Edited by J Myron Atkin and Janet E.

Coffey

Arlington, Virginia
Claire Reinburg, Director
J. Andrew Cocke, Associate Editor
Judy Cusick, Associate Editor
Betty Smith, Associate Editor

ART AND DESIGN Linda Olliver, Director

NSTA WEB Tim Weber, Webmaster
PERIODICALS PUBLISHING Shelley Carey, Director
PRINTING AND PRODUCTION Catherine Lorrain, Director
Nguyet Tran, Assistant Production Manager
Jack Parker, Electronic Prepress Technician
PUBLICATIONS OPERATIONs Erin Miller, Manager
sciLINKS Tyson Brown, Manager
David Anderson, Web and Development Coordinator

NATIONAL SCIENCE TEACHERS ASSOCIATION

Gerald F. Wheeler, Executive Director
David Beacom, Publisher

Copyright © 2003 by the National Science Teachers Association. Chapter 2, “Learning through
Assessment: Assessment for Learning in the Science Classroom,” copyright © 2003 by the National
Science Teachers Association and Anne Davies.
All rights reserved. Printed in the United States of America by Victor Graphics, Inc.

Science Educators’ Essay Collection

Everyday Assessment in the Science Classroom
NSTA Stock Number: PB172X
08 07 06 5 4 3
Library of Congress Cataloging-in-Publication Data
Everyday assessment in the science classroom / J. Myron Atkin and Janet E. Coffey, editors.
p. cm.— (Science educators’ essay collection)
Includes bibliographical references and index.
ISBN 0-87355-217-2
1. Science—Study and teaching (Elementary) —Evaluation. 2. Science—Study and teaching (Second-
ary)—Evaluation. 3. Science—Ability testing. I. Atkin, J. Myron. II. Coffey, Janet E. III. National Science
Teachers Association. IV. Series.
LB1585.E97 2003
507'.1—dc21 2003000907

NSTA is committed to publishing quality materials that promote the best in inquiry-based science education. However,
conditions of actual use may vary and the safety procedures and practices described in this book are intended to serve
only as a guide. Additional precautionary measures may be required. NSTA and the author(s) do not warrant or
represent that the procedures and practices in this book meet any safety code or standard or federal, state, or local
regulations. NSTA and the author(s) disclaim any liability for personal injury or damage to property arising out of or
relating to the use of this book including any of the recommendations, instructions, or materials contained therein.

Permission is granted in advance for photocopying brief excerpts for one-time use in a classroom or
workshop. Requests involving electronic reproduction should be directed to Permissions/NSTA Press,
1840 Wilson Blvd., Arlington, VA 22201-3000; fax 703-526-9754. Permissions requests for coursepacks,
textbooks, and other commercial uses should be directed to Copyright Clearance Center, 222
Rosewood Dr., Danvers, MA 01923; fax 978-646-8600; www.copyright.com.
 Contents
Acknowledgments ............................................................................................ ix
About the Editors .............................................................................................. x
Introduction
J Myron Atkin and Janet E. Coffey .................................................................... xi

1 The Importance of Everyday Assessment

Paul Black ............................................................................................................ 1

Assessment for learning is set in the context of conflicts and synergies with the
other purposes of assessments. The core ideas are that it is characterized by the
day-to-day use of evidence to guide students’ learning and that everyday practice
must be grounded in theories of how people learn. Its development can change the
classroom roles of both teachers and students. The ways in which practice varies
when broad aims of science education change are illustrated in relation to
practices in other school subjects.

2 Learning through Assessment: Assessment for Learning

in the Science Classroom
Anne Davies ...................................................................................................... 13

This chapter presents an extended example from a middle school science

classroom of what assessment that supports learning looks like. In the example, the
teacher models assessment for learning by talking about learning with her
students; showing samples of quality work; setting and using criteria; helping
students self-assess and set goals; providing specific, descriptive feedback; and
helping students to collect evidence of learning and to use that evidence to
communicate with peers and adults.

3 Examining Students’ Work

Cary I. Sneider .................................................................................................. 27

Examining student work is an essential aspect of teaching, yet it is easy to miss

opportunities to learn about how students are interpreting—or misinterpreting—
the lessons we present to them. In this chapter the author shares insights
concerning the techniques he has found to be most effective in tuning in to his
students, so that he can adjust his teaching methods and content in order to be a
more effective teacher.

v
 4 Assessment of Inquiry
Richard A. Duschl ............................................................................................. 41

This chapter provides an overview of frameworks that teachers can use to conduct
assessments of students’ engagement in scientific inquiry. The author examines two
factors that are central to such assessment. One factor is the design of classroom
learning environments, including curriculum and instruction. The second factor is
the use of strategies for engaging students in thinking about the structure and
communication of scientific information and knowledge. The chapter includes an
up-to-date description of nine National Science Foundation–supported inquiry
science units.

5 Using Questioning to Assess and Foster Student Thinking

Jim Minstrell and Emily van Zee ...................................................................... 61

Questioning can be used to probe for understanding, to initiate inquiry, and to

promote development of understanding. The results from questioning, listening,
and assessment also can be used by teachers to promote their own growth as
professionals. This chapter presents a transcript of a class discussion in which
questioning is used to assess and foster student thinking. After developing this
context for questioning, the authors discuss purposes and kinds of questions, then
revisit the context to demonstrate how the results of assessment through
questioning can be used to guide the adaptation of curriculum and instruction.

6 Involving Students in Assessment

Janet E. Coffey .................................................................................................. 75

While much of the responsibility for classroom assessment lies with teachers,
students also play an important role in meaningful assessment activity. Bringing
students into the assessment process is a critical dimension of facilitating student
learning and in helping to provide students with the tools to take responsibility for
their own learning. The author examines the role students can play in assessment
through a closer look at a middle school program where students actively and
explicitly engage in all stages of the assessment process.

7 Reporting Progress to Parents and Others: Beyond Grades

Mark Wilson and Kathleen Scalise ................................................................... 89

As science education moves increasingly in the direction of teaching to standards,

teachers call for classroom assessment techniques that provide a richer source of
“rigorous and wise diagnostic information.” Student-to-student comparisons and
single grades are no longer enough, and here the authors describe a new type of
criterion-based assessment to track individual learning trajectories. It can be
embedded in the curriculum, easily used in the classroom, customized by grade
level, subject area, and standard set, and controlled by the classroom teacher.

vi
8 Working with Teachers in Assessment-Related Professional Development
Mistilina Sato .................................................................................................. 109

Professional development related to everyday classroom interactions can require a

shift in the teacher’s priorities in the classroom from a focus on managing activity
and behavior to a mind-set of managing learning opportunities. This essay looks
closely at a professional development approach that sees the teacher not only as a
professional engaged in learning and implementing new strategies for assessing
students, but also as an individual who is undergoing personal change in beliefs.

9 Reconsidering Large-Scale Assessment to Heighten Its Relevance

to Learning
Lorrie A. Shepard ............................................................................................ 121

In contrast to classroom assessments that can provide immediate feedback in the

context of ongoing instruction, large-scale assessments are necessarily broader
survey instruments, administered once-per-year and standardized to ensure
comparability across contexts. Classroom and large-scale assessments must each
be tailored in design to serve their respective purposes, but they can be symbiotic
if they share a common model of what it means to do good work in a discipline and
how that expertise develops over time. Three purposes of large-scale assessment
programs are addressed—exemplification of learning goals, program “diagnosis,”
and certification or “screening” of individual student achievement. Particular
attention is given to the ways that assessments should be redesigned to heighten
their contribution to student learning. In addition, large-scale assessments are
considered as both the site and impetus for professional development.

10 Reflections on Assessment
F. James Rutherford ........................................................................................ 147

As a context for thinking about the claims made in this book, some of the
circumstances that have influenced the demand for and character of assessment in
general are noted. The argument is then made that the substantial lack of
coherence in today’s assessment scene is due in large part to policies and
practices that fail to recognize that there is no one best approach to assessment
and that assessment and purpose must be closely coupled.

Index ............................................................................................................... 159

vii
 Acknowledgments
W e wish to acknowledge our debt to Rodger Bybee and Harold Pratt. A National
Science Teachers Association (NSTA) “yearbook” was Rodger’s vision, and
he served as the editor of the first volume, Learning Science and the Science of
Learning. Rodger and Harold helped to identify classroom assessment as a focus for
a subsequent volume. Their discussions were the impetus for this collection. Many
people within the NSTA organization supported the launch of the annual series and
this volume—particularly Gerry Wheeler, executive director of NSTA, and David
Beacom, NSTA publisher. We also thank Carolyn Randolph, current NSTA presi-
dent, for her support.
The authors of the separate chapters have been extraordinarily responsive to
suggestions from us and from the reviewers and have made their revisions in a timely
manner. Their dedication to improving science education and their desire to engage
in a dialogue with and for teachers are inspiring.
We also thank the following individuals for their reviews of this volume:
Ann Fieldhouse, Edmund Burke School, Washington, D.C.
Patricia Rourke, Potomac School, McLean, Virginia
Rick Stiggins, Assessment Training Institute, Portland, Oregon
The reviewers provided thorough and thoughtful feedback. The volume is better
for their efforts.
Special acknowledgments are due for Claire Reinburg and Judy Cusick at NSTA.
Claire initiated early communication and helped with logistics from the outset. Judy
Cusick has provided support in innumerable ways, not least as careful production
coordinator for the entire volume. Claire and Judy have guided us smoothly through
the process, coordinated reviews, and served as invaluable resources. We thank them
both for their time and effort.
Although the reviewers provided constructive feedback and made many sugges-
tions, they did not see the final draft before release. Responsibility for the final text
rests entirely with the chapter authors and the editors.

ix
 About the Editors
J Myron (Mike) Atkin is a professor of education and human biology at Stanford
University. He directs the National Science Foundation–supported Classroom As-
sessment Project to Improve Teaching and Learning (CAPITAL) at Stanford. He
formerly chaired the National Research Council committee that prepared an adden-
dum to the National Science Education Standards on assessment in the science class-
room. In addition to having taught at the elementary and secondary school levels, he
has served as dean of education at both the University of Illinois and at Stanford. For
much of his career, he has emphasized the central role of teachers in designing high-
quality science education programs and now focuses on strengthening that role in
any comprehensive system of assessment and accountability.

Janet E. Coffey, a former middle school science teacher, currently works with teachers
on the Classroom Assessment Project to Improve Teaching and Learning (CAPI-
TAL) at Stanford University. CAPITAL is a National Science Foundation–funded
effort that seeks to better understand teachers’ assessment practices as they strive to
improve their practices. She worked at the National Research Council as a staff mem-
ber on the development of the National Science Education Standards. She earned her
Ph.D. in science education from Stanford University.

x
 Introduction
J Myron Atkin and Janet E. Coffey

T he assessment that occurs each day in the science classroom is often overlooked
amidst calls for accountability in education and renewed debates about external
testing. We believe that daily assessment should be moved to the foreground in these
broad discussions and receive greater attention in policy circles. Research points to
the positive influence that improved, ongoing classroom assessment can have on
learning. Documents that offer visions for science education—such as the National
Science Education Standards (NRC 1996) and materials associated with Project 2061
(AAAS 1993)—strongly echo that sentiment.
Too often, assessment is used synonymously with measurement (in the form, for
example, of tests and quizzes). The misconception that they are the same minimizes
the complexity and range of purposes of assessment. As teachers are aware, assess-
ment in any classroom is rich and complicated. It includes tests and quizzes, of course,
but also students’ other work, students’ utterances made while conducting lab inves-
tigations, and class discussions in which students share their explanations for what
they have observed. It raises issues of quality and of what counts as evidence for
learning. It happens in reflection, in exchanges that occur countless times a day
among teachers and students, and in feedback on work and performance. When we
reduce assessment to a specific event or “thing”—the test, the lab practical, the grade—
it is easy to overlook the interactive assessment that occurs each day in the class-
room.
Assessment operates to improve student learning, not solely measure it, when it
is used to move the student from his or her current understanding to where the stu-
dent would like to be (or where the teacher would like the student to be). To cross
that gap, the teacher and student must use feedback from assessments. Quality and
use of information become crucial in that process. Sometimes the way to bridge that
gap is clear, with obvious starting and destination points. Sometimes, however, it is
rendered more complex when the destination point is not so clear, as in inquiry-
based science investigations. In inquiry learning, students ask their own questions
and face multiple paths to answering those questions. Students must continually re-
flect on what they are doing and ask themselves, Where am I now? Where do I need
to go? How do I get there? What is my evidence?
Everyday assessment is local and contextual. It depends more on the skills, knowl-
edge, attentiveness, and priorities of teachers and students than on any particular set

xi
 of protocols or strategies. Opportunities for meaningful assessment occur numerous
times each day—within interactions, in conversations, by way of observations, and
even as part of traditional assessment. A test review or a discussion about criteria of
good lab reports each provides opportunities for assessment-related conversations.
When the K–12 teacher attends closely to questions and responses and observes
students as they engage in inquiry, he or she will gather important assessment infor-
mation on a daily basis.
The challenge for teachers and students is to maintain a classroom assessment
culture that supports learning for all students. In this culture, teaching and assess-
ment are so closely entwined that the two become difficult to differentiate; students
engage actively and productively in assessment; clear, meaningful criteria exist; and
teachers provide high-quality, regular feedback. In classrooms with supportive as-
sessment cultures, the focus is on learning rather than on grades, on progress rather
than on fixed achievement, on next steps rather than on past accomplishments.
Achievement and accomplishments serve an assessment purpose, as can grades; how-
ever, they alone do not assessment make.
In any discussion of assessment, issues associated with teaching, learning, and
curriculum quickly arise, as do questions of equity, fairness, and what counts as
knowing and understanding. These interconnections, after all, constitute one reason
that the topic of assessment is so important. In this volume, the authors address the
interconnections and provide guidance and illuminate challenges—at the same time
that they maintain a sense of the bigger picture. Although our primary audience is
classroom teachers, we also hope that this book will be informative and useful to
those who work with teachers, in either professional development or teacher educa-
tion capacities, and to school administrators, program designers, policy makers, and
parents.
In Chapter 1, Paul Black frames the assessment that occurs every day in the
classroom within the broader realm of assessment and its multiplicity of purposes.
For everyday assessment to contribute to improved learning, he argues, action must
be taken based on evidence. This action could come in any number of forms, not the
least of which include altering teaching, modifying curriculum, or providing stu-
dents with useful feedback. Even the most informative data do little good to students
and teachers if the data do not feed into learning activities. Black provides an over-
view of some theoretical assumptions embedded in a view of assessment that sup-
ports learning, and he highlights issues related to assessment in science classrooms.
He also addresses the tensions and synergistic opportunities that exist when trying to
manage the many purposes assessment must serve.
In Chapter 2, Anne Davies discusses the critical role teachers play in ongoing
assessment. Teachers not only identify and articulate learning goals, they also find
samples of work to meet those goals, consider the type of work that would serve as
evidence of attainment, and assist students in developing an understanding of the
goals. Her discussion highlights the relationships among assessment, curriculum,

xii
and teaching. As we see in many of these chapters, the three topics are, at times,
difficult to distinguish from one another.
Cary Sneider, author of Chapter 3, examines what can be gained through careful
consideration of student work, which includes what students do, say, and produce.
Through reflecting on his own career as a teacher, which began as an Upward Bound
tutor, his work as a curriculum developer, and now his present position in an educa-
tion department at a science museum, Sneider makes a case for tuning into what
students do and say to gain insights into the nature of their ideas and understandings.
He reminds us that to teach is not necessarily to learn. Assessment serves as a critical
link between the two processes. In this essay, Sneider encourages teachers to move
beyond the use of written responses and final products and to see the value of listen-
ing to, conversing with, and observing students as they engage in activities as well.
The focus of Chapter 4 is on assessing inquiry science. Richard Duschl offers a
metaphor of “listening to inquiry” to guide teachers as they support inquiry in their
classrooms. The organic nature of inquiry excludes more traditional tests as a means
of helping students move forward in pursuit of investigation. With inquiry, there are
few clear destinations or explicit goals toward which to steer. Duschl looks at the
design of learning environments that engage students in thinking about the structure
and communication of scientific ways of thinking. Deliberations about what counts
as data, evidence, and explanation would give inquiry a voice, to which student and
teacher alike can listen.
Chapter 5 explores issues related to questioning in the classroom. Specifically,
Jim Minstrell and Emily van Zee highlight the importance of developing and asking
questions that help students consider scientific phenomena and elicit thoughtful and
in-depth responses that reveal insight into student understandings. The authors dis-
cuss ways in which these insights can be used by teachers to plan additional activi-
ties, modify existing ones, and inform future questions. A message that emerges in
this chapter is the important role that subject matter plays when teachers are listen-
ing to student responses and following those responses with further questions.
Much of the classroom assessment literature focuses on the roles of teachers.
Chapter 6 shifts the focus to students. Janet Coffey addresses the integral role stu-
dents can play in everyday classroom assessment. Student participation in and with
assessment activities can help clarify and establish standards for quality work and
help students to identify the bigger picture of what they are learning. Through les-
sons learned from a middle school program where students had opportunities and
expectations to participate actively in assessment, Coffey identifies ways to support
students as they become self-directed with respect to assessment.
Mark Wilson and Kathleen Scalise take on issues related to grading in Chapter 7.
Grades quickly can become the centerpiece of any discussion of assessment at the
classroom level. Wilson and Scalise discuss the meanings that underlie grades, the
information they convey, and what they often represent. Grades, they argue, often
reflect a teacher’s perception of a student’s effort rather than what the student has

xiii
 learned. Any time a large amount of information is reduced to a single letter grade,
much of the useful information behind the grade gets lost. The authors offer a frame-
work for assessment tools that can generate useful assessment data for classroom
purposes and for reporting purposes. Assessment tools such as the ones they share
can yield useful and high-quality assessment data for teachers, students, parents, and
other interested parties. An element of their model includes teacher “moderation
meetings,” where teachers discuss student work, scores, and interpretations. These
deliberations can provide powerful professional development opportunities.
In Chapter 8, Mistilina Sato discusses assessment-related professional develop-
ment for teachers. She advocates for a change in the teacher’s image from that of
monitor and maker of judgments to that of manager of learning opportunities. For
lasting change, she argues, teacher professional development must go beyond learn-
ing new strategies and skills to take into consideration the teacher as a person. Sato
points out that teachers enter the classroom as individuals with beliefs, values, back-
grounds, assumptions, and past experiences that shape who they are in the classroom
and the actions they take. Reform efforts that overlook these personal dimensions of
teachers will find minimal success. She shares lessons from a National Science Foun-
dation–funded assessment effort currently underway with Bay Area middle school
science teachers.
Lorrie Shepard explores the assessment landscape beyond the classroom in Chap-
ter 9. Specifically, she discusses the ways in which large-scale assessment could be
redesigned to heighten contributions to student learning. Even within the realm of
large-scale, external testing, a myriad of purposes clamor for attention. Due to con-
straints such as time and cost, these assessments often take the form of traditional
tests. Shepard points out that all tests are not the same. The intended purposes of a
test shape the content, criteria for evaluating, and technical requirements; all needs
cannot be met through a single test. Shepard calls for external tests to embody im-
portant learning goals, such as those set forth in the National Research Council’s
National Science Education Standards (1996) or AAAS’s Benchmarks for Science
Literacy (1993). Examples of actual efforts underway in districts and states help
show possibilities for lessening tensions.
James Rutherford provides a historical overview of educational assessment and
reform in science education in Chapter 10. The shifting focus of school, district, and
national reform efforts has made sustained attention to any one initiative difficult
and frustrating at best. Assessment is no exception. We are quick to react to “cri-
ses”—real or imagined—rather than take proactive steps, with a long-term view and
guidance from solid research literature, toward higher quality science instruction.
Rutherford proposes that teachers, parents, and others within the educational com-
munity assess assessments by asking critical questions about the assessments as well
as the information they provide. He concludes his chapter by offering questions for
all of us to consider. In doing so, he sets a frame for use of this book as a tool for
professional development. Generating discussion among teachers about some of the

xiv
ideas raised and addressed in any or all of these chapters would be a valuable out-
come of the book.
As Rutherford indicates, this collection may raise more questions than it an-
swers. We, too, hope that this volume contributes to the practical and professional
development needs of teachers. We hope it illuminates the importance of attending
to everyday assessment, raises issues worthy of reflection and consideration, and
offers some practical suggestions. Thinking and acting more deliberately with re-
gard to the ongoing assessment that occurs each and every day in the classroom can
go a long way to making our classrooms more conducive to learning for all students.

