173).

03_TIGHT_DRSS_CH_02.indd 16 18/03/2019 4:22:54 PM

 DOCUMENTS AND DOCUMENTARY RESEARCH 17

9. Teh, Hotte and Sumaila (2017) were concerned with whether fishery buybacks could achieve
positive socio-economic results. They carried out four case studies of selected Australian,
American, Canadian and Norwegian fisheries. They ‘first conducted a desktop review of
existing literature on the selected case studies. This information was then supplemented with
in-depth telephone interviews conducted between December 2007 and February 2008 with
nine experts who were either involved in the management, or had extensive knowledge and
research experience about the respective fishery buybacks’ (pp. 3–4).
10. Urbano and Keeton (2017) researched carbon dynamics in recovering secondary forests in
northeastern USA. They ‘employed a longitudinal study based on twelve years of empirical
data (2001–2013) collected from 60 permanent monitoring plots within 16 reference stands
at Marsh-Billings-Rockefeller National Historical Park in Woodstock, VT … [and] 150 years of
documentary data from park management records’ (p. 21). The latter data was both qualitative
and quantitative, and covered the 1880–2013 period.

Thus, the disciplines and/or topics represented include education, energy, fisher-
ies, forestry, health care, politics, social work and sports. As with the examples given
in Box 2.1, these go well beyond the social sciences. The documents being analysed
include case files, internal institutional documents, manifestos, planning reports, policy
documents, research reports, strategic plans and surveys. And the other methods being
used range from interviews and focus groups to questionnaire surveys and participant
and non-participant observation, and to longitudinal studies, reflective dialogues, site
visits and workshops.
The question arises, of course, as to why you might or should use documents as well
as other forms of data in research; does this not mean making more work for yourself?
The answer is, in part, as has already been suggested, that this is what is expected:
some discussion of the existing research and/or policy literature is a standard element of
research projects in the social sciences and beyond. Otherwise, how would you be able
to put your research in context, demonstrate what its contribution to the field of study
is, and, possibly, avoid unnecessarily repeating research that had already been done?
The other, and perhaps more important, part of the answer to this question is that
using more than one form of data and/or form of analysis adds value to a research pro-
ject. It allows for what social science researchers call triangulation to take place (this
is discussed in more detail in Chapter 14). For example, if you research a particular
topic using documents and interviews, and the different sources of data suggest much
the same conclusions, supporting each other, this makes your analysis and argument
rather stronger and more convincing. Or, alternatively, if your sources and analyses
suggest conflicting interpretations, this is also useful (if perhaps a little frustrating) as it
alerts you to possible errors – in the data or in your analysis of it – or indicates that the
issue you are researching is rather more complex or multifaceted than you previously
thought.

03_TIGHT_DRSS_CH_02.indd 17 18/03/2019 4:22:54 PM

 18 STARTING DOCUMENTARY RESEARCH

The Purpose and Usage of Documents

[W]e cannot treat records – however ‘official’ – as firm evidence of what they report … We
have to approach them for what they are and what they are used to accomplish. (Atkinson and
Coffey 1997, p. 47)

Undertaking research into documents – or records, as Atkinson and Coffey term

them – requires an appreciation and understanding of their purpose(s) and usage(s).
No document should be treated as an objective tabula rasa or ‘tablet of stone’; rather,
they have all been created by highly subjective and interested individuals such as
you and me. Documents need to be treated with care: ‘if we are to get to grips with
the nature of documents then we have to move away from a consideration of them
as stable, static and pre-defined artefacts’ (Prior 2003, p. 2).
Document analysis, then, is:

a systematic procedure for reviewing or evaluating documents – both printed and electronic
(computer-based and Internet-transmitted) material. Like other analytical methods in
qualitative research, document analysis requires that data be examined and interpreted in
order to elicit meaning, gain understanding, and develop empirical knowledge. Documents
contain text (words) and images that have been recorded without a researcher’s intervention.
(Bowen 2009, p. 27)

While I would disagree with Bowen’s emphasis on the use of qualitative research methods –
documents may just as well be analysed quantitatively, or using a mixture of qualitative
and quantitative methods (see Chapters 12, 13 and 14) – he is right to emphasise that
documents, unlike other forms of research data, have normally been created without
the researcher’s involvement.
This means that we need to take particular care in their interpretation, just as
we would, for example, in reading a party election manifesto that dropped through
our letterbox, or an investment opportunity that arrived in our inbox. This would
involve:

• considering who created the document, for what purposes and in what context
• reviewing how the document has been used and interpreted since its creation (what impact has it
had?)
• examining the document in relation to similar and related documents (what is its relationship to
other documents?)
• assessing what the document does and does not ‘say’ (‘reading between the lines’), and how it
says it.

These points are, of course, analogous to the processes that competent social researchers
would go through in carrying out other forms of social research. They will be discussed
further in the next chapter, Chapter 3, and returned to throughout the book.

03_TIGHT_DRSS_CH_02.indd 18 18/03/2019 4:22:54 PM

 DOCUMENTS AND DOCUMENTARY RESEARCH 19

Conclusion
This chapter has provided a brief introduction to documents and documentary research.
It has argued that documentary research differs from other forms of research in that the
objects of research (the data) exist already (in the form of documents). However, the
processes involved in carrying out documentary research are analogous to, and should
be just as rigorous as, those employed in other forms of social research.
Documents exist and are available in a huge variety, and social (and other)
researchers should make much more use of them – on their own or in conjunction
with other forms of research – than they currently do. As empirical research becomes
more problematic, it behoves us as social researchers to fully exploit the data that
already exists, and is often freely available.

KEY READINGS
Bowen, G (2009) Document Analysis as a Qualitative Research Method. Qualitative
Research Journal, 9, 2, pp. 27–40.
A very accessible account, albeit focusing on qualitative analysis only.

McCulloch, G (2004) Documentary Research in Education, History and the Social Sciences.
London, Routledge.
One of the few recent books on the topic, and still in print.

Prior, L (2003) Using Documents in Social Research. London, Sage.

Also still in print, and very readable.

Scott, J (1990) A Matter of Record: Documentary source in social research. Cambridge,

Polity Press.
A classic in the field. While inevitably somewhat out of date, particularly with the
massive expansion of online sources, this text still has much to offer.

03_TIGHT_DRSS_CH_02.indd 19 18/03/2019 4:22:54 PM

03_TIGHT_DRSS_CH_02.indd 20 18/03/2019 4:22:54 PM