References
American Association for the Advancement of Science (AAAS). 1993. Benchmarks for science lit-
eracy. New York: Oxford University Press.
National Research Council (NRC). 1996. National science education standards. Washington, D.C.:
National Academy Press.

xv
 CHAPTER 1

The Importance of Everyday Assessment

Paul Black

Paul Black is Emeritus Professor of Science Education at King’s College London. He retired in
1995, having spent much of his career in curriculum development and research in science edu-
cation. In 1987–1988 he was chair of the government’s Task Group on Assessment and Testing,
which set out the basis for the United Kingdom’s national testing. More recently, he has served
on advisory groups of the National Science Foundation and is a visiting professor at Stanford
University. His recent research with colleagues at King’s has focused on teachers’ classroom
assessments. This work has had a significant impact on school and national policies in the UK.

The Context: Conflicts and Synergies of Purpose

I f assessment is understood in a broad sense—that is, to signify all those processes

and products that provide evidence about what is happening—it is immediately
evident that in education, assessments are all-pervasive. They influence, even rule,
the context within which teachers work, but are also an essential part of the everyday
minutiae of teachers’ work with their students. This broad interpretation leads to a
need to impose some structure on any discussion, so I start here with a discussion of
purposes of assessment.
Three main purposes can be distinguished. Assessment can serve accountability,
certification, or learning.
• For accountability, the evidence has to be broad in scope and designed to high-
light needs that might be met through policy actions. This purpose can be served
by testing samples of the student population, as well as by collecting a range of
other data so that interpretation might be served by exploring interrelationships.
Such a broad program might be called an evaluation, to distinguish it from as-
sessment, which is seen as only one of its components. Such surveys as the
National Assessment of Educational Progress (NAEP), the Third International
Mathematics and Science Study (TIMSS), and the Program for International
Student Assessment (PISA) are examples. However, an accountability exercise
can be designed for a different purpose—to drive improvement by linking the
results to public exposure and other rewards or punishments. It is usual, albeit
unnecessary, to test every student.
• For certification, the audience is the individual students and those who care about
them, together with potential employers and those controlling admission to the
further stages of education. For both accountability and certification, the evi-
dence is often limited to results of formal written tests drawn up and marked by
agencies external to the school. However, it is possible, and indeed normal in
some countries, to use evidence provided from within schools to help meet these
purposes. The collection of evidence need not then be limited to formal, timed
tests; the broader term assessment is then appropriate.

1
CHAPTER 1

• For learning, the dominant term can be assessment, but evaluation and diagnosis
sometimes creep in. The purpose is clear enough in principle, while action based
on the evidence can range from minute-by-minute feedback to adjustment of the
lesson plans for next year.
Over all of this spectrum, the concepts of reliability and validity are central. To
claim reliability, one has to be sure that if the student were to take a parallel form of the
same assessment on another occasion, the result would be the same. This claim is
rarely supported by comprehensive evidence. For certification, and internal school
decisions about tracking and grading, weak reliability is serious because the effects on
students are hard to reverse. In assessment for learning, reliability is harder to achieve,
but it is less serious an issue provided that the teacher’s approach is flexible so that the
effects of wrong decisions can be discerned and corrected within a short time.
Validity is a more serious issue. The key to the concept is that the inferences that
are made on the basis of an assessment result are defensible (Messick 1989). In the
case of a formal written test, any inference that goes beyond saying that the student
did well on this test on this date requires justification. But inferences do go far be-
yond such limitation, for it is often assumed that the student could do well in any test
in any part of the subject on any occasion in the future, understands the nature and
structure of the subject, is competent in exercising the discipline of the subject, and
is well-equipped to benefit from more advanced study.
One source of the limitation on validity is that formal tests have to be short and
inexpensive to administer and mark. However, many aspects of understanding and
competence can only be displayed over extended periods of time in unconstrained
conditions. It is often claimed that a written test might serve as a surrogate for such
activity, but those claims usually lack empirical support (Black 1990; Baxter and
Shavelson 1994). For this reason, significant efforts have been invested in develop-
ing a broader range of methods of assessment—for example, new types of evidence
for performance (e.g., notebooks written by students about a science investigation)
and new procedures to improve and attest to the evidence that teachers can gather
from their more extensive interactions with their students. However, what then needs
to be made clear is whether these innovations are designed to serve accountability
and certification, or to serve learning, or to serve all three purposes.
Demands of accountability are often seen to impose such high-stakes pressures
on teachers that assessment for learning is not seriously considered. When the public
believes that wholly external written tests are the only trustworthy evidence for stu-
dents’ achievement, and then teachers believe that the pressures to succeed can only
be met by ad hoc and inadequate methods of learning, dominance of the accountabil-
ity purpose seems inevitable. Yet critical scrutiny of the claims for the superiority of
formal tests is not supported by careful scrutiny of their reliability and validity. It has
been demonstrated that when any such test is newly introduced, performances will
rise steadily for a few years and then level out. If the test is then changed, perfor-

2 National Science Teachers Association

 CHAPTER 1

mance will drop sharply and then rise again to level off in a further few years (Linn
2000). Thus, the high-stakes performances that ensue are, at least in part, artifacts of
the pressures of the particular test being used rather than valid educational measures.
Ironically, the belief of most teachers that they have to “teach to the test” rather than
teach for sound understanding is also unjustified, for there is evidence that the latter
strategy actually leads to better performances even on the tests to which the former,
narrow teaching approach is directly aligned.
The problem for policy makers is to stand back from current assumptions and
radically rethink their approach to reconciling the different purposes of assessment. A
key feature of any such appraisal must be to strengthen teachers’ own skills in assess-
ment so that the public can have confidence in the capacity of teachers to serve all three
purposes in a valid and rigorous way. The problem for teachers and schools is to im-
prove practices, to be clear about the purposes that those practices are designed to
serve, and to resolve any conflicts as best they can within today’s constraints.

How Do People Learn?

Three common assumptions about learning, which have their origins in behaviorist
psychology (Collins 2002), are that (1) a complex skill can be taught by breaking it
up and teaching and testing the pieces separately; (2) an idea that is common to
action in many contexts can be taught most economically by presenting it in abstract
isolation so that it can then be deployed in many situations; and (3) it is best to just
learn about new things first and not try for understanding—that will come later. A
test composed of many short, “atomized,” out-of-context questions and the practice
of “teaching to the test” are both consistent with those assumptions.
Contemporary understanding of the ways that children learn looks at the process
quite differently (Wood 1998). A first important lesson is illustrated by the following
quotation:

… even comprehension of simple texts requires a process of inferring and

thinking about what the text means. Children who are drilled in number facts,
algorithms, decoding skills or vocabulary lists without developing a basic
conceptual model or seeing the meaning of what they are doing have a very
difficult time retaining information (because all the bits are disconnected)
and are unable to apply what they have memorized (because it makes no
sense). (Shepard 1992, 303)

Current “constructivist” theories focus attention on the models that a learner em-
ploys when responding to new information or to new problems. Even for the restricted
task of trying to memorize something, one can do better if one already has some scheme
built on relevant understanding and tries to link the new knowledge with existing pat-
terns. It appears that memory is rather like a filing cabinet—that is, the storage is only
useful insofar as the filing system makes sense so that one knows where to look.

Everyday Assessment in the Science Classroom 3

CHAPTER 1

More generally, learning always involves analyzing and transforming any new
information. Piaget stressed that such transformation depends on the mind’s capac-
ity to learn from experience—within any one context, we learn by actions, by self-
directed problem-solving aimed at trying to control the world. Abstract thought evolves
from such concrete action. It follows that

… teaching that teaches children only how to manipulate abstract procedures

(e.g., learning how to solve equations) without first establishing the deep
connections between such procedures and the activities involved in the
solution of practical concrete problems (which the procedures serve to
represent at a more abstract level) is bound to fail. (Wood 1998, 9)

Here, context is important. An individual’s general capacity for abstract thought

may be exhibited in, say, family relationships but be quite absent in, say, physics
concepts. It is also evident that transformations of incoming ideas can only be achieved
in light of what the learner already knows and understands, so the reception of new
knowledge depends on existing knowledge and understanding. It follows that

… learning is enhanced when teachers pay attention to the knowledge and

beliefs that learners bring to a learning task, use this knowledge as a starting
point for instruction, and monitor students’ changing conceptions as
instruction proceeds. (Bransford, Brown, and Cocking 1999, 11)

Research in the learning of science has shown that many learners resist changes
in their everyday and naive views of how the natural world works, despite being able
to play back the “correct” science explanations in formal tests. So teaching must
start by exploring existing ideas and encouraging expression and defense of them in
argument, for unless learners make their thinking explicit to others, and so to them-
selves, they cannot become aware of the need for conceptual modification. The next
step is to find ways to challenge ideas, usually through examples and experiences
that are new to pupils and that expose the limitations of their ideas. It follows that
assessment for learning must be directed at the outset to reveal important aspects of
understanding and then be developed, within contexts that challenge pupils’ ideas, to
explore responses to those challenges.
Such classroom activities can be a basis for learning development at a more
strategic level. Research studies have shown that those who progress better in learn-
ing turn out to have better self-awareness and better strategies for self-regulation
than their slower learning peers (see, e.g., Brown and Ferrara 1985). Thus self-as-
sessment becomes an important focus of assessment for learning. Peer-assessment
also deserves priority, for it is by engaging in critical discussion of their work with
their peers that learners are most likely to come to be objective about the strengths
and weaknesses of their work. The main message is that students need to understand

4 National Science Teachers Association

 CHAPTER 1

what it means to learn. They need to monitor how they go about planning and revis-
ing, to reflect on their learning, and to learn to determine for themselves if they
understand. Such skills enhance metacognition, which is the essential strategic com-
petence for learning.
When the teacher starts from where the learners are, helps them to take respon-
sibility for their learning, and develops peer- and self-assessment to promote
metacognition, the teacher becomes a supporter rather than a director of learning.
This idea was taken further by Vygotsky (1962), who emphasized that because learning
proceeds by an interaction between the teacher and the learner, the terms and con-
ventions of the discourse are socially determined, and its effectiveness depends on
the extent to which these terms and conventions are shared. His influence can be
seen in the following statement:

Participation in social practice is a fundamental form of learning. Learning

involves becoming attuned to the constraints and resources, the limits and
possibilities, that are involved in the practices of the community. Learning is
promoted by the social norms that value the search for understanding.
(Bransford, Brown, and Cocking 1999, xii)

Wood, Bruner, and Ross (1976) developed this approach by introducing the
metaphor of “scaffolding”—the teacher provides the scaffold for the building, but
the building itself can only be constructed by the learner. In this supportive role, the
teacher has to discern the potential of the learner to advance in understanding, so that
new challenges are neither too trivial nor too demanding. Vygotsky called the gap
between what learners can do on their own and what they can do with the help of
others the “zone of proximal development.” One function of assessment is to help to
identify this zone accurately and to explore progress within it.
All of this shows how important it is for the teacher to develop a classroom
discourse through which all students learn to internalize and use the language and
the norms of argument used by scientists to explain phenomena and to solve prob-
lems (Bransford, Brown, and Cocking 1999, 171–75). Thus, the way students talk
about science, both in informal and formal terms, is important formative assessment
material for teachers (Lemke 1990).
Because a learner’s response will be sensitive to the language and social context
of any communication, it follows that assessments, whether formative or summative,
have to be very carefully framed, both in their language and context of presentation,
if they are to avoid bias (i.e., unfair effects on those from particular gender, social,
ethnic, or linguistic groups). The importance of context is also a far-reaching issue.
For example, tests that ask questions about mathematics that might be used in soci-
ety in general might be failed by a student who can use the same mathematics in a
familiar school context, and vice versa (Boaler 2000).

Everyday Assessment in the Science Classroom 5

CHAPTER 1

The discussion in this section has been limited to the cognitive aspect of links
between assessment and student response. Other important elements will be explored
in the next section.

Assessment for Learning

In the broadest sense of the word, assessment is something that we do all the time.
We encounter a new situation, make a judgment about the meaning of what is hap-
pening, and decide what to do next. The evidence of our encounters continually
shapes and reshapes our actions. Our actions may be more effective if we are flex-
ible—that is, if we are prepared to modify our intentions in the light of events. They
might also be more effective if we probe the situation carefully in order to ensure
that we understand what is going on before jumping to conclusions.
All of this applies in particular to life inside the classroom. The teacher has some
understanding of the state of the students’ learning, and must decide what to do next.
This understanding is bound to be imperfect, but it can be refined by setting up
activities through which the students will provide more evidence. So the cycle is to
evoke or explore, to interpret the feedback, and then to modify the teaching actions.
The key to formative assessment lies in this flexibility—the capacity to change
what was planned in order to meet the needs exposed by the evidence. The prospects
are improved by finding ways to so elicit evidence that key features of the learning
are illuminated; this can be called assessment for learning. However, there is little
point in doing this if the evidence is not used to fashion what happens next; only
when such refashioning occurs does the assessment become formative assessment
(Black and Wiliam 1998). It is necessary to stress this feature; some teachers believe
that they are engaged in formative assessment when, even though they are listening
to their students, they then proceed with a lesson plan despite what they have heard.
The concept of assessment is a very general one—in the classroom it is happen-
ing all the time. When the looks on the faces of the students, or their written work, or
their oral answers to a question are appraised, the teacher is assessing. A quiz or
written test is also an occasion for assessment, but it is only one among many possi-
bilities. As outlined above, the quality of the assessment feedback will depend on the
quality of the interventions that evoke that feedback. It is here that the theories of
learning become relevant. A question that asks about a technical term—for example,
“What is the unit of current?”—serves a very different purpose from one that probes
understanding—for example, “Does the current get used up as it goes through the
light bulb?” The former question looks for recall, and there is little that can be done
with the response apart from noting whether or not it is correct.
The latter question probes rather deeply, for it has a basis in research evidence
that the notion of “current used up” is a common misconception. One way for a
teacher to respond is to listen to an answer and then to tell the class the right answer;
such action, however, is not responsive to the evidence. A second way is to explore
opinions among a class to stimulate a discussion about the concept of current, which

6 National Science Teachers Association

 CHAPTER 1

could lead to a test with a simple circuit with ammeters on either side of the bulb.
Table 2: Professional Development Standard B
This second way is formative, for it explores understanding, involves students
actively in the learning process, and follows the learning principle that effective
learning starts from where learners are and helps them to see the need to change.
Furthermore, insofar as discussion is evoked, the learning is in the context of a dis-
course in a learning community rather than being a one-way transmission. Thus,
there is an intimate connection between good formative assessment and the imple-
mentation in the classroom of sound principles of learning.
Similar arguments can be applied to other learning activities, notably the mark-
ing of written work, the use of peer- and self-assessment, the possibilities for the
formative use of written tests, and so on. As teachers change to make formative
assessment a constant feature of their work, they will inevitably be changing their
roles as teachers. They have to be more interactive with their students, and they have
to give them more responsibility for learning. This leads to a change in role, from
directing students to empowering them (Black et al. 2002).

Assessment and the Student

Change in the role of teachers must lead, in the formative classroom, to changes in
the roles of students. One type of change will be cognitive. As questions become
more searching, and as the classroom routine is altered to depend more on the active
involvement of the learner, students will find that they have to think more and take
responsibility for doing more of the work themselves. Because formative work re-
quires the elicitation of students’ ideas, they will also have to be more willing to
expose these ideas and to submit them to discussion and challenge by their peers as
well as by their teachers. This calls for a change in the expectations of students, and
such change may disconcert many, who are likely to resist. Thus it becomes impor-
tant to build a supportive environment. Students must learn to listen to one another,
to respect one another’s opinions, and to understand that learning works through
exploration and challenge, not by rewarding those who are right and labeling those
who are wrong.
However, learning is not just a cognitive exercise—it involves the whole person.
The need to motivate pupils is evident, but it is often assumed that motivation should
consist of extrinsic rewards, such as merits, grades, gold stars, and prizes. Ample
evidence challenges this assumption. If a learning exercise is seen as a competition,
then everyone is aware that there will be losers as well as winners; those who have a
track record as losers will see little point in trying. Thus, the problem is to motivate
everyone, even though some are bound to achieve less than others. In tackling this
problem, teachers need to realize that the type of feedback they give is very impor-
tant. Many research studies support this assertion, as the following citations attest:
• Pupils told that feedback “will help you to learn” learn more than those told that
“how you do tells us how smart you are and what grades you’ll get”; the differ-
ence is greatest for low attainers (Newman and Schwager 1995).

Everyday Assessment in the Science Classroom 7

CHAPTER 1

• Those given marks as feedback are likely to see the marks as a way to compare
themselves with others (ego-involvement); those given only comments see such
feedback as helping them to improve (task-involvement). The latter group out-
performs the former (Butler 1987).
• In a competitive system, low attainers attribute their performance to lack of “abil-
ity” and high attainers attribute their performance to effort. In a task-oriented
system, all attribute their performance to effort, and learning is improved, par-
ticularly among low attainers (Craven, Marsh, and Debus 1991).
• A comprehensive review of research studies of feedback showed that feedback
improved performance in 60 percent of the studies. In the cases where it was not
helpful, the feedback turned out to be merely a judgment or grading with no
indication of how to improve (Kluger and DeNisi 1996).
In general, feedback in the form of rewards or grades enhances ego rather than task
involvement. It can focus pupils’ attention on their “ability” rather than on the im-
portance of effort, thereby damaging the self-esteem of low attainers and leading to
problems of “learned helplessness” (Dweck 1986). Feedback that focuses on what
needs to be done can encourage all students to believe that they can improve. Such
feedback enhances learning, both directly through the effort that can ensue and indi-
rectly by supporting the motivation to invest such effort.

Assessment across Subjects

Everyday assessment is not an abstract idea; it is a concrete activity that the science
teacher conducts in and through the stuff of science education. While there are ge-
neric principles applicable to any learning, practical implementation is bound to be
different in the teaching of science and the teaching of, say, history.
The formulation of insightful oral or written questions and the subsequent devel-
opment of dialogue through which students become involved in their own learning
are essential components of formative assessment. For example, a useful question
about heat transfer can be based on a picture of three imaginary children arguing
about the melting of a snowman. The scenario is that the sun is shining, and there is
a breeze blowing, and child A suggests that they wrap a black coat around their
snowman to stop the sun from melting it. Child B objects that the black coat will
warm up the snowman, and child C says that it all depends on the wind. The class
can be asked to say what they think about the arguments of these three children. The
question is conceptually rich in that it can be used to open up discussion of radiation,
conduction, and convection. But it has two other features. One is that it has the
potential to elicit a well-known misconception—namely, that a coat actively warms
you up rather than reducing the outward flow of heat. The second feature of the
question is that it is likely to interest the children because it uses a context, and a
practical need for decision, with which they might identify. Such knowledge, about
the way children might think and might be interested, is called pedagogical content

8 National Science Teachers Association

 CHAPTER 1

knowledge. Science teachers can only plan and conduct an exercise of this type on
the basis of understanding of the concepts involved and of the relevant pedagogical
content knowledge.
In this snowman example, there is a set of “correct” ideas and a “right answer,”
albeit a complex one. More generally, there is in any intellectual discipline a spec-
trum of possible issues, with those issues having well-defined answers lying at one
end and those with a variety of good answers lying at the other. Emphasis on inquiry
works at the latter end. It calls for a pedagogy that promotes investigations in which
students have to exercise initiative and follow a variety of paths and for which the
criteria of achievement are concerned with the quality of strategies and arguments
rather than with attainment of a single, well-defined outcome. The type of formative
“scaffolding” needed here is more akin to that of the teacher of English who is help-
ing students to develop their individual writing styles than to that of the same teacher
concerned with basic rules of punctuation.
The current trend of reform in science education is to shift the balance of learn-
ing toward the open-ended. This calls for emphasis on inquiry skills, but there is a
further and broader aim, which is to engage students in discussion of the effects of
the advances of science on society, issues that are, or ought to be, of evident impor-
tance to them. Such issues are almost always areas of controversy. A discussion of,
say, genetic variation might start with clarification of the concepts, but will change
style when the discussion moves on to whether to develop genetically modified crops
and the sale of genetically modified foods. What matters here is the quality of the
arguments that are deployed, and the teacher has to guide students, in the midst of
controversy, both to argue and to listen carefully (Osborne 2000). Teachers in the
social sciences might be more experienced in such work than their science colleagues.
Teachers have to be flexible in varying their classroom styles to promote differ-
ent types of learning. Indeed, many science teachers who are accustomed to dealing
with “fixed-answer” topics find it very hard to cope with discussions for which the
aim is to help learners be critical about the quality of the arguments, rather than
about the correctness of the outcome. The shift from “delivery” modes of learning to
interactive modes is an essential step in developing this flexibility. As hinted above,
science teachers struggling with the new challenges presented by the standards re-
forms might find help through study of the classroom practices of colleagues who
teach other subjects.

Summary of the Issues

Two main themes emerge from this chapter. The first is that tensions and synergies
exist between different purposes of assessment, raising problems for both policy
makers and classroom teachers. It is obvious that insofar as there is tension, priority
must be given to assessment for learning—the measurement of the outcome of school-
ing can hardly be more important than the means to secure that outcome.

Everyday Assessment in the Science Classroom 9

CHAPTER 1

If this is accepted, then all concerned must find ways to achieve a new equilib-
rium between the formative and summative purposes, which involves finding ways
in which summative systems can serve rather than damage good learning. One
obstacle to such achievement has been revealed by many surveys of teachers’ class-
room practices, for these have shown that assessment practices are one of the weak-
est aspects of many teachers’ work. In part this may well be due to the dominance of
summative tests, which provide many teachers with models for assessment—models
that are unhelpful because they are not designed as aids to everyday learning.
So there is a need to reconstruct, to build up, everyday assessment so that it
serves its first purpose—to elicit and to serve the needs of each learner. A secondary
reason why such a development is of central importance is that only when teachers
are more skilled and more confident in their own assessment practices might it be
possible for them to command the public confidence that is needed so that they can
play a strong part in any reconstruction of assessment and testing as a whole.
The second theme of this chapter is the necessity for reform of the classroom as
a learning environment. Four principles have been set out for the design of learning
environments (Bransford, Brown, and Cocking 1999, xvi–xvii): They should be
learner-centered, knowledge-centered, community-centered, and strong in assess-
ment for learning. To create a learner-centered environment, the teacher must start
from where the learner is and work by interaction and formative feedback to pro-
mote learning. Teachers must also attend to the learners’ understanding of how they
can learn and the learners’ confidence that they can all learn.
For a school environment to be knowledge-centered, the focus has to be on a
coherent approach to developing important knowledge and skills, an approach in
which all students can grasp what the purposes and values of science are and can feel
that they are actually taking part, albeit in a modest way, in sharing in the practices of
scientists. So active participation of all is clearly essential.
To be community-centered, the environment must again be one in which active
participation by everyone in the serious business of learning about science is a priority.
Through lively discourse in the classroom, the understanding of all students is ad-
vanced and refined and their power to participate in scientific argument is developed.
The dominance of competition through testing, which creates winners and losers, and
of labeling pupils as “bright” or “dull,” ought to be replaced by a shared belief that all
students can make progress—and can help one another to do so, as well.
Formative assessment is an essential ingredient for each of these aspects of a
learning environment.

References
Baxter, G. P., and R. J. Shavelson. 1994. Science performance assessments: Benchmarks and surro-
gates. International Journal of Educational Research 21: 279–98.
Black, P. J. 1990. APU science: The past and the future. School Science Review 72 (258): 13–28.

10 National Science Teachers Association

 CHAPTER 1

Black, P., and D. Wiliam. 1998. Inside the black box: Raising standards through classroom assess-
ment. Phi Delta Kappan 80 (2): 139–48.
Black, P., C. Harrison, C. Lee, B. Marshall, and D. Wiliam. 2002. Working inside the black box:
Assessment for learning in the classroom. London, UK: King’s College London, Department of
Education and Professional Studies.
Boaler, J. 2000. Introduction: Intricacies of knowledge, practice and theory. In Multiple perspectives
on mathematics teaching and learning, ed. J. Boaler, 1–17. Westport, NJ: Ablex.
Bransford, J. A., A. Brown, and R. Cocking. 1999. How people learn: Brain, mind, experience and
school. Washington, DC: National Academy Press.
Brown, A. L., and R. A. Ferrara.1985. Diagnosing zones of proximal development. In Culture, com-
munication and cognition: Vygotskian perspectives, ed. J. V. Wersch, 273–305. Cambridge: Cam-
bridge University Press.
Butler, R. 1987. Task-involving and ego-involving properties of evaluation: Effects of different feed-
back conditions on motivational perceptions, interest and performance. Journal of Educational
Psychology 79 (4): 474–82.
Collins, A. 2002. How students learn and how teachers teach. In Learning science and the science of
learning, ed. R. W. Bybee, 3–11. Arlington, VA: National Science Teachers Association.
Craven, R. G., H. W. Marsh, and R. L. Debus. 1991. Effects of internally focused feedback on en-
hancement of academic self-concept. Journal of Educational Psychology 83 (1): 17–27.
Dweck, C. S. 1986. Motivational processes affecting learning. American Psychologist (Special Issue:
Psychological science and education) 41 (10): 1040–1048.
Kluger, A. N., and A. DeNisi. 1996. The effects of feedback interventions on performance: A histori-
cal review, a meta-analysis, and a preliminary feedback intervention theory. Psychological Bulle-
tin 119 (2): 254–84.
Lemke, J. L. 1990. Talking science: Language, learning and values. Norwood, NJ: Ablex.
Linn, R. L. 2000. Assessments and accountability. Educational Researcher 29 (2): 4–14.
Messick, S. 1989. Validity. In Educational Measurement (3rd ed.), ed. R. L. Linn, 13–103. London:
Macmillan.
Newman, R. S., and M. T. Schwager. 1995. Students’ help seeking during problem solving: Effects of
grade, goal, and prior achievement. American Educational Research Journal 32 (2): 352–76.
Osborne, J. 2000. Science for citizenship. In Good practice in science teaching: What research has to
say, eds. M. Monk and J. Osborne, 225–40. Philadelphia, PA: Open University Press.
Shepard, L. A. 1992. Commentary: What policy makers who mandate tests should know about the
new psychology of intellectual ability and learning. In Changing assessments: Alternative views
of aptitude, achievement and instruction, eds. B. R. Gifford and M. C. O’Connor, 301–28. Boston
and Dordrecht: Kluwer.
Vygotsky, L. S.1962. Thought and language. New York: Wiley.
Wood, D. 1998. How children think and learn: The social contexts of cognitive development (2nd ed.).
Oxford: Blackwell.
Wood, D., J. S. Bruner, and G. Ross. 1976. The role of tutoring in problem solving. Journal of Child
Psychology and Psychiatry 17: 89–100.

Everyday Assessment in the Science Classroom 11

 CHAPTER 2

Learning through Assessment: Assessment

for Learning in the Science Classroom1
Anne Davies

Anne Davies has taught both elementary- and university-level students and has worked as a
school administrator and a district consultant. Currently, she is involved in research projects; is
preparing a facilitator’s guide to her book Making Classroom Assessment Work (Merville, BC:
Connections Publishing, 2000); and is conducting workshops and teaching graduate courses
internationally. She has been a recipient of the Hilroy Fellowship for Innovative Teaching as well
as a Social Sciences and Humanities Council doctoral fellowship.

A classroom assessment process that supports student learning is made up of

three general parts. First, teachers review the curriculum and standards docu-
ments and summarize in their own words the learning that students are expected to
accomplish. They collect and review samples and models that show what the learn-
ing looks like for students of a particular age range, and they think through what
kinds of evidence their students could produce to show they have learned what they
needed to learn. This summary guides their work with students and helps them ar-
ticulate the learning destination clearly and simply to others.
Once the big picture is established, the second task is for teachers to work with
students to bring them into the assessment process. They do this by talking about the
learning, showing samples and discussing what the evidence might look like, setting
criteria with students, asking them to self-assess, and giving them time to learn.
Third, as students become involved in assessment for learning, they learn to be-
come partners in a continuous assessment cycle that supports their learning. In this
cycle, students receive and give themselves descriptive feedback as they learn in rela-
tion to the criteria. From time to time they discuss their own learning in relation to what
needs to be learned, they self-assess, and set goals. They revisit and reset the criteria as
they learn more. They collect evidence of learning, present their evidence to others,
and receive feedback. This leads to more self-assessing, goal setting, and learning.

Seeing It Work
What does this process look like in a middle school science classroom? The rest of
this chapter presents an example of assessment that leads to learning. In this ex-
ample, students are learning how to collect evidence as part of a scientific research
project in their area.2

1
Copyright © 2003 by the National Science Teachers Association and Anne Davies. This chapter is
adapted from Chapter 1 in A. Davies. 2000. Making classroom assessment work. Merville, BC: Con-
nections Publishing.
2
My thanks to Debbie Jamieson, Pembroke School, Pembroke, Maine, for providing this example
from her classroom.

13
CHAPTER 2

Talking about the Learning

“Our class has been asked to help with an international environmental

problem. The problem is an invasion of green crabs. Have you noticed them
at the beaches? People around North America are watching and tracking the
invasion, trying to figure out the impact on local species, finding ways to
minimize the impact, and helping to identify possible solutions in advance of
their arrival elsewhere. Our class can help. It will mean spending time at the
beach collecting evidence and reporting on the impact here in our coastal
areas. Are you interested in becoming involved? There is a website devoted to
this project. Let’s go online and look at the data being collected, what people
are learning, and how we can help.”

When students engage in conversation before any learning activity or task, the
talk clarifies options, highlights possible plans, and encourages sharing of informa-
tion. As students work with teachers to define what learning is and what it looks like,
they shift from being passive learners to being actively involved in their own learn-
ing. By being engaged, they use and build neural pathways in their brains. This
means they are more likely to be able to access their learning easily and for a longer
period of time—way beyond the end of the unit or test.
Relevance is key to student learning. Why bother? Is it worth it? Does it count?
These are questions students ask as they are trying to figure out whether to commit to
learning. When students and teachers talk about why the learning is relevant to stu-
dents’ lives, students begin to understand more fully what needs to be learned.
When students are in partnership with teachers and others in support of learning,
more learning is possible because learning is socially mediated (Vygotsky 1978).
Berger and Luckman (1967, 3) explain: “All human knowledge is developed, trans-
mitted, and maintained in social situations.” Talking about the learning provides an
opportunity for collaborative feedback—from student and teacher perspectives. What
was learned? What evidence is there? What worked? What didn’t? What might be
done differently next time?

Showing Samples

“I’m really glad you are interested in becoming part of this project. As we
gather evidence about the green crabs it will be really important that we do it
properly so the data can be collated across research sites. Did anyone notice
what kind of data was being reported? What do you think we will need to
collect?”

When we give students samples to review and when we talk with them about
what is important in their learning, we help them build mental models of what suc-
cess looks like. When teachers spend time with students, sharing samples as well as

14 National Science Teachers Association

 CHAPTER 2

connecting what students already know to what they need to know, students’ under-
standing of what they will be learning and of what will be assessed increases. When
we involve students in this way, they use their prior knowledge and learn more about
the language of learning and assessment.
Before teachers show samples of student work that illustrate learning destina-
tions, they should decide on the purpose for the learning. For example:
1. Do you want all students to do the same thing in the same way? In collecting
scientific evidence there is a quality standard that needs to be adhered to. The
samples need to help students understand exactly what quality looks like. There
may be specific scientific criteria to demonstrate.
2. Do you want students to show what they know in a variety of ways? Teachers can
provide numerous samples that help students understand the range of what is
possible. This open-ended destination supports “all kinds of minds.”
3. Do you want students to show individual progress over time? Does quality have
many different looks or just one? When teachers know that different students are
at different points in their development, they can use samples that show a range
of what the learning looks like over time. In this case, samples become a kind of
road map of quality on the journey to learning.
Samples are particularly important for the students who struggle the most in our
classrooms. When samples represent work that is too far away from what students
know and are able to do, students cannot see how to get from where they are to where
they need to be. If samples are limited to showing what students already know and can
do, they fail to orient students toward what they need to know next. Selecting samples
needs to be a process that is carefully linked to the purpose of the learning and the
learning needs of students. When we use samples to illustrate the learning destination,
more students are able to participate in setting criteria, in giving themselves feedback,
and in assessing their own way to quality and success.

Setting and Using Criteria

“Our collected data is important to the success of this project. I want you to
think about what is important as you collect the data about the green crabs.
Let’s brainstorm a list. What’s important when we are collecting?” [Students
call out their ideas, and Ms. J. takes notes on the board until their ideas run
out.]

Make sure you have the right species.

Count more than once to make sure.

Count the green crabs.

Everyday Assessment in the Science Classroom 15

CHAPTER 2

Measure the square on the beach.

Be accurate when counting.

Count EVERY species in that area (even the barnacles!).

Survey all the animals in different spots.

Be exact about everything.

“I think that is a really good beginning. Different projects require different

kinds of data. Dr. R. is coming this afternoon. She is the local project
coordinator from the university. We will ask her what else is important as we
collect the data.”

Some of our students know what teachers want without it ever being made ex-
plicit. Some students simply don’t get it or have expectations and perceptions em-
bedded in their personal backgrounds (perhaps culture and family experience) that
do not match those a teacher may assume are understood by all students (Busick
2001; Delpit 1995). When we make the criteria explicit, share the process of learning
with each other, and assess (give descriptive feedback) according to the agreed crite-
ria, we give more students the opportunity to learn. We begin to make more of the
implicit expectations explicit.
Giving students time to discover what they already know, what they are learning,
and what they can learn from each other provides a scaffold for future learning.
Knowing what they are learning, the varied looks it can have, and what the desired
level of performance looks like gives students the information they need to assess
themselves as they learn—to keep themselves on track. Setting criteria with students
helps teachers know more about students’ prior knowledge as well as giving students
the language needed to self-monitor progress.
When criteria for success are set with learners, they can check their thinking and
performance and develop deeper understandings. Perhaps most significantly, students’
expectations of themselves and their beliefs in what they are capable of accomplishing
increase dramatically. Using criteria or creating rubrics that describe levels of quality
in relation to criteria, samples, or models results in more learning. Simply put, stu-
dents’ involvement in establishing and using performance criteria teaches them what
“good work” is, what it looks like, and how to produce and distinguish it.

Self-Assessing and Goal Setting

“Now that we are back from the beach and our first data collection, I want
you to review the criteria for collecting data that we created and think about
what you did that helped you to be successful and what you need to do

16 National Science Teachers Association

 CHAPTER 2

differently next time. Record your ideas. Here are two different procedures.
Choose one.”

Procedure #1:
Complete the following—

Here is a sample of the data I collected:

One thing that worked was …

One thing that didn’t work was …

One thing I might do differently next time is …

Procedure #2:
Take your copy of the criteria and use a highlighter pen to note the things
you did that worked and in another color one thing you need to do differently
next time. Select a sample of your collected data to show evidence of your
work. When you are finished, compare your ideas with your partner.

When we think about what we’ve done, we may come to understand it in a

different way. Self-assessment gives learners the opportunity to think about their
thinking and their learning—a process called metacognition. Michael Fullan puts it
this way: “An event is not an experience until you reflect upon it.” When students
and teachers self-assess, they confirm, consolidate, and integrate new knowledge.
Students, through the “assessment conversation,” help other students in the class
learn. When students are engaged in discussion with teachers about the learning
expected before any learning activity or task, the talk clarifies options, highlights
possible plans, and encourages sharing of information. Knowing what they are learning
and what it looks like gives students the information they need.
When conversations about learning take place in the group, learners can check
their thinking and performance, develop deeper understandings of their learning,
and become more strategic in their planning and monitoring. Researchers studying
the role of emotions and the brain say that experiences such as these prepare learners
to take the risks necessary for learning.

Revisiting Criteria

“As you were reviewing the criteria and thinking about collecting data, was
there something you did that worked that isn’t on the criteria sheet? What do
we need to add to the criteria for collecting data now that we’ve collected
some data about the green crabs?”

Everyday Assessment in the Science Classroom 17

CHAPTER 2

As students learn and assess, they define and redefine the criteria with teachers,
each time trying to make them more specific and accurate. When students are in-
volved, criteria become more specific as they learn more about high-quality work.
The source of ideas for criteria may come from anywhere—hands-on learning expe-
riences, classroom lessons, or experiences outside the classroom. The important thing
is that student learning is acknowledged and that the criteria continue to change as
students learn more. Like samples, criteria vary to reflect the learning journey and
destination. Consider the following purposes:
1. Do you want all students to do the same thing in the same way? For example, in
the case of collecting scientific evidence there is a quality standard that needs to
be adhered to. In this case the criteria need to describe that quality standard. It
would be inappropriate to aim for anything less.
2. Do you want students to show what they know in a variety of ways? For example,
students may choose a different way to express what was learned. It doesn’t
matter how the information is presented, but accuracy should not be sacrificed.
3. Do you want students to show individual progress over time? Does quality have
many different looks or just one? For example, if the focus of the criteria was
searching for and analyzing information, there may be an acceptable range. In
this case a rubric describing development may be helpful.
Criteria can become a rubric when different levels of development, quality, or
achievement over time are described. Rubrics that support student learning
• describe what students do at each level of quality or performance,
• are written in simple, clear language that students can understand,
• provide information about incremental steps students can take that will lead to
greater success, and
• focus on what the student needs to do to be more successful.
For example, the preferred description would state, “needs complete ideas, needs
more details” rather than “incomplete ideas, lack of detail.” The former description
helps students understand what to do next time. It is an example of assessment for
learning. The latter description merely judges the work and does not give students
information they can use to improve performance. It is an example of assessment of
learning. If the rubric you are considering using describes ways students can com-
pletely miss the learning destination, consider modifying it so it supports learning.

18 National Science Teachers Association

 CHAPTER 2

Providing Specific, Descriptive Feedback

“When you’re finished talking with your partner, please hand in your science
notebooks, open to some of the evidence you’ve collected, and attach your self-
assessment. I am going to review your work thus far and give you feedback.”

Specific, descriptive feedback is essential for learning (Hattie in press; Sadler

1989; Senge 1990; Shepard 2000; Stiggins 1996; Sylwester 1995). Descriptive feed-
back serves three goals: (1) it describes strengths on which a student can build; (2) it
articulates the manner in which performance falls short of desired criteria with an
eye to suggesting how that can be improved; and (3) it gives information that enables
the learner to adjust what he or she is doing in order to get better. Specific, descrip-
tive feedback that focuses on what was done successfully and points the way to
improvement has a positive effect on learning.
Evaluative feedback, particularly summary feedback, is very different. It tells the
learner how she or he has performed as compared to others or to some standard. Evalu-
ative feedback is highly reduced, often communicated using letters, numbers, checks,
or other symbols. It is encoded. This creates problems for students—particularly for
students who are struggling. Summary feedback does not address students’ needs or
the manner in which further growth and development can be realized and may affect
students’ motivation to learn. Students with poor marks are more likely to see them-
selves as failures. Students who see themselves as failures may be less motivated, and
therefore less likely to succeed as learners, than their peers with high marks.
If we evaluate too early, we limit descriptive feedback and risk interrupting the
learning. When we assess during the learning and evaluate at the end of the learning,
we give students time to practice and learn before we judge the evidence.

Setting Goals

“Before we go to the beach for today’s data collection, I want you to review
your notes from last time about what worked and what didn’t when we were
down there last collecting data. Given our experiences last time collecting
data, can we make a list of advice we would give other students doing similar
work? I’ll record your ideas. [Teacher writes down ideas for everyone to
see.] Before we go to the beach, write down one idea you are going to use to
make your data collection better this time. Write it at the top of your data
collection sheet. When we return, you will need to show evidence that you
tried to improve your data collection using the idea you have selected.”

When students work together to set criteria, self-assess, and reset criteria, they
come to understand the process of assessment and begin to learn the language of
assessment. Students gain a clear picture of what they need to learn and where they

Everyday Assessment in the Science Classroom 19

CHAPTER 2

are in relation to where they need to be, and they get an opportunity to begin to
identify possible next steps in their learning. Research indicates that closing in on a
goal triggers a part of the brain linked to motivation (e.g., Csikszentmihalyi 1990;
Pert 1997; Pinker 1997). Setting goals is a powerful way to focus students’ learning.
Students involved in self-assessment and goal setting in relation to criteria learn
more. When students self-assess, they gain insights about their learning that help
them learn. These insights help them monitor their learning and provide practice in
giving themselves descriptive feedback. When student self-assessments are shared
with teachers, teachers gain a better understanding about where students are in rela-
tion to where they need to be. When students share their thinking with teachers,
teachers can teach better.
Doing things more than once is essential for learning. In education we tend to
jump on one bandwagon and then jump on the next one to come along. We would
learn more as a profession if we just stuck to one thing and learned how to do it well.
The same is true for our students. It is when they do something the second and third
time that they learn what they know and what they need to know. Students need
practice time to learn. It is when students practice that they are able to take what they
are learning and apply it at deeper and deeper levels.

Collecting Evidence of Learning

“Congratulations. Our first data collection is over. The data have been
submitted. We will be doing this again in the spring. We have made our report
to the local environmental group. They were very impressed. They got some
important information. They also have some ideas for us to improve our data
collection next time. Also the new video camera will help us record more
evidence. Now that we are finished this first part of the project, I want you to
collect evidence that you have learned to show your parents at the end of
term. Remember our criteria. What evidence do you have that shows you
have met the criteria in a quality fashion? Who has some ideas about what
that evidence might include?”

Until recently, classroom assessment information—evidence of learning—re-

ferred exclusively to traditional paper and pencil tests. Such classroom tests can
measure some kinds of learning; they cannot measure all the knowledge, abilities,
and skills students are responsible for learning. Because learning is more than recita-
tion, assessment is more than recall tasks. Assessment evidence is more than tests; it
includes a range of physical evidence as well as observations and conversations with
learners about the learning. One way classroom assessment can be usefully under-
stood is as a particular kind of classroom research that adapts research methods from
anthropology and the human sciences. It is research into what students know and can
do, their strengths and areas of need.

20 National Science Teachers Association

 CHAPTER 2

This involves gathering evidence of learning from a variety of sources over time
and looking for patterns and trends. Such information gathering is one way to in-
crease the reliability and validity of classroom assessment findings. This process has
a history of use in the social sciences and is called triangulation. As students learn,
there are three ways to acquire evidence of learning:
1. Collecting products, including tests, assignments, students’ writings, projects,
notebooks, constructions, images, demonstrations, and video- and audiotapes.
2. Observing the process of learning, including observation notes regarding hands-
on, minds-on learning activities as well as learning journals and performances of
various kinds across all subject areas.
3. Communicating with students about their learning, including conferences, read-
ing written self-assessments, and interviews.
Collecting products, observing the process of learning, and communicating with
students will provide a considerable range of evidence over time. This not only in-
creases the potential for instructionally relevant insights into learning, it also pro-
vides for more accurate assessment of learning and therefore enhances the reliability
and validity of the evidence thus assembled. Collections of student work—evidence
of learning—may include anything that is relevant to learning and may vary from
student to student. Collecting the same information from all students may not be fair
and equitable because students show what they know in different ways. Allowing for
a range of evidence encourages students to represent what they know in a variety of
ways and gives teachers a way to fairly and more completely assess the learning.

Communicating Using Evidence of Learning

“Today your parents and other invited guests will be viewing the data we
collected concerning the invasion of the green crabs as well as looking at your
individual collections of evidence for this project and for science in general
this term. You have all worked hard. What do you want people to notice about
your work? What kind of feedback do you want to ask for from people who see
your work? Could you please fill in a ‘Please notice …’ card for your work?
Would you like to ask people to fill in a comment card telling you what they
liked and one suggestion they might have for next time?”

When we give students a chance to share their knowledge with each other and
with us, they learn and we learn. Celebrating our accomplishments by sharing our
work with others is part of the process of learning. The audience can be other people
in the class, other classes, parents and guardians, or community members. When the
learning is captured in print, on videotape, audiotape, or electronically, it becomes
concrete evidence of learning.

Everyday Assessment in the Science Classroom 21

CHAPTER 2

Acknowledging our accomplishments by sharing our work with others extends

the learning. When students are invited (and assisted) to collect and organize evi-
dence of learning for the purpose of sharing that with others, they learn about
• what they have learned,
• what they need to learn, and
• what kind of support may be available and necessary to them to extend their
learning.
It seems the presence of others influences what we attend to and what we know,
and forces us to step back and reflect—to think about and assess what we’re doing.
When students reveal evidence of their learning to an audience (e.g., peers, teachers,
parents), the students receive feedback and recognition from themselves and from
others. When learners talk about their learning to people whose opinion they value,
they learn more and are more likely to become resilient learners.
Students can use their collections of evidence of learning and portfolios to com-
municate in a variety of settings, including
• student-teacher conferences, in which students meet with teachers to present
evidence of their learning and the conversation is between teacher and student in
relation to the course goals;
• student-parent conferences that involve students and parents who meet at school
or at home to review collections of evidence of learning that show growth or
learning over time;
• student-parent-teacher conferences that complement written reports; and
• formal conference settings or exhibitions where students present evidence of
learning and answer questions from a panel of community members, parents,
and peers.
All of these ways of communicating have one thing in common: The student is
actively involved in presenting a range of evidence of learning. The teacher assists
by providing information regarding criteria and evidence of quality. Sometimes this
is done through using samples that reflect a continuum of learning over time. These
samples provide a reference point for conversation about student development and
achievement. Teachers use samples of work that represent levels of quality to show
parents where the student is in relation to the expected standard. This helps the teacher
respond to the question many parents ask: “How is my child doing compared to the
other students?” In a “verbal report card,” parents are involved participants in under-
standing the evidence and in the “reporting” on the child’s strengths, areas needing
improvement, and the setting of goals.

22 National Science Teachers Association

 CHAPTER 2

Continuing the Learning

“This is the last time this year we will be collecting data on the green crabs
on our beaches. Before we collect our data and prepare our report, I want to
revisit our criteria and identify previous work we have done that meets or
exceeds the criteria. Look at the criteria we set. Who has evidence from last
time that illustrates what quality looks like? What other kind of evidence
could prove you have met the criteria? You need to show what you’ve
learned. You do that when you provide evidence that you have met, to a high
degree of quality, each part of the criteria.”

When students are involved in the assessment process they are motivated to learn.
They learn how to think about their learning and how to self-assess—key aspects of
metacognition. Also, when students are involved in assessment they have opportuni-
ties to share their learning with others whose opinions they care about. An audience
gives purpose and creates a sense of responsibility for the learning. This increases
the authenticity of the task.
In summary, students need to be involved in the classroom assessment process
for the following reasons:
• Learners construct their own understandings. Therefore, learning how to learn—
becoming an independent, self-directed, lifelong learner—involves learning how
to assess and learning to use assessment information and insights to adjust learn-
ing behaviors and improve performance.
• Students’ sense of quality in performance and expectations of their own perfor-
mance are increased as a result of their engagement in the assessment process.
• Students can create more comprehensive collections of evidence to demonstrate
their learning because they know and can represent what they’ve learned in vari-
ous ways to serve various purposes.
• The validity and reliability of classroom assessment are increased when students
are involved in collecting evidence of learning. Also, the collections are likely to
be more complete and comprehensive than if teachers alone collect evidence of
learning.

Assessment for Learning in the Science Classroom

“Remember that the person doing the work is the one growing the dendrites,” writes
Pat Wolfe (2001, 187). The assessment for learning process described in this chapter
can be overwhelming if teachers believe it is their job to do all the work. Teachers
who understand assessment for learning are shifting more and more of the responsi-
bility for learning and assessment to students. This process can be a slow one as
students learn how to learn by becoming fully involved in the assessment process.
What does it look like when it happens? Students, in partnership with teachers,

Everyday Assessment in the Science Classroom 23

CHAPTER 2

• are able to articulate the learning destination,

• collect and refer to samples that show quality work,
• are able to describe what evidence of learning might look like,
• set criteria defining quality,
• have time to learn,
• receive and give themselves specific, descriptive feedback as they learn,
• debrief their learning with their peers and others,
• self-assess and set goals,
• revisit and reset the criteria as they learn more,
• collect evidence of their own learning, and
• present their evidence of learning to others and receive feedback.
Classroom teachers will know they are well on their way to assessment for learn-
ing when they can state, “I involve students in ongoing assessment for learning. I
summarize in my own words the learning that students are expected to accomplish. I
collect and review samples and models to show what the learning looks like for
students of a particular age range. I think about what kinds of evidence students
could produce to show they have learned what they needed to learn. I ensure that the
evidence of learning is valid and reliable by using the process of triangulation. I
collect evidence over time so emerging trends and patterns can be identified. I use
classroom assessment information to fine-tune instruction and the learning environ-
ment. Students are fully involved in the assessment process. They are working harder
than I am and learning more than ever before.”

References
Berger, P. L., and T. Luckman. 1967. The social construction of reality. Garden City, NY: Anchor
Books.
Busick, K. Conversation with author, Kaneohe, HI, 2001.
Csikszentmihalyi, M. 1990. Flow: The psychology of optimal experience. New York: Harper and
Row.
Davies, A. 2000. Making classroom assessment work. Merville, BC: Connections Publishing.
Delpit, L. 1995. Other people’s children: Cultural conflict in the classroom. New York: New Press.
Hattie, J. In press. The power of feedback for enhancing learning. Auckland, NZ: University of
Auckland.
Pert, C. 1997. Molecules of emotion: Why you feel the way you feel. New York: Simon and Schuster.
Pinker, S. 1997. How the mind works. New York: Harper Collins.
Sadler, R. 1989. Formative assessment and the design of instructional systems. Instructional Science
18: 119–44.

24 National Science Teachers Association

 CHAPTER 2

Senge, P. M. 1990. The fifth discipline: The art and practice of the learning organization. New York:
Doubleday.
Shepard, L. 2000. The role of assessment in a learning culture. Educational Researcher 29 (7): 4–14.
Stiggins R. 1996. Remarks made during keynote presentation, Assessment Training Institute Annual
Conference, Portland, OR.
Sylwester, R. 1995. A celebration of neurons: An educator’s guide to the brain. Alexandria, VA:
Association for Supervision and Curriculum Development.
Vygotsky, L. S. 1978. Mind in society: The development of higher psychological processes, eds. M.
Cole, V. John-Steiner, S. Scribner, and E. Souberman. Cambridge, MA: Harvard University Press.
Wolfe, P. 2001. Brain matters: Translating research into classroom practice. Alexandria, VA: Asso-
ciation for Supervision and Curriculum Development.

Everyday Assessment in the Science Classroom 25

 CHAPTER 3

Examining Students’ Work

Cary I. Sneider

Cary I. Sneider is vice president for programs at the Museum of Science in Boston, where he
oversees live programming for approximately 1.6 million visitors each year. In addition to his
work at the museum, he is currently chair of the Committee on Science Education K–12 at the
Center for Education, National Research Council. Cary has taught science in grades 5–12 in
Massachusetts, Maine, California, Costa Rica, and Micronesia. He received a teaching certifi-
cate and Ph.D. in science education from the University of California at Berkeley. His research
interests have focused on helping students unravel their misconceptions in science and on new
ways to link science centers and schools to promote student inquiry. His publications include K–
12 teachers’ guides, articles about the instructional uses of computers, and research studies on
how children learn science. He helped to develop the National Science Education Standards
(NRC 1996) and contributed to Designing Professional Development for Teachers of Mathemat-
ics and Science (Loucks-Horsley et al. 1998). In 1997 he received the NSTA Distinguished Infor-
mal Science Education award.

I have been fortunate to teach science in a wide variety of situations, and I’ve learned
a great deal from each of them. In this chapter I’ll share some of those experi-
ences, with particular reference to what I’ve learned from examining students’ work.
I’ll begin with my first teaching experience, as a tutor at an Upward Bound summer
program in Cambridge, Massachusetts.

Learning to Teach
My assignment that summer was teaching high school math to a group of about ten
sophomores. I chose a topic that the students would be likely to encounter in the
following school year—trigonometric functions. I approached the teaching of math-
ematics through induction. I wanted my students to “discover” the trigonometric
functions by measuring the edges of right triangles and taking ratios. So, I assigned
the students the task of constructing several similar right triangles and dividing the
length of the opposite side by the adjacent side (the tangent). I expected the answers
to vary somewhat due to inaccuracies in measurement, but for the ratios to be ap-
proximately the same. Then I’d assign them to construct several more similar right
triangles, but with a different angle, and so on, until they had constructed a rough
table of tangents.
The lesson was planned beautifully. I expected the students to go on to learn
about the sine and cosine functions, then to apply the tables they constructed to word
problems. So, when the students encountered these functions in school, they would
understand how the tables of trigonometric functions were derived and would soon
master the subject.
That was when I learned that lessons never go as planned. The students were
able to construct triangles and take the ratios of the sides, but their answers for sev-
eral similar triangles were wildly different! I could see the problem when I walked
around the room and looked carefully at what they had done. They were all using

27
CHAPTER 3

their protractors in different ways. Apparently, the students had not been taught (or
did not remember) how to use a protractor. I changed the objective of the lesson from
teaching about tangents to teaching about protractors; for many of the students, a
lesson about how angles were measured in degrees was also necessary. In an ap-
proach that many science teachers know all too well, I had to “back up” to where my
students were before I could progress to the original learning objective.
We eventually did get around to creating a rough table of tangents for 10° inter-
vals, but we never would have arrived at that point if I had not been able to see how
they were holding their protractors. I had previously thought hands-on activities rep-
resented an important way for students to learn concepts and skills. While that may
be true, the valuable insight I gained was that hands-on activities also allowed me to
see how my students were thinking—and that was absolutely essential if I was to
adjust my teaching objectives to suit their needs. In brief, I learned two important
things about assessment from my first teaching experience:

Insight #1 We need to be prepared to modify not only our approach, but also
our teaching objectives in response to feedback on student thinking.

Insight #2 Hands-on activities provide opportunities for us to see what our

students are thinking.

How Directive Should We Be?

My third insight about examining student work came during my first full-time posi-
tion in rural Maine. My high school teaching assignment included two sections of
physics for seniors and three sections of general science for freshmen, as well as
being the adviser to the student council and band. I quickly learned that as band
director I needed to be in charge. The drum section simply could not be allowed to
move faster than the woodwinds. But as a science teacher, taking charge was often
counterproductive.
I was on fairly firm ground with the senior physics classes since I had the oppor-
tunity to use the Harvard Project Physics program. When I lectured and stopped to
ask questions I could find out what one or two students were thinking, but it was not
possible to find out what the rest of the students were thinking with Q&A alone.
However, there were many lab activities that enabled me to engage my students’
thinking so that I could work with them in small groups. Since I was not pressured to
finish the text, I assigned many of the labs twice. The first time the students com-
pleted the lab as prescribed. The second time involved an original application of the
concepts and tools.
For instance, to study motion we used Polaroid cameras and stroboscopes as
well as some old ticker-tape timers that I dug out of the stockroom. In round one the
students experimented with dynamics carts to measure and graph uniform and accel-
erated motion. In round two they selected some other motion to study. Balls thrown

28 National Science Teachers Association

 CHAPTER 3

into the air, model rockets, and a good strong punch stand out in my mind as the
motions they chose to study. Perhaps the most interesting was a windup baby doll
that crawled along the floor. To study the motion of the doll, the students used a
ticker-tape timer, which imprints a dot on a strip of paper every sixtieth of a second.
The students taped one end of the paper strip to the doll and turned on the timer. On
the strip of paper they saw a complex cyclic pattern of dots, in some places getting
further apart (acceleration) and in other places closer together (deceleration). Listen-
ing to the students interpret a graph of these patterns was far more revealing of their
understanding of motion than their work with the dynamics carts alone. I realized
that by allowing the students some room for creativity, they became more deeply
engaged in the project and it was easier for me to evaluate what they had learned and
what they still found confusing (Sneider 1972).
There is a paradox here that’s worth pointing out. On the one hand, as teachers
we are encouraged to treat each student fairly in evaluating his or her work. That
means we should ask all students the same questions and evaluate their performances
by the same set of rules. That is why teachers give all of their students the same test
at the end of a unit, and it is why school districts and states use standardized tests to
measure success districtwide. On the other hand, such tests leave virtually no room
for creativity, so we may not be seeing what students can do when motivated by their
own curiosity. The paradox is resolved if we use creative problem-solving for the
purpose of providing direct feedback to individuals and teams and modify instruc-
tion accordingly. As discussed in other chapters of this book, records of creative
student work can be evaluated with the use of a rubric and therefore contribute to
grades, but using such data for accountability of schools and teachers may be prob-
lematic. The main point here is that creative tasks provide invaluable opportunities
to examine student thinking, and they should not be neglected in an effort to use
assessment data for too many different purposes.

Insight #3 We should provide activities for students to apply what they have
learned in new situations that allow some free choice or creativity.

The Importance of Listening

My major challenge during that first year of full-time teaching was the general sci-
ence classes for freshmen. Since I was teaching with a provisional credential, I had
virtually no experience in the instructional materials that were just then being devel-
oped with National Science Foundation support. And the textbooks that were avail-
able were hopelessly out of date. After floundering for a while, I settled on a method
of teaching that provided even more insight into how my students’ thinking evolved
over time. It involved laboratory and discussion components in which I played the
role of listener and coach rather than leader.
The labs all started with an open-ended challenge. “Find out all you can about
falling objects, and make a list of what you discover.” Or, “Explore shadows and

Everyday Assessment in the Science Classroom 29

CHAPTER 3

make a list of what you find out.” The students worked in lab groups of three or four
students per team. I gathered materials and suggested ways for the students to ex-
plore these phenomena and helped them articulate their findings in writing, using
large sheets of butcher paper. Then, the students sat in a large circle and discussed
each other’s lists, one statement at a time. I explained that the purpose of the discus-
sion was to (1) understand the phenomenon that was being described; (2) determine
if it was consistent with what they had experienced; (3) look at additional evidence if
necessary; (4) modify the written statement if needed to be accurate; and (5) vote on
whether or not they thought the statement was true as finally worded. If a large
majority of the students agreed, we would call it a “natural law.” We called these
discussions “scientific conventions.” Various students facilitated, so that I could sit
on the side taking notes and acting as coach. The discussions were absolutely fasci-
nating and sometimes quite surprising. For example, one topic the students investi-
gated was “wave motion,” including waves traveling in springs and ripple tanks. In
the following transcription of a tape recording, the students are trying to decide among
several statements to determine which best explains what occurs when a wave hits a
barrier:
“Just don’t tell me the wave is absorbed because I don’t believe it.”
“It fades away.”
“Fades away, fades away, that’s no excuse.”
“What happens when your voice dies out?”
“Your voice dies out?”
“That’s what I want to know.”
“It dampens. It goes away.”
“Where does it go?”
“It goes to voice heaven, right next to the wave heaven.”
Some of the students in the class recognized that the energy from the wave must
go somewhere, while others were content with the statement that it just disappeared.
Eventually, the class voted in favor of two statements that they were willing to call
“laws.” The first, accepted almost unanimously, was “Waves pass through each other.”
The students had seen evidence of waves in springs and in water passing right through
each other, so there was no question that it was true. The second statement, which
commanded only a two-thirds vote, was “When a wave hits a barrier, part of it will
bounce back, part of it will be absorbed by the barrier, and part of it may continue
on.” Although the statement made logical sense, fewer students were willing to sup-
port it because there was no direct evidence that it was always true.

30 National Science Teachers Association

 CHAPTER 3

After one of these discussion sessions it was not difficult to see what had to be
done next. Often I’d present a mini-lecture to resolve conflicts, or offer an additional
lab activity to test one of the students’ ideas. Unless I was worried that the students
would leave a class with more misconceptions than when they came, I refrained
from pronouncing their statements to be true or false. I wanted the students to realize
that scientific laws are not clear statements in the teacher’s head or on the pages of a
book but are hard-won discoveries gained from experimentation and logical argu-
ment (Sneider 1971).
The insight I learned that year was that it is extremely valuable for the teacher to
sit on the sidelines as students discuss their interpretations of the phenomena. In
these “minds-on” activities, the students seemed less inhibited in talking with each
other than in answering the teacher’s questions, and I could really focus on what they
were saying, rather than planning what to say next. Later I developed more detailed
content objectives, but I always tried to make time for these open-ended discussions
about the phenomena that the students had been studying, taking myself out of the
center of the discussion whenever possible. To summarize:

Insight #4 Structuring class discussions so students address each other

rather than the teacher encourages students to express their own ideas and
allows the teacher an opportunity to make observations and take notes on
their thinking.

Culture Is Critical
In subsequent years I had the opportunity to attend graduate school at the University
of California at Berkeley to earn a teaching credential and to learn about some of the
excellent new curriculum materials that were just becoming available. Then, when I
returned to teaching, I had the chance to integrate what I learned about examining
student work with new curriculum materials in such diverse locations as San Jose,
Costa Rica; the island of Pohnpei in Micronesia; and Coalinga, California.
An important insight that I gained from these experiences is that approaches devel-
oped within one cultural context do not necessarily translate to another. One of the
clearest examples was my work in Micronesia, since the culture there was so different
from life in the USA. The purpose of that program was to prepare high school students
for colleges in the United States. But teaching classes in the American style was diffi-
cult since Micronesian teenagers are very shy. It was almost impossible to get a roaring
class discussion going, even among college-bound seniors. I understood why this was
so when I found that young people were actively discouraged from taking part in dis-
cussions at family gatherings. The right to speak in public was reserved for men over
age 30. While hands-on activities were valuable in engaging the students’ attention, it
was difficult to understand the conversations among small groups of students when
they preferred to speak their native language rather than English.

Everyday Assessment in the Science Classroom 31

CHAPTER 3

In my view, the most effective lessons were those in which the students created
maps of their island, researched native crafts, and wrote stories and drew pictures
about the rich history of their island. Especially compelling was a story about the
origin of the island and its people, and the overall structure of the Pohnpean uni-
verse. Some of these were formatted as newsletters, translated from English into
Pohnpean for the students to share with their parents and grandparents. Naturally it
was not possible for us to check the students’ Pohnpean translations, so we asked one
of the parents for assistance. He made quite a few corrections and explained to us
that in many cases the words used by young people to talk with each other were
different from words used to address elders. Many of the corrections took into ac-
count who was talking to whom. By encouraging our students to undertake tasks that
were meaningful for them, we learned enough about their culture to shift our meth-
ods and teaching objectives so as to create a bridge between the two cultures. The
following statement sums up several teaching experiences that I’ve had where cul-
tural differences have been an important factor.

Insight #5 The more we learn about our students’ cultural backgrounds, the
more successful we are likely to be in adjusting our teaching methods to meet
their needs.

Study Group to Examine Student Work

The next chapter in my learning about how to really listen to students came when I
returned to UC Berkeley for a graduate degree program. I happened to attend a semi-
nar by Yossi Nussbaum, a visiting educator from Israel. He presented his research,
conducted in collaboration with Jerome Novak from Cornell, that showed that chil-
dren in the elementary–middle school range, in both Israel and New York, had very
unusual notions about the spherical Earth concept. When asked about the shape of
the Earth, nearly all children would say it is “round.” However, on further probing,
Dr. Nussbaum found a wide variety of different ideas. One child thought the Earth
was round like an island. She explained that Columbus proved the Earth is round by
sailing out of a port in Spain, going around the island, and back to the same port.
Another realized that the Earth is round like a ball, but thought people lived just on
top of the ball. When the sun went “under” the Earth, it sometimes caused volcanoes
(Nussbaum and Novak 1976).
At the time I heard about Dr. Nussbaum’s research I was teaching astronomy to a
group of teachers. When I told them about these results they were incredulous. They
explained to me that students learn the Earth is a sphere in the first or second grade, and
they simply did not believe that any of their students would have difficulty with the
concept. Three of the teachers agreed to work with me to conduct studies in their own
classrooms. We formed a study group and conducted more than two hundred indi-
vidual interviews and developed a paper-and-pencil questionnaire. Our study included
children in grades four through nine. We examined the interview records and the writ-

32 National Science Teachers Association

 CHAPTER 3

ten questionnaires very closely, both individually and as a group. Based on their re-
sponses to a series of open-ended questions, we classified the children’s conceptions
into several categories of ideas about the Earth’s shape and gravity.
Sure enough, our results from the San Francisco Bay Area were consistent with
Dr. Nussbaum’s findings in New York and Israel. While there was deeper under-
standing of the spherical Earth concept among older students than younger students,
there was a significant number of students at all age levels who failed to understand
that we do indeed live on the outside surface of a ball and that objects fall “down”
toward the center of the Earth (Sneider and Pulos 1983; Sneider et al. 1986).
The questionnaire had four questions. Question one asked why the Earth looks flat
even though it is supposed to be round. Possible choices were taken from Nussbaum’s
paper and pilot studies. The second question asked children to imagine the Earth is
made of glass so they could look right through it and to indicate what direction they
would look to see people in far-off countries like China and India. The third question
showed a drawing of Earth with enlarged humans standing all around it, each holding
a rock. The students were asked to draw what would happen to the rock when each
person dropped it and to explain their answers. The last question showed a hole drilled
through the entire Earth, from pole to pole. The students were asked to draw the path of
a rock dropped into the hole, and then to explain their answers.
While the results varied tremendously from class to class, a typical fifth-grade
class had only five or six students who could answer all questions the way a scientist
would. For Q1 these students would say that the “Earth is round like a ball but looks
flat because we see only a small part of the ball.” For Q2 they would show people
looking downward, through the transparent Earth, to see people in far-off countries.
For Q3 they would draw the rocks falling down to the surface of Earth and stopping
at the people’s feet. And for Q4 they would show the rock falling into the hole. Some
would show the rock falling through the middle, then back again, finally settling in
the middle. Others showed the rock simply falling to the middle and stopping.
(Newton’s laws of motion lead to the prediction that the rock will pass through the
middle, then fall back and forth, settling in the middle. Most of the students who
drew the rock falling to the center and stopping explained their answer by saying
that’s where the “center of gravity” is located.)
What was striking about the students whose answers did not agree with the sci-
entists’ was that they nonetheless reflected a consistent view of the world. For ex-
ample, a student who indicated that Earth is like a flat, round island on the first
question would indicate that a person would need to look parallel to the ground to
see people in far-off countries. The student would then show rocks falling down, to
the bottom of the page, in the next two questions. These answers are perfectly con-
sistent with the conception of the Earth as a round island. Similarly, a student who
indicated that the “Earth is round like a ball but looks flat because we see only a
small part of the ball” might believe that people live just on top of the ball. That
conception could be inferred by noting that they would look parallel to the ground to

Everyday Assessment in the Science Classroom 33

CHAPTER 3

see people in far-off countries, and indicate rocks falling down to the bottom of the
page in response to questions three and four.
At about that time I read a paper by Driver and Easley (1978) that recommended
calling such ideas “alternative frameworks” rather than “misconceptions.” Their per-
spective was that children’s ideas were not wrong, even though their answers dif-
fered from those that would be given by scientists. When these students saw a globe
in the classroom and were told by their teacher and textbook that the Earth is round,
they created a personal theory of the world that integrated this new information with
the flat Earth of their everyday experience. Some of their ideas that we uncovered
through the interviews were marvelously inventive. Several of the children believed
that the Earth is indeed shaped like a ball but that people live in the “flat part in the
middle,” with a hemispherical sky above and a hemispherical solid Earth below.
That idea corresponded with illustrations in certain textbooks that illustrated how
the path of the sun varied during the year. That same concept was reported by Dr.
Nussbaum as a result of his interviews of children in Israel and New York.
Eventually, working with many teachers and colleagues at the Lawrence Hall of
Science, we developed activities that enabled students to compare and contrast their
ideas and to begin to recognize some of their own alternative frameworks. Using the
questionnaire as a pre- and posttest revealed that most students from fourth through
ninth grade who were involved in those activities gradually developed ideas that
were more similar to scientists (Sneider and Ohadi 1998).
I gained two insights from those very detailed observations and long-term efforts
to develop lessons that would change students’ ideas about the Earth’s shape and grav-
ity. The first was that interviewing students and working closely with other teachers to
examine the results of the interviews and written questionnaires is a tremendously
rewarding activity. As individual teachers, we had never fully analyzed what our stu-
dents were thinking. While it took quite a few evenings and weekends, we all agreed
that forming a study group was worth the effort in terms of our own professional devel-
opment and the insights it gave us into how our students think. In brief:

Insight #6 Study groups are an excellent way to compare interpretations of

students’ work.

Insight #7 Children are like scientists. They formulate theories to synthesize

what they see and what they learn from others so that they have a consistent
worldview. As teachers, our job is to help them articulate their current
understanding and move to a more fruitful way of thinking.

Theories Are Not Easy to Change

The insight about how children construct theories of their world was reinforced in a
very fortunate collaboration with two other astronomy educators, Varda Bar from
Israel and Nathalie Martimbeau from Canada (Bar et al. 1994). During a casual

34 National Science Teachers Association

 CHAPTER 3

conversation Dr. Bar mentioned a curious alternative conception about gravity that
she found to be very common at the upper–elementary school level in Israel: that
there is no gravity in space because there is no air up there (Bar et al. 1994). To
support this belief many children refer to astronauts that they have seen on television
who are floating around in space. Others refer to the smaller amount of gravity on
the Moon, again assuming that is because there is no air on the Moon. Similar find-
ings were reported by other investigators (Riggiero et al. 1985; Gunstone and White
1980; Watts 1982).
We designed a series of activities that could take place during a single class
period that would help students articulate their current ideas about gravity, and expe-
rience logical conflicts that we hoped would lead them to abandon the idea that
gravity requires air. We worked with two sixth-grade classes that had just finished a
unit on astronomy. We interviewed a sample of ten students before and after present-
ing the unit, and we gave paper-and-pencil pre- and posttests to all of the students.
Here is what two of the students said about gravity prior to the lesson about gravity:
Raymond (sixth grade, age 11)
Why does this pencil fall when you drop it?
“Gravity is pulling it down.”
Why does a cloud not fall?
“It is very high in the atmosphere and it is not touching the gravity.”
Why do the Moon and Sun not fall?
“The Moon is very high, past the Earth’s atmosphere. The gravity stops at the
top of the atmosphere. The same for the Sun.”
What is gravity?
“It is the force that keeps us down. It is in the air.”

Emily (sixth grade, age 11)

Why does this pencil fall when you drop it?
“Because of the gravity pulling it down.”
Why does a cloud not fall?
“They are higher in the sky. They are made of condensation of water. They do
not fall because gravity does not reach them.”
Why do the Moon and Sun not fall?
“They are in space. There is no gravity there.”

Everyday Assessment in the Science Classroom 35

CHAPTER 3

What is gravity?
“It is a pull, a thing that keeps … pushes things down on Earth or a planet. Air is
around the Earth and keeps gravity. The atmosphere keeps the gravity inside.”
We started class by having the students roll balls off tables, watching the curved
path followed by the ball as it left the table. They saw that by pushing the ball faster and
faster they could make it go further from the table before it hit the floor. The students
were able to articulate that the path of the ball was due to its forward motion and the
downward pull of gravity. We next showed a series of transparencies in which a base-
ball player hits a ball. The students explained that the ball would follow the same
curved path as the balls that they rolled off of tables earlier in the class (though it might
go further upward if it’s a fly ball). The harder the ball is hit, the further it goes.
The baseball player in the transparency is then replaced by a cannon that is fired
parallel to the Earth. As it is fired with greater and greater force, we asked the stu-
dents what would happen to the cannon ball. Eventually the students saw that the
cannon ball would circle the Earth completely, and go into orbit. We replaced the
cannon ball with a space shuttle, and then the Moon, allowing the students to discuss
these situations with each other, and providing coaching whenever we saw a good
opportunity to do so. Near the end of the class we asked the students about gravity
and air. If the space shuttle continues to circle the Earth as a result of gravity, then
how could gravity depend on air? The same is true of the Moon. After the class, we
gave the students a written questionnaire. We asked the students to write answers
that indicated their personal beliefs, not just what other people said. The key ques-
tion was: “Does gravity act in space where there is no air? Why or why not?” Here
are the results of the students’ answers.

Table 1. Responses to the Question “Does gravity act in space where

there is no air? (Why or why not?)”

Does gravity act in space where there is no air? Pretest Posttest

Yes 13 23
No 31 10
Not Much 0 3
Only Near Planets 0 4
No Answer, or I Don’t Know 4 8
Total 48 48

Note: The pretest was given at the beginning of the class period and the posttest was given one
week after the class period. The subjects are 48 students in two groups of sixth graders.

Interviews with the same ten students whom we interviewed prior to the lesson
confirmed the results of the questionnaire. On the posttest, there were more than

36 National Science Teachers Association

 CHAPTER 3

twice as many “yes” answers as “no” answers. We have some confidence that stu-
dents were giving us their honest opinions because most could give convincing rea-
sons why gravity acts through space—for example, “Since the Earth keeps the Moon
in orbit, there must be gravity in space.” Another said, “This is why the planets go
around the Sun.”
Nonetheless, some students did not change their answers as a result of the les-
son. They understood the first part of the activity, which concerned the rolling balls,
but did not generalize it to include the notion of satellites around the Earth. However,
even achieving this goal is a step in the right direction since, as we learn from the
history of science, changing ideas about how things move here on Earth is the first
step in learning about how things move in space.
An insight that we can draw from this experience is that it is very difficult for
our students to change deep-seated ideas about the world that they have learned
through a lifetime of experiences and that explain things in a way that is perfectly
satisfactory to them. The students’ ideas “work” for them, but they are quite differ-
ent from what scientists have in mind when they speak of subjects such as orbits
and weightlessness. For this reason we were not discouraged when we found that
only some of our students were able to come to a more adequate understanding of
how moons and space satellites stay in their orbits. For factual information or sim-
pler concepts we might expect closer to 100 percent success; but for the develop-
ment of fundamental concepts like gravity, we have to be more patient and expect
that more such experiences, over a longer period of time, will be necessary to
reach all of our students. To summarize:

Insight #8 Students are not likely to give up their deep-seated concepts of the
world easily. Most will require a variety of learning experiences.

Insights and Outcomes

One of my most enriching experiences is the work that I have done for more than 15
years on a project at the Lawrence Hall of Science, UC Berkeley, entitled Great
Explorations in Mathematics and Science (GEMS). In contrast to my science teach-
ing experiences, in which I needed to have new lessons each day, working on the
GEMS curriculum team gave me opportunities to repeat the same lessons several
times with different groups of students and to observe the same science activities
taught by different teachers. Several of us on the team compared notes, and together
we would decide which changes should be made.
Among the many GEMS guides that I worked on, one of the most rewarding was
about how to assess student learning. GEMS guides tend to focus on broad concepts,
science and math thinking skills, and unifying ideas—and these are difficult to mea-
sure. The entire GEMS team collaborated on the development of a teacher’s guide to
the assessment of hands-on, minds-on activities. Our guide was entitled Insights and
Outcomes: Assessments for Great Explorations in Math and Science (Barber et al.

Everyday Assessment in the Science Classroom 37

CHAPTER 3

1995). In the book we used 17 detailed cases studies to illustrate a variety of assess-
ment methods, along with suggested rubrics for carefully examining and assigning
values to students’ work. From this work I learned the following:

Insight #9 There is a wide variety of ways to assess inquiry-based learning

so as to measure progress against goals, and to inform next steps in
instruction.

Assessment or Instruction?
Now that I’m a museum administrator I tend to use the word “visitor” rather than
“student” since we have a much greater diversity of learners than in a classroom.
Learners for a given program or exhibit can range from five- or six-year-olds to
teenagers, parents, and grandparents! While we attempt to design layers of meaning
into our programs so there is something for everyone, we depend on a wide variety
of techniques to examine our visitors’ thinking. Here is one technique that works
particularly well, even with a large group of museum visitors.
Ken Pauley, one of the museum’s seasoned presenters of live programs, intro-
duces a large snake and asks, “How much of this snake is tail, and how much is
body?” To encourage visitors to respond, he explains that some people think it’s all
body; others say that it’s all tail. During his presentation he allows time for just about
everyone to respond to the question and leads them to discover knowledge that they
already have to answer the question. (“Think of a dog or cat. What can we observe
where the tail joins the body? Can we find similar structures on a snake? Would
some volunteers care to come up and help me look for those structures?”) Similarly,
Jessica Swanson presents a mystery skull to the visitors and uses a series of ques-
tions to help them figure out what kind of animal it came from, based on such
evidence as the positions of the eyes and kinds of teeth. The reader will probably
recognize this teaching style as the Socratic questioning approach. Socrates may
have been the first advocate in history for close examination of student work. What I
love about the Socratic method is that it’s impossible to distinguish between assess-
ment and instruction. The answer to each question leads to the next question in a
seamless flow of ideas. That reminds me of yet another insight:

Insight #10 Assessment can and should be an essential and inseparable

aspect of instruction.

Examining Cases
It seems appropriate to close with a recent project that is just coming to fruition. It is
a video library created by WestEd, the Museum of Science, and WGBH-TV, with
National Science Foundation support. The series of videos is entitled Teachers as
Learners: Professional Development in Science and Mathematics. The purpose of
the project is to provide vivid examples of a variety of strategies for designing pro-

38 National Science Teachers Association

 CHAPTER 3

fessional development programs for teachers of science and mathematics. Working

with the team that developed this video library greatly expanded my repertoire of
professional development strategies. One of these was a case-study approach to closely
examining student work. Used in this sense, a case study is an example of student
work selected by a team of educators for the purpose of bringing to light key aspects
of the teaching-learning process.
One tape in the series shows a group of about 20 teachers in a workshop orga-
nized by the Arizona Tribal Coalition. In the video we see a facilitator lead the group
in applying a scoring rubric to the work of students for whom English is a second
language. As various teachers suggest interpretations for what the students mean by
their written answers, it becomes clear that they are learning a lot from each other. It
is also clear that there is a great advantage if teachers understand the students’ native
language, so that errors in English are not interpreted as errors in scientific thinking.
I gained two insights from that experience:

Insight #11 Examining prepared cases is a very powerful approach to

acquiring skills in interpreting student work.

Insight #12 In the case of English language learners, it is helpful if we know

the students’ native language, so that errors in English are not interpreted as
errors in scientific thinking.

Our Role as Teachers

The previous example is an appropriate way to end the chapter as it turns the spotlight
from examining students’ work to examining teachers’ work. And the kind of teachers’
work that is illuminated is the subject of this chapter—understanding what our stu-
dents are thinking, so that we can help them develop more powerful ways of viewing
the world. While it is clearly important for us to understand the concepts and skills we
wish to teach, without a way of stepping inside our students’ shoes it is impossible for
us to communicate those concepts and skills in a deep and meaningful way.

References
Bar, V., B. Zinn, R. Goldmuntz, and C. Sneider. 1994. Children’s concepts about weight and free fall.
Science Education 78(3): 149–69.
Barber, J., L. Bergman, J. Goodman, K. Hosoume, L. Lipner, C. Sneider, and L.Tucker. 1995. Insights
and outcomes: Assessments for great explorations in math and science. Berkeley: Lawrence Hall
of Science, University of California at Berkeley.
Driver, R., and J. Easley. 1978. Pupils and paradigms: A review of literature related to concept devel-
opment in adolescent science students. Studies in Science Education 5: 61–84.
Gunstone, R. F., and R. T. White. 1980. Understanding gravity. Science Education 65(3): 294–99.
Loucks-Horsley, S., P. Hewson, N. Love, and K. Stiles, with H. Dyasi, S. Friel, J. Mumme, C. Sneider,
and K. Worth. 1998. Designing professional development for teachers of science and mathemat-
ics. Thousand Oaks, CA: Corwin Press.

Everyday Assessment in the Science Classroom 39

CHAPTER 3

National Research Council (NRC). 1996. National science education standards. Washington, DC:
National Academy Press.
Nussbaum, Y., and J. Novak. 1976. An assessment of children’s concepts of the Earth utilizing struc-
tured interviews. Science Education 60(4): 535–50.
Riggiero, S., A. Cartelli, F. Dupre, and M.Vincentini. 1985. Weight, gravity and air pressure: Mental
representations by Italian middle school pupils. European Journal of Science Education 7(12):
181–94.
Sneider, C. 1971. A laboratory and discussion approach to high school science teaching. The Physics
Teacher 9(1): 2–24.
———. 1972. A different discovery approach. The Physics Teacher 10(6): 327–29.
Sneider, C. I., and M. Ohadi. 1998. Unraveling students’ misconceptions about the Earth’s shape and
gravity. Science Education 82: 265–84.
Sneider, C., and S. Pulos. 1983. Children’s cosmologies: Understanding the Earth’s shape and grav-
ity. Science Education 67(2): 205–21.
Sneider, C., S. Pulos, E. Freenor, J. Porter, and B. Templeton. 1986. Understanding the Earth’s shape
and gravity. Learning ’86 14(6): 43–47.
Teachers as learners: Professional development in science and mathematics. Video Library. In press.
Thousand Oaks, CA: Corwin Press.
Watts, D. M. 1982. Gravity—Don’t take it for granted! Physics Education 17(4): 116–21.

Additional Resources
The following books provide additional information, ideas, and activities for examining student work.
Atkin, J. M., P. Black, and J. Coffey, eds. 2001. Classroom assessment and the national science
education standards. Washington, DC: National Academy Press.
Bar, V., C. Sneider, and N. Martimbeau. 1997. Is there gravity in space? Science and Children 34(7):
38–43.
Brown, J. H., and R. J. Shavelson. 1996. Assessing hands-on science: A teacher’s guide to perfor-
mance assessment. Thousand Oaks, CA: Corwin Press.
Doris, E. 1991. Doing what scientists do: Children learn to investigate their world. Portsmouth, NH:
Heinemann.
Hein, G., and S. Price. 1994. Active assessment for active science. Portsmouth, NH: Heinemann.
Stead, K., and R. Osborne. 1981. What is gravity? Some children’s ideas. New Zealand Science Teacher
30: 5–12.
Wolf, D., J. Bixby, J. Glenn III, and H. Gardner. 1992. To use their minds well: Investigating new
forms of student assessment. Review of Research in Education 17: 31–74.

40 National Science Teachers Association

 CHAPTER 4

Assessment of Inquiry
Richard A. Duschl

Richard Duschl, professor of science education at King’s College London, received his under-
graduate and doctorate degrees in Earth science education and science education, respectively,
from the University of Maryland–College Park. A former high school teacher, he served for 10
years as the editor of Science Education. His present research focuses on establishing science
assessment learning environments that focus on students’ argumentation processes. This research
is an extension of the National Science Foundation–funded Project SEPIA (Science Education
through Portfolio Instruction and Assessment) research. Project SEPIA investigated the dynamic
structures of implementing full-inquiry units. Each unit is carefully written and researched to
involve students in solving a problem while developing and evaluating a scientific claim (e.g.,
causal explanation, model, argument). A focus of instruction is the epistemic criteria used to
evaluate knowledge claims.

“Of one thing I am convinced. I have never seen anybody improve in the art
and techniques of inquiry by any means other than engaging in inquiry.”
—Jerome Bruner, 1961

T eaching science through inquiry has been a distinguishing feature of innovative

science education programs since the 1960s science curriculum reform move-
ment. Then, as now, however, the pedagogical challenges and material costs of teach-
ing science as inquiry or science through inquiry minimized the number of classrooms
in which students learn science and learn about science through inquiry. Student-cen-
tered inquiry science teaching is complex and challenging. Assessing inquiry only
adds to the complexity and to the challenge. Part of the implementation problems teachers
face is deciding what type of instruction promotes inquiry teaching and what type of
instructional sequences and lessons count as inquiry curriculum.
Inquiry means many things to many different individuals. In fact, Table 2.6 in
Inquiry and the National Science Education Standards (NRC 2000) bears the title
“Essential Features of Classroom Inquiry and Their Variations.” Some advocate the
use of inquiry methods to teach concepts while others propose their use to teach pro-
cess skills. Still others associate inquiry learning with the completion of laboratory
investigations. But when we reflect on the five essential features of inquiry and on the
two National Science Education Standards K–12 inquiry goals (NRC 1996), namely,
“doing scientific inquiry” and “understanding scientific inquiry,” the basic elements of
scientific inquiry become clearer. The focus on scientific inquiry needs to be on attain-
ment of evidence and how it is used to generate and justify explanations.
Since the 1960s, research and scholarly findings in cognitive and social psychol-
ogy (Bransford, Brown, and Cocking 2000; Pellegrino, Chudowsky, and Glaser 2001),
in history and philosophy of science (Magnani and Nersessian 1999), and in educa-
tional research (Linn and Hsi 2000; Millar, Leach, and Osborne 2001; Minstrell and
van Zee 2000) have developed new models of thinking and reasoning scientifically,
new models of science as a way of knowing, and new models of learning and teach-

41
CHAPTER 4

ing science. When we synthesize this research, the messages we receive are the fol-
lowing:
1. The incorporation and assessment of scientific inquiry in educational contexts
need to focus on three integrated domains:
• the conceptual structures and cognitive processes used when reasoning scien-
tifically,
• the epistemic frameworks used when developing and evaluating scientific
knowledge, and
• the social processes and forums that shape how knowledge is communicated,
represented, argued, and debated.
2. The conditions for science inquiry learning and assessment improve through the
establishment of
• learning environments that promote student-centered learning,
• instructional sequences that promote integrating science learning across each
of the three domains in item 1, above,
• activities and tasks that make students’ thinking visible in each of the three
domains, and
• teacher-designed assessment practices that monitor learning and provide feed-
back on thinking and learning in each of the three domains.
Research informs us that students come to our classrooms with a diversity of
beliefs about the natural world and a diversity of skills for learning about the natural
world. Research also tells us that accomplished teachers know how to take advan-
tage of this diversity to enhance student learning and skill development. Accom-
plished teachers also know how to mediate student learning, that is, provide the
helping hand and probing questions that enable students to move ahead in their un-
derstanding of conceptual structures, criteria for evaluating the status of knowledge
claims, and strategies to communicate knowledge claims to others. Furthermore,
accomplished teachers know how to create a classroom climate that promotes the
sharing and display of students’ ideas—thus making learners’ thinking visible and
teachers’ assessment of inquiry possible.
The purpose of this chapter is to provide an overview of possible frameworks
teachers can use to conduct assessments of students’ engagement in scientific in-
quiry. In thinking about ways to conduct assessments around students’ inquiry em-
ploying evidence to generate explanation, I will examine two factors. One is the
design of classroom learning environments, by which is meant the curriculum, in-
struction, and assessment models that promote inquiry. The other factor is how to
engage and facilitate students in thinking about the structure and communication of

42 National Science Teachers Association

 CHAPTER 4

scientific information and knowledge (i.e., the conceptual, epistemic, and social do-
mains). Herein, I maintain, lie the fundamentals of assessing scientific inquiry. The
teacher needs to create a classroom learning environment that makes it possible to
“listen to inquiry” and then employ a set of strategies that make it possible to give
feedback on students’ use of scientific information and the construction and evalua-
tion of scientific knowledge claims. Adapting the opening quote by Bruner, I can
think of no other way for teachers to learn how to assess inquiry other than engaging
in the task of assessing inquiry.

Listening to Inquiry
Teaching in a manner that promotes scientific inquiry is much more than providing
and managing materials and activities for students to conduct investigations. Engag-
ing students in kit-based science or lab investigations in and of itself is not inquiry.
More often than not, kit-based and lab science lessons are designed to confirm through
a final-form science approach what it is we already know. The kind of inquiry I want
us to consider is that which occurs when students examine how scientists have come
to know what they believe to be scientific knowledge and why they believe this
knowledge over other competing knowledge claims. Such instruction is grounded in
a consideration of what counts as evidence and as explanations.
Scientific inquiry that goes beyond teaching the facts or what we know provides
opportunities for students to share, discuss, debate, and argue the complex set of
ideas, information, beliefs, and questions they bring to or learn in the classroom.
Designing, establishing, and managing assessment-driven science inquiry learning
environments, however, is difficult (Krajcik et al. 1994; Duschl and Gitomer 1997;
Black and Wiliam 1998). Consider the following excerpt from the National Science
Education Standards:

Students at all grade levels and in every domain of science should have the
opportunity to use scientific inquiry and develop the ability to think and act
in ways associated with inquiry, including asking questions, planning and
conducting investigations, using appropriate tools and techniques to gather
data, thinking critically and logically about relations between evidence and
explanations, constructing and analyzing alternative explanations, and
communicating scientific arguments. (NRC 1996, 105)

A prerequisite for assessing students’ inquiry is creating a learning environment

that gives inquiry a voice. Assessing inquiry requires creating conditions that allow
teachers, and eventually the students themselves, to listen to inquiry. The string of
inquiry abilities in the above excerpt gives us some guidance about what we need to
accomplish in the design of our classroom learning environments. The key ideas in
the quote are from the “Changing Emphases to Promote Inquiry” table found in the
National Science Education Standards, presented here as Figure 1.

Everyday Assessment in the Science Classroom 43

CHAPTER 4

Figure 1. Changing Emphases to Promote Inquiry

More Emphasis on
• Activities that investigate and analyze science questions
• Investigations over extended periods of time
• Process skills in context
• Using multiple process skills—manipulation, cognitive, procedural
• Using evidence and strategies for developing or revising an explanation
• Science as argument and explanation
• Communicating science explanations
• Groups of students often analyzing and synthesizing data after defending conclusions
• Doing more investigations in order to develop understanding, ability, values of inquiry and
knowledge of science content
• Applying the results of experiments to scientific arguments and explanations
• Management of ideas and information
• Public communication of student ideas and work to classmates

Source: National Research Council (NRC). 1996. National science education standards.
Washington, DC: National Academy Press, p. 113.

The assessment of inquiry is best thought of as a set of elements that place em-
phasis on examining the processes of engaging in scientific knowing and learning, as
opposed to the products or outcomes of scientific knowing and learning. Scientific
inquiry is fundamentally ill-structured. We may have an initial question and some
sense of the data we need, but with time and experience the inquiry is shaped and
modified. Inquiry is foremost a journey that begins with a set of guiding conceptions
and questions that lead us through the attainment and examination of data. When we
look at the list of “more emphasis on” items in Figure 1, we get a sense of what needs
to happen to make listening to inquiry possible. Investigations are long term and
process skills are not limited to the traditional Science—A Process Approach (SAPA)
skills for conducting investigations (e.g., observing, using numbers, generating hy-
potheses, making and reading graphs, etc.) but also include consideration of the cog-
nitive skills and epistemic criteria that frame scientific reasoning and decision-
making. For an excellent review of research on young children’s inquiry abilities,
see Metz (1998), who writes,

Children at the first level of schooling are capable of independent empirical

investigations that, while falling short of older children and adults in
adequacy of experimental design and inference, support both fruitful theory
construction and the improvement of the inquiry process itself. Even before
instruction, many aspects of children’s theorising resemble the theorising of
scientists. (85)

44 National Science Teachers Association

 CHAPTER 4

The crux of the matter is simple to state but complex to implement and manage.
Teachers need to develop new ways of listening to and monitoring students’ scientific
reasoning and thinking. A key component in the design of inquiry learning environments
is developing or using extended (two- to four-week-long) instructional contexts. Table 1
presents information on a set of science inquiry units developed with support from the
National Science Foundation. Most are web-based, thus providing easy access for re-
view. What you will find in all of these inquiry units is a compelling inquiry context or
problem to facilitate students’ meaningful engagement in inquiry lessons. Another im-
portant design feature of the units in Table 1 is the incorporation of formative assessment
tasks and activities into the instructional sequence that help make visible students’ think-
ing. Here the challenge is developing students’ abilities for communicating and repre-
senting what they know and how and why they believe what they know. White and
Gunstone’s Probing Understanding (1992) is a useful resource here for teachers. Each of
the main chapters outlines how a different instruction strategy (e.g., concept maps, draw-
ings, interviews, demonstrating) can be used to probe and monitor students’ understand-
ing. Many examples of student work are provided. As students become more adept at
communicating and representing scientific information, teachers will, in turn, be in a
better position to listen and give feedback. It is a symbiotic relationship.
Yet another important learning environment component to support inquiry is
that of teachers, and eventually students themselves, developing and employing ana-
lytical insights and criteria for assessing the thinking and the scientific inquiry being
revealed. Here is where my adaptation of Bruner’s quote applies. Teachers need to
begin listening to inquiry and then develop and apply criteria that engage learners in
deciding “what counts” as evidence and as explanations.
In summary, an important aspect of assessing inquiry is designing the learning
environment to promote inquiry activities and students’ reporting and sharing infor-
mation and ideas. In the next sections of the chapter we will examine two strategies
for targeting assessment of inquiry. First, we consider in more detail how science
inquiry is a process of transition from evidence to explanation. Then we consider
strategies for supporting assessment conversations with students. Here we consider
the importance of conducting assessments in the three domains—conceptual,
epistemic, and social—that frame scientific inquiry.

Evidence and Explanation

Science distinguishes itself from other ways of knowing by appealing to evidence
that is deemed objective by its practitioners and then using the evidence to put forth
testable explanations. Scientific ideas and information are rooted in evidence and
guided by our best-reasoned beliefs in the form of the scientific theories that frame
investigations and inquiries. All elements of science—questions, methods, evidence,
and explanations—are open to scrutiny, examination, justification, and verification.
Inquiry and the National Science Education Standards (NRC 2000) identifies five
essential features of classroom inquiry, which are presented in Figure 2.

Everyday Assessment in the Science Classroom 45

CHAPTER 4

Table 1. National Science Foundation–Supported Full-Inquiry Science

Units

Curriculum Project and

Web Address Project Leader Description of Units

LeTUS—Learning Technologies in Louis Gomez, Middle- and secondary-level

Urban Schools Brian Reiser, and inquiry-based units on climate,
www.letus.org Daniel Edelson, air quality, solar energy, Create-a-
Northwestern U. World.

Learning by Design Janet Kolodner, Middle school; physical and

www.cc.gatech.edu/projects/lbd/ Georgia Tech U. Earth science units designed
around case-based reasoning.

WISE—Web-based Inquiry Science Marcia Linn and Grades 5–12; 2-week capstone
Environment Jim Slotta units; 13 projects in physical,
wise.berkeley.edu U. of California, Berkeley life, and Earth science.

One Sky, Many Voices Nancy Songer, K–12; environmental focus;

www.onesky.umich.edu U. of Michigan 4–8-week units; Kids as
Global Scientists, BioKids.

MUSE—Modeling for Jim Stewart, Secondary biology and Earth

Understanding in U. of Wisconsin science; extended 9-week courses;
Science Education units on genetics, evolution.
www.wcer.wisc.edu

SEPIA—Science Education Richard Duschl, Middle school; 4–6-week problem-

through Portfolio Instruction King’s College London based inquiry units; units on
and Assessment flotation, acid/base chemistry,
e-mail: richard.duschl@kcl.ac.uk earthquakes, and volcanoes.

BioLogica Concord Consortium High school; independent inquiry

biologica.concord.org and scientific reasoning;
units on genetics, cell, DNA.

Hi-CE—Center for Highly Joseph Krajcik, Middle and high school; modeling
Interactive Computing in U. of Michigan software (e.g., Science Laboratory)
Education to investigate complex problems.
hi-ce.eecs.umich.edu

BGuILE—Biology GUided Brian Reiser, Middle and high school; scientific

Inquiry Learning Environments Northwestern U. investigation and argumentation
www.letus.org/bguile/ using puzzling and authentic
problems; units on evolution,
animal behavior.

46 National Science Teachers Association

 CHAPTER 4

Figure 2. Essential Features of Classroom Inquiry

• Learners are engaged by scientifically oriented questions.

• Learners give priority to evidence, which allows them to develop and evaluate explanations
that address scientifically oriented questions.
• Learners formulate explanations from evidence to address scientifically oriented questions.
• Learners evaluate their explanations in light of alternative explanations, particularly those
reflecting scientific understanding.
• Learners communicate and justify their proposed explanations.

Source: National Research Council (NRC). 1996. National science education standards.
Washington, DC: National Academy Press, p. 25.

The emphasis in Figure 2 on the words evidence and explanations appears in the
original. Science at its core is fundamentally about acquiring data and then trans-
forming that data first into evidence and then into explanations. The point I want to
make here is that preparation for making scientific discoveries and engaging in sci-
entific inquiry is linked to students’ opportunities to examine the development and
unfolding or transformations of data across the evidence-explanation (EE) continuum.
The EE continuum and three transformations are presented in Figure 3. The strategy
I propose is to allow students to make and report judgments, reasons, and decisions

Figure 3. Scientific Inquiry and Communication Processes

Everyday Assessment in the Science Classroom 47

CHAPTER 4

during three critical transformations in the EE continuum. One is selecting data to

become evidence. Two is analyzing evidence to generate models and/or locate pat-
terns of evidence. Three is locating or otherwise determining the scientific explana-
tions that account for the models and patterns of evidence.
At each of the three transformations, students are encouraged to share their think-
ing by engaging in argument, representation and communication, and modeling and
theorizing. At each transformation, teachers are to engage in assessments of inquiry
by comparing and contrasting student responses to each other and, importantly, to
the instructional aims, knowledge structures, and goals of the science unit. In other
words, effective assessments of science inquiry examine students’ beliefs and deci-
sion-making concerning the transformations of data to evidence, evidence to pat-
terns or models, and patterns/models to explanation.
Another reason to adopt the EE continuum as an instructional framework for guid-
ing the planning and/or design of curriculum, instruction, and assessment models is
that it helps slow down the pace of instruction and, thus, helps facilitate assessment of
inquiry. The unfolding of data takes time and this is another reason why effective
inquiry units are longer in length. When teachers pause to allow students to come
together and discuss and debate what they know, what they believe, and what evidence
they have to support their ideas, thinking is made visible, thus enabling monitoring and
assessments of the communication of information and of the thinking.
The commitment here is to curriculum frameworks that (1) promote full or ex-
tended instructional sequences rather than partial or short, single-lesson instructional
sequences and (2) intentionally embed into the instructional sequence assessment-
driven activities that facilitate feedback on the conceptual, epistemic, and social goals
of the unit. Again, the units presented in Table 1 are good examples of curriculum
material that do this. Typically, we are looking at units that are two to four weeks in
length, sometimes longer. The additional time is needed to make room for student
conversations and representations of reasoning that, in turn, make possible assess-
ments of inquiry. The significant trade-off that needs to be made is holding down the
number of concepts, science terms, and labels so that the data-driven elements of
scientific inquiry can be examined and debated. The National Science Education
Standards recommend that at each grade level K–12 students be given the opportu-
nity to complete at least one full-inquiry unit.
When core scientific concepts are kept to a minimum (e.g., in the curriculum
designs for Project SEPIA [Science Education through Portfolio Instruction and As-
sessment] [Duschl and Gitomer 1997] a maximum of 20 core concepts frame the
inquiry), students’ thinking can shift from memorization of terms to the important
tasks of knowledge integration and the development of criteria for evaluating scien-
tific knowledge claims. Let’s consider in more detail how the three transformations
promote assessment opportunities. During full-inquiry instructional sequences we
want students to probe how raw data from investigations are analyzed and selected to
be evidence, how evidence is selected and analyzed to identify patterns, trends, and

48 National Science Teachers Association

 CHAPTER 4

models, and how patterns, trends, and models are selected and analyzed to generate
scientific explanations. Each is an important “transitional” step or transformation in
doing science. Each transition involves asking students to commit to judgments about
“what counts” and thus makes visible students’ thinking and beliefs about “what
counts.” Assessments of inquiry on each of the three domains—conceptual, epistemic,
and social—can now take place.
Fundamentally important to the success of conducting assessments along the EE
continuum is capturing the diversity of thinking found in students’ judgments and
decisions. One important dynamic is to always ask students to provide reasons and
evidence to back up the judgments and decisions they make. When students are asked
to explain or justify their results, judgments, and decisions with reasons and evi-
dence, their thinking is made visible and several important assessments of inquiry
can occur. For example, a common strategy for making students’ thinking visible is
to ask each student or group of students, upon completion of an investigation, to
place their data into a class data table on a board or overhead transparency or in a
computer data file. Making public the display of data facilitates discussion about
what data to use as evidence. The class data table may reveal, among other things,
errors and successes in measurement and in data recording (e.g., placement of deci-
mal points, use of formulas). Students can begin to sense patterns in the data (e.g.,
the second transformation in the EE continuum). Discussion about the class data
table may reveal that more data are needed to complete the inquiry or that the data
being collected can’t be used to answer the questions being posed.
Through the discussion of data, mediated and guided by the teacher, students
begin to develop a sense of the criteria for good data. Students begin to learn that
scientific inquiry involves asking questions of the data and using the data to ask new
questions. Full-inquiry instructional sequences make provisions for just such occur-
rences. An excellent source of instructional strategies to use with students for han-
dling the analysis and reporting of data is Investigating Real Data in the Classroom
(Lehrer and Schauble 2002).
Another instructional strategy that promotes assessments of what counts as data
and evidence is providing students with options for obtaining data and then asking
them to justify their choices. The choices and the reasons provided to support choices
create another kind of assessment opportunity about students’ thinking and reason-
ing. Many kit-based science investigations, in the interest of allowing time to “cover”
the content, are structured so that all students use the same equipment, probe the
same question, and use the same materials. When the outcomes are the same, what is
there to discuss and debate? Monolithic knowledge does not engender scientific
reasoning or critical thinking. The goal during science inquiry lessons is to create
conditions that stimulate diversity among students’ responses. A key dynamic to
effective assessment of inquiry is exposing the different reasons and beliefs that
students hold in order to encourage communication of ideas and argumentation about
ideas, both essential features of scientific inquiry.

Everyday Assessment in the Science Classroom 49

CHAPTER 4

Consider the example of the “Acids and Bases” middle school unit (Erduran
1999) designed as part of Project SEPIA. To promote diversity of outcomes and thus
create assessment conversation opportunities, we decided to provide students with
six different techniques for determining acidity and alkalinity early in the unit. Mak-
ing students’ thinking visible at this point in the instructional sequence allowed for
an assessment of students’ comprehension of the unit goals and their thinking about
“what counts” as good data.
The performance goals for the six-week, full-inquiry “Acids and Bases” unit were
to identify five unknown substances as either acids or bases and then devise a strategy
for safe disposal. A core concept was neutralization since safe disposal depends on
neutralization. The six techniques were used to test known substances, and students
were told that the goal of the lesson was to decide which technique(s) would be best to
use with the unknown samples. Only two of the six techniques for testing the un-
knowns could determine if a solution was neutral. After testing all six techniques,
students were asked to select, and give reasons for choosing, the technique they be-
lieved to be best.
The students’ lab reports and the oral discussions that followed were quite re-
vealing. First, each of the techniques was considered “best” by at least some of the
students, and thus we achieved the requisite diversity of thinking to hold an assess-
ment conversation. For some, litmus paper was the choice because it was fast and
easy. Bromothymol blue and yellow were the choice of others because these were
judged to be pretty and fascinating. These were interesting perspectives but not
grounded in scientific reasoning nor attentive to the inquiry goals of the unit’s prob-
lem. The second way the students’ reports were revealing is they made visible stu-
dents’ thinking, that is, that some students had lost sight of the goal of the full-
inquiry. Given the research that shows teachers’ goals for lessons and students’
perceptions of the goals of lessons are frequently not the same (Osborne and Freyberg
1985), we felt the diversity of student responses provided a timely chance to conduct
an assessment conversation that focused on long-term conceptual and epistemic goals
of the inquiry unit.
Neutralization is the important guiding concept for this problem-based inquiry
unit, but only about 25 percent of the students demonstrated such awareness or under-
standing. The teacher and I sat down and sorted the lab reports into piles based on
students’ reasons for choice of method. Examining students’ work and deciding how it
will be used to promote learning are the first steps in carrying out an assessment of
inquiry. By surveying and selecting samples of student work that reflected the diversity
of thinking, we were able to plan a whole class discussion on the best methods for
gathering data. The second step in carrying out an assessment of inquiry is to make the
diversity of students’ thinking visible to the class. Asking students themselves to exam-
ine the range of responses—followed by a teacher-led class discussion—made stu-
dents’ thinking visible. Through the oral testimonies of students and via debates over
alternative perspectives, students who were off-track on the unit goals were brought

50 National Science Teachers Association

 CHAPTER 4

back on track. Additionally, the assessments of students’ thinking by the students them-
selves help them see how the goals of the inquiry (e.g., understanding neutralization
for safe disposal) help determine “what counts” as good data. Subtle, yes, but an im-
portant step in aiding students to make progress on the standard “Developing students’
understandings about scientific inquiry” (NRC 1996).
The third way students’ reports from the six-techniques activity were revealing
had to do with the introduction of new concepts by the students into the classroom
learning environment. When given the chance to communicate what they know and
believe, students will surprise you in what they bring to a discussion. Furthermore,
the fact that concepts come from students, rather than the textbook or the teacher,
means that the inquiry begins to rest with the students and important connections are
made. This is a powerful dynamic in shaping a science inquiry learning environment
because students begin to own the ideas. In this lesson, students who selected pH
paper or Universal Indicator did so because, in the words of one student, “It tells you
how much of an acid it is” and of another, “You know how much acid or base is in it.”
To the astute teacher who knows acid and base chemistry, such comments are
steppingstones to developing an understanding of the pH scale and the concept of
ions (H+, OH-).
Instructional sequences can be designed to promote assessments of scientific
inquiry by having students report on their ideas, beliefs, and reasons. A key element
of science inquiry units that promote formative assessments is designing activities
and tasks that promote teacher “listening” and do so over time in each of the three
assessment domains—conceptual, epistemic, and social. Fill-in-the-blank worksheets
and responses to multiple-choice questions may help with feedback on simple con-
cept learning but do little to promote feedback on the conceptual, epistemic, or so-
cial domains associated with scientific inquiries that seek links between evidence
and explanation.

Assessment Conversations in Three-Part Harmony

Up to this point, the emphasis has been on designing learning environment condi-
tions that help teachers to carry out assessments of inquiry. Let’s now turn our atten-
tion to some practical issues—namely, the domains of assessments a teacher needs
to consider and a strategy teachers can use to facilitate assessment conversations on
scientific inquiry. The chapter opened with references to three important domains to
consider when setting out to establish a classroom learning environment that will
promote the assessment of inquiry.
• Conceptual Domain. First, there is the need to have access to information that
will allow you to assess both (a) the conceptual frameworks you want students to
learn and use and (b) the alternative conceptual frameworks students naturally
bring with them to lessons. Students, like scientists, come to any inquiry with a
set of beliefs and commitments that guide them in reasoning and decision-

Everyday Assessment in the Science Classroom 51

CHAPTER 4

making. Effective science teachers will continually assess pupils’ conceptual

understandings. Scientific inquiry typically requires knowledge integration across
several areas of science and ways of reasoning. The design of curriculum activi-
ties and tasks should promote teachers’ access to students’ reasoning about sci-
ence concepts and to students’ use of science concepts.
• Epistemic Domain. Second, there is the need to have access to information that
will allow you to assess the epistemic frameworks students use during an in-
quiry. Epistemic frameworks are the knowledge structures (e.g., data, evidence,
principles, theory) and the rules and criteria used in scientific inquiry for deter-
mining “what counts” as a good experiment, measurement, graph, explanation,
or argument, to name but a few products of inquiry. When we ask students to
develop and evaluate experiments, evidence, hypotheses, explanations, models,
or arguments, we are asking them to employ epistemic frameworks.
• Social Domain. Third, there is the need to have access to information that will
allow the teacher to assess the representation and communication frameworks
students will need to use while engaging in inquiry. Here the goal is to make
thinking visible through oral, written, and pictorial presentations of concepts,
conceptual frameworks, and criteria that frame the inquiry. The more accom-
plished students get at communicating and showing what they know, then the
more accomplished teachers can get at listening to students’ thinking and pro-
viding assessments of inquiry. Deliberation and discourse need to be a central
part of the classroom learning environment.
Let’s turn once again to the Essential Features of Classroom Inquiry as presented
in Figure 2 and consider how each inquiry feature is related to the three assessment
domains. The conceptual frameworks students use will influence each of the five fea-
tures, but, in particular, students’ conceptual frameworks will affect Feature 1. The
research on alternative conceptions tells us how important it is to work with students’
ideas. Construction of concept maps is one instructional strategy for making thinking
visible and assessing how learners are integrating knowledge. Other useful strategies
for promoting assessment of the conceptual frameworks students bring to science les-
sons are activities that engage them in design activities or problem-solving while em-
ploying scientific ideas and information. When students are given assignments that ask
them to provide reasons or evidence to support design or problem-solving decisions,
conceptual commitments are exposed and feedback is possible.
Yet another way to make visible students’ thinking about conceptual frameworks
is through question generation. After establishing the initial “focus question” for a
scientific inquiry, making room in lessons to allow students to ask new questions and
hear others asking subsequent questions sheds light on students’ reasoning. But each
of these strategies has to be modeled by the teacher and nurtured through assessment
practices. Students do not arrive knowing how to do concept maps or how to ask

52 National Science Teachers Association

 CHAPTER 4

questions. Teachers need to help them learn how to communicate and represent ideas
with concept maps. Here is an instance where there is overlap with the conceptual
and social domain.
The epistemic frameworks address Features 2, 3, and 4. With a focus on three of
the five essential features, epistemic frameworks are clearly important. Again, provid-
ing opportunities for students to give reasons is a strategy that makes thinking visible.
The epistemic frameworks teachers need to develop among students are such things as
the notion of a fair test or why we use models to tell us things about the real world
(Schauble et al. 1995). Epistemic frameworks function at each of the three data trans-
formations. Decisions about evidence, patterns in evidence, and explanations of
evidence are all influenced by choices that determine “what counts” as good/sound
evidence, patterns, and explanations. We know, for example, that when confronted
with a set of data or evidence, many students will seek information that confirms a
personal view and ignore the information that refutes their point of view (Metz 1998).
Assessment practices can help make such faulty thinking visible to learners.
Epistemic frameworks also include the development of the criteria students are
to use to make judgments about ideas and information. A simple criterion is to ask
students if they can give an example or cite an instance when they have seen an idea
or a phenomenon before. A more sophisticated criterion is to ask students to statisti-
cally determine if a reported finding is typical or if it is an outlier for the data being
generated.
Clearly, the representation and communication frameworks are related to Fea-
ture 5. A critically important challenge for teachers seeking to develop their abilities
to assess scientific inquiry is to learn the ways to display and represent data and
information. Visualization techniques for the presentation and representation of data
and information help promote thinking and making thinking visible. There is grow-
ing interest in the use of exploratory data analysis (EDA) techniques with K–12
science students. Initially developed for statisticians, EDA techniques such as stem-
and-leaf and box-and-whisker diagrams are proving successful with helping stu-
dents see the shape of data and the typical patterns in data. Again, Lehrer and
Schauble’s book (2002) is an excellent source of examples of elementary-level stu-
dents using various EDA techniques to discuss data. In turn, these displays of data
help promote assessments that can, for example, help students learn about math-
ematical modeling and statistical probabilistic reasoning.
The best teachers are those who can assess where a learner is in his or her knowl-
edge development and then provide the right intervention to advance the learner.
Assessing and advancing learning is mediation of learning. The general consensus
among educational researchers (Bransford, Brown, and Cocking 2000; Pellegrino,
Chudowsky, and Glaser 2001) is that the most effective learning environments are
those that support mediation of learning. In school science, the enterprise of address-
ing epistemic connections and other elements of inquiry is about carefully designed
learning sequences that engage students in doing science (e.g., investigations, prob-

Everyday Assessment in the Science Classroom 53

CHAPTER 4

lems, and projects). But doing science is also about engaging students in colloquiums
or conversations around the investigations, problems, and projects. As we have al-
ready discussed, there are important inquiry steps between obtaining the raw data of
observation and acquiring the theory-informed selection of evidence used to develop
explanations, theories, and models. Richard Grandy (1997), in a critique of the popular
constructivist practices in science classroom, warns that

[w]hat is missing [from most constructivist teaching models] are the

epistemic connections that relate theory to supporting data, to conflicting
theories, to anomalous data, to equivocal data…. What can be taken as data
and what is disqualified, what is strong evidence and what is weak evidence,
is always judged against the background provided by the community’s
experience with the theories, the data domain, and the instruments in
question…. The demarcation between what counts and what does not … is
critical to the ongoing enterprise. (49–50)

Grandy is reinforcing the ideas presented above that stress the notion that good
science teaching requires a focus on the refinement and extension of ideas about
“what counts.” Up to this point the discussion has been on evidence and explanation.
Now we turn our attention to the language of the classroom. We are talking about
developing among students the discriminating language that promotes doing scien-
tific inquiry and understanding scientific inquiry. It is a language, for example, where
students are comfortable speaking about the shape, spread, range, mean, mode, me-
dian, and outliers of data sets. It is a language that builds arguments supported by
premises that, in turn, are supported by backings and evidence.
The call for conversations is a recognition of the value and importance that repre-
sentation, communication, and evaluation play in science learning. I use conversation
in a very broad sense to include, among others ideas, argumentation, debate, modeling,
drawing, writing, and other genres of language. Such an expanded repertoire helps us
to consider an important domain of research in both formal and informal science learn-
ing settings, namely, how to mediate the learning experiences. The position advanced
by Schauble, Leinhardt, and Martin (1997), and adopted here, is that such learning
mediations should focus on promoting talk, activity structures, signs and symbol sys-
tems, or, collectively, what I wish to call conversations. For science learning, the con-
versations should mediate the transitions from evidence to explanations, do so within
the salient conceptual domain, and, thereby, unfold discovery and inquiry.
The idea of conversations or colloquiums in science classrooms is taken from
Lansdown, Blackwood, and Brandwein (1971). Grounded in Vygotsky’s theory of
learning that claims meaning is obtained through language, colloquiums are “speaking
together” opportunities that begin with “a pooling of observations, getting a collection
of facts into the arena, so to speak, to make individuals aware of common data seen
from different viewpoints. This is the beginning of speaking together.” (Lansdown,

54 National Science Teachers Association

 CHAPTER 4

Blackwood, and Brandwein 1971, 120). Speaking together activities can be whole
class events, small group activities, or interactive, computer-supported learning con-
texts. In each instance, the conversations present “listening to inquiry” opportunities.
Speaking together opportunities, when properly planned and managed, are also won-
derful occasions for making thinking visible. You will find that the curriculum units
listed in Table 1 (p. 45) have designed formats for supporting science conversations.
In Project SEPIA, the speaking together occasions take place during instructional
episodes we call assessment conversation activities. The assessment conversation is an
idealized model of teaching practice. These conversations are structured discussions in
which student products and reasoning are made public, recognized, and used to de-
velop questions and activities that can (a) promote conceptual growth for students and
(b) provide assessment information to the teachers. The assessment conversation has
three general stages, presented in Table 2. The first stage is to receive student ideas.
This requires that students be allowed to represent their understanding as fully as pos-
sible. To this end, SEPIA instruction incorporates detailed writing, drawing of anno-
tated pictures, linkages between drawings and writings, construction of storyboards,
and many other techniques that allow students to show what they know.

Table 2. Stages of the Assessment Conversation

Stage Description

Stage 1: Receiving Information • Individual or group efforts on specialized

tasks that by design bring about a diversity
of student responses and range of representation or
ideas.
• Teacher and students make explicit and
publicly display via posters, presentations,
charts, overheads, and so forth the diversity
of students’ efforts, representations of
meanings and understandings, and
performances on the tasks.

Stage 2: Recognizing Information • Teacher examines critically and makes an

appraisal of the diversity of student efforts,
meanings, understanding, and performances
and selects student work according to goal
domains (conceptual, epistemic, social) to
be addressed.
• Teacher and students work toward a synthesis
of what counts or stands for appropriate efforts,
meanings, understandings, and performances
employing agreed upon criteria.

Stage 3: Using Information • Applying what has been learned to an evaluation

of previous efforts, meanings, understandings,
and performances or to the design of a new
investigation for advancing efforts, meanings,
understandings, and performances in the
present domain of inquiry.

Everyday Assessment in the Science Classroom 55

CHAPTER 4

Once the students have represented their understanding, it is the responsibility of

the teacher to recognize the ideas (Stage 2) in the classroom in relation to unit or
lesson goals. Inevitably, there will be a diversity of ideas. In traditional classrooms,
this diversity is quickly constrained through an appeal to find the “best answer.” In a
full-inquiry unit that supports assessment conversations, diversity is made public
and resolved through a discussion that is governed by scientific criteria and evi-
dence. In this second step of an assessment conversation, the students’ ideas and
information are publicly shared (e.g., displayed on overhead, copied and distributed
to student groups) and discussed. Here the teacher needs to intervene to select the
samples of student work to bring to the attention of students. The teacher also needs
to decide the focus of the assessment conversation depending on the inquiry goals
(e.g., conceptual, epistemic, social) that she or he wishes to pursue at this time.
In recognizing the diversity of student work, teachers need to select work that
differs on dimensions relevant to the conceptual, epistemic, and social domain(s)
being explored. The teacher must guide the class through a discussion in which the
critical differences in student representations and reasoning are highlighted. Con-
sider, for example, the storyboards in Figure 4. The three storyboards are from a
seventh-grade class that was studying flotation and buoyancy using the “Vessels”
SEPIA unit. Students were directed to draw (storyboard), in as many frames as they
wished, the action of and forces acting on a ball released under water. As you can
see, each is quite different and revealing. The two principal goals for the “Vessels”
SEPIA unit are (1) designing a vessel hull that maximizes carrying capacity and (2)
developing a causal explanation for why vessels can float while carrying a load.
With regard to the conceptual goals of the unit, we can examine Storyboard #3 and
discuss the omission of gravity forces whenever the ball is under the water. Is this
accurate? The other two storyboards show gravity acting on the ball when it is under
the water. Which depiction is correct? We can also ask what role water displacement
has in the physics of flotation and buoyancy. With regard to the social goals of the unit,
we can compare Storyboard #1 with #2 and #3. Storyboards #2 and #3, respectively,
use length of arrow and number of arrows to describe the forces acting on the ball.
Further, Storyboard #3 uses a key (G = gravity). Students can discuss how the use of
arrows, labels, symbols, and narration helps to communicate scientific ideas.
When we ask students about the information in their drawings, thinking is made
visible and mediated learning occurs. Middle school students, however, are quite
reluctant to discuss each other’s work in public. Through our research, we have found
it useful for teachers to import student work from other classrooms for purposes of
assessment conversations. When the author is anonymous, students are good to go.
Once the diversity of ideas is public, the teacher can use that diversity (Stage 3)
as a basis for achieving a consensus view in the classroom. The teacher poses ques-
tions and facilitates a discussion that results in the consensus. Generally, one group
of students’ representation or reasoning is viewed as satisfying operating criteria
more than other students’ (e.g., it is more consistent with the evidence). A final use

56 National Science Teachers Association

 CHAPTER 4

Figure 4. Storyboards: Students Draw the Action of and Forces

Acting on a Ball under Water
Storyboard #1

Storyboard #2

Storyboard #3

Everyday Assessment in the Science Classroom 57

CHAPTER 4

of student understanding is to entertain how the accepted view may generalize to

new and different situations. In Storyboards #1 and #2 the arrows are drawn straight
up and down while in Storyboard #3 the forces are depicted acting as a cluster around
the submerged object in one upward direction but buoyant pressure acts on a sub-
merged object in all directions. This is a subtle but important distinction for develop-
ing a causal explanation for why a vessel can float carrying a load. Storyboard #3
can be used to advance students’ thinking about the role water pressure has on flota-
tion and buoyancy (e.g., the greater the depth, the greater the water pressure) and on
the design of a vessel that incorporates high sides.

Summary
The assessment of inquiry is a complex task. We have discussed the importance of
organizing classroom learning environments and using instructional sequences that
support “listening to inquiry.” We have examined some characteristics of instruc-
tional sequences that promote inquiry. We have discussed how teachers need to sup-
port assessments across three domains—conceptual, epistemic, and social—to make
thinking about doing science visible.
Successful assessment-driven learning environments create a symbiotic relation-
ship between teacher and students. As the teacher mediates learning, students’ knowl-
edge and skills for communicating develop. This, in turn, enables the teacher to
mediate learning at new, more complex levels of inquiry. In time, the students will
learn from each other as well. Effective mediation on the part of the teacher, though,
in addition to everything that has been discussed, also demands a strong understand-
ing of the subject matter and of the inquiry goals for the instructional sequence.
One key to success with assessing inquiry is to give students a voice and then
teach them how to use that voice. Another key to success is for teachers to learn how
to listen to and make sense of student voices. In our science and technology world,
our students know much more than we give them credit for. Deficit, “my students
can’t do that” models of learners must be replaced by more open and supportive
images. When you allow the conversation of science to open up along the evidence-
to-explanation continuum and examine the three transformations, you will be amazed
at the depth and breadth of ideas and information your students will bring to the
debate. Thus, the assessment of inquiry must be situated in carefully designed in-
structional sequences that support conversations about evidence and explanations
and that offer ongoing assessment opportunities. So, take a look at the instructional
units in Table 1 and get in the game. The only way to get better at the art of assessing
inquiry is to begin assessing inquiry.

References
Black, P., and D. Wiliam. 1998. Assessment and classroom learning. Assessment in Education 5(1):
7–73.
Bransford, J., A. Brown, and R. Cocking, eds. 2000. How people learn: Brain, mind, experience, and
school. Washington, DC: National Academy Press (http://www.nap.edu).

58 National Science Teachers Association

 CHAPTER 4

Bruner, J. 1961. The act of discovery. Harvard Educational Review 31: 21–32.
Duschl, R., and D. Gitomer. 1997. Strategies and challenges to changing the focus of assessment and
instruction in science classrooms. Educational Assesssment 4(1): 37–73.
Erduran, S. 1999. Merging curriculum design with chemical epistemology: A case of teaching and
learning chemistry through modeling. Ph.D. diss., Vanderbilt University.
Grandy, R. 1997. Constructivism and objectivity: Disentangling metaphysics from pedagogy. Sci-
ence and Education 6 (1, 2): 43–53.
Krajcik, J., P. Blumenfeld, R. Marx, and E. Soloway. 1994. A collaborative model for helping teach-
ers learning project-based instruction. Elementary School Journal 94(5): 483–98.
Lansdown, B., P. Blackwood, and P. Brandwein. 1971. Teaching elementary science: Through inves-
tigation and colloquium. New York: Harcourt Brace Jovanovich.
Lehrer, R., and L. Schauble, eds. 2002. Investigating real data in the classroom: Expanding children’s
understanding of math and science. New York: Teachers College Press.
Linn, M., and S. Hsi. 2000. Computers, teachers, peers: Science learning partners. Mahwah, NJ:
Lawrence Erlbaum.
Magnani, L., and N. Nersessian, eds. 1999. Model-based reasoning in scientific discovery. New York:
Kluwer Academic/Plenum.
Metz, K. 1998. Scientific inquiry within reach of young children. In International Handbook of Sci-
ence Education, eds. B. Fraser and K. Tobin, 81–96. London: Kluwer Academic.
Millar, R., J. Leach, and J. Osborne, eds. 2001. Improving science education: The contribution of
research. Philadelphia: Open University Press.
Minstrell, J., and E. van Zee, eds. 2000. Inquiring into inquiry learning and teaching in science.
Washington, DC: American Association for the Advancement of Science.
National Research Council (NRC). 1996. National science education standards. Washington, DC:
National Academy Press (http://www.nap.edu).
———. 2000. Inquiry and the national science education standards. Washington, DC: National Acad-
emy Press (http://www.nap.edu).
Osborne, R., and P. Freyberg. 1985. Learning in science: The implications of children’s science.
Auckland, New Zealand: Heinemann.
Pellegrino, J., G. Baxter, and R. Glaser. 1999. Addressing the “two discipline” problem: Linking
theories of cognition and learning with assessment and instructional practice. In Review of re-
search in education (Volume 24), eds. A. Iran-Nejad and P. D. Pearson, 307–53. Washington, DC:
American Educational Research Association.
Pellegrino, J., J. Chudowsky, and R. Glaser, eds. 2001. Knowing what students know: The science and
design of educational assessment. Washington, DC: National Academy Press (http://www.nap.edu).
Schauble, L., G. Leinhardt, and L. Martin. 1997. A framework for organizing a cumulative research
agenda in informal learning contexts. Journal of Museum Education 22(2, 3): 3–11.
Schauble, L., R. Glaser, R. Duschl, S. Schulze, and J. John. 1995. Students’ understanding of the
objectives and procedures of experimentation in the science classroom. The Journal of the Learn-
ing Sciences 4(2): 131–66.
White, R., and R. Gunstone. 1992. Probing understanding. London: Falmer Press.

Everyday Assessment in the Science Classroom 59

 CHAPTER 5

Using Questioning to Assess and Foster

Student Thinking
Jim Minstrell and Emily van Zee

Jim Minstrell taught science and mathematics for thirty-one years in the public schools, retiring
from teaching in 1993. For over twenty-five years he has also conducted research on learning
and teaching in his own classroom and the classrooms of many teacher colleagues. He contin-
ues doing research and development on assessment of understanding and skills, professional
development of teachers, and evaluation of educational programs. He and his colleagues build
technical tools to help teachers monitor students’ thinking in science and mathematics.

Emily van Zee is an associate professor of science education at the University of Maryland,
College Park. She has taught middle school science, contributed to the development of physics
curricula, and assisted in physics programs for teachers. She conducted her postdoctoral re-
search in Jim Minstrell’s high school physics classroom. Currently she is collaborating with teachers
in the development of case studies of science learning and teaching. Her research has focused
on student and teacher questioning during conversations about science.

T he National Science Education Standards recommend that teachers shift empha-

sis from assessing memorization of information to assessing scientific under-
standing, reasoning, and utilization of knowledge (NRC 1996, 82). This shift in
emphasis underlies Teaching Standard B: “Teachers of science guide and facilitate
learning. In doing this, teachers … orchestrate discourse among students about scien-
tific ideas” (NRC 1996, 32). A key aspect of such discourse is appropriate questioning.
In many science classrooms, questioning typically involves a three-turn exchange
in which the teacher asks a question, a student answers, and the teacher evaluates the
answer. Student questions are rarely heard in such classrooms. In too many classes,
teachers exercise the traditional authority of their position to get students to accept
“the” answer rather than engaging students in a reasoned discussion of issues and
ideas. One alternative is to involve students in extended conversations. Rather than
trying to elicit the “right” answer, teachers can use questioning to elicit student think-
ing, to find out what conceptions and ways of reasoning their students are using.
A shift toward ongoing assessment involves teachers in establishing norms of scien-
tific argumentation in their classrooms. Scientific argumentation involves knowing how
to generate new knowledge from observations and being able to trace the reasoning
through which we go from our observations to our conclusions and beyond. Knowing
how we know is at least as important as knowing what we know. Ongoing assessment
through the use of different sorts of questions can help guide learners through their
thinking. One way to use questioning for ongoing assessment is to ask questions, first to
diagnose the state of students’ thinking and then to prescribe an appropriate next step to
guide students toward a deeper understanding. A comment from one of the students
captures the philosophy of this approach to instruction: “Seems like I have to think about
everything and understand about everything in order to take another step further.”

61
CHAPTER 5

In this chapter, we first present and discuss an example of using questioning to

assess and foster student thinking through a class discussion. The lesson is drawn
from a videotape of one of the authors’ (JM) high school physics classes. After building
a context for questions in the first part, we will summarize some of the sorts of
questions that are recommended for ongoing assessment of students’ understanding,
reasoning, and use of knowledge. Finally we will use an example from the experi-
ence of one of the authors to show how results of assessment can be used to guide the
adaptation of curriculum and instruction in the classroom.

A Context: Example of Ongoing Assessment through Questioning

Sometimes even successful science students maintain two separate systems for inter-
preting natural phenomena—what they themselves believe and what they have memo-
rized as the accepted view. The lesson described here was a “benchmark lesson”
whose purpose was to help students “make sense” of formal representations of physical
systems by enlarging their own ways of thinking (diSessa and Minstrell 1998). The
conceptual underpinnings of an early version of this lesson have been described
elsewhere by the first author (Minstrell 1982).
The lesson began with an ungraded diagnostic question to elicit students’ thinking
about forces and proceeded through a series of episodes initiated by questions from the
students as well as the teacher. This is an example of an exploration of conceptual
understanding in which the teacher followed the students’ lead in thinking.

Elicitation Questions
Elicitation questions serve several functions. They provide an opportunity for stu-
dents to articulate the conceptions and reasoning with which they begin their study
of a particular topic. They highlight central issues in the upcoming unit and alert
students to what they should be thinking about in the days ahead. Often the elicita-
tion questions and follow-up class discussions involve demonstrations that enable
students to view physical phenomena more carefully than they may have previously.
The follow-up discussions are designed to enable students to examine contexts within
which their intuitions are valid and to recognize contexts within which they may
need to modify their thinking. By providing the opportunity to articulate their initial
conceptions and to clarify these ideas, the elicitation questions and subsequent dis-
cussions help students begin building new, more powerful conceptions.
An elicitation question on forces is shown in Figure 1a. The question asks stu-
dents, individually at first, to identify the forces acting upon a book that is lying on a
table and to draw and label their representation of the force(s) acting in that situa-
tion. In typical responses, a vertical arrow pointing downward represents the gravita-
tional force exerted on the book by the Earth, as shown in Figure 1b. However, in
about half the papers no upward force is shown. A response more consistent with the
view of scientists is shown in Figure 1c, which represents an upward force by the
table, balancing the downward force exerted on the book by the Earth.

62 National Science Teachers Association

 CHAPTER 5

Figure 1a. Elicitation Question Related to Forces on a Book at Rest

A book is at rest on a table.

Using arrows, indicate the one or more forces that keep the book at rest on the table. Length
and direction of the arrow(s) indicate the relative size and direction
of each of the forces on the book. Next to each arrow write the source of each force
(i.e., force by ____).

Figure 1b. Typical Student Response to Elicitation Question in

Figure 1a.

Figure 1c. Typical Scientists’ Response to Elicitation Question in

Figure 1a.

Many students begin the study of force with the idea that only animate objects
are capable of exerting forces, that the table cannot exert an upward force on the
book because it is not able to push actively on the book. The elicitation question and
subsequent discussion help students recognize that an active agent is not necessary,
that inanimate objects can passively exert forces such as the upward force on a book
by a table. We believe that such issues must be explicitly addressed if we want stu-
dents to “make sense” for themselves of ways in which scientists understand and
represent the world.

Everyday Assessment in the Science Classroom 63

CHAPTER 5

Questioning Episodes
The following is an edited transcript of a videotaped, follow-up lesson on forces; it
shows highlights of a discussion that typically requires about 30–45 minutes to con-
duct. We have partially annotated the transcript below with comments that discuss
what is happening conceptually and/or linguistically through the teacher and student
utterances. We divided the transcript into seven episodes that involve questioning by
the teacher and/or students: invitation to students to speak, student initiation of ques-
tioning, student comparison of beliefs, negotiation of meaning between student and
teacher, class poll, group demonstration, and summarizing mini-lecture.

Episode 1: Invitation to students to speak. In the first episode, Minstrell repeated a

statement made by one of the students as a way to invite other students to participate
in the discussion:

Teacher: Does it make sense to everybody that the table is exerting an

upward force on that book? [four-second pause] No? OK. Want to argue
against it?

Students should actively engage in thinking about an idea until they can articu-
late reasoning that “makes sense” to them. The long pause gave students time to
think about the issue and to venture a response (Rowe 1974). By the questioning
“No?”, Minstrell reflected back his interpretation of the students’ nonverbal responses,
that there were some for whom this did not make sense. “OK. Want to argue against
it?” invited students to express their own ideas if they were in conflict with what
another student had stated. Such discussions encourage students to think about is-
sues central to scientific literacy, such as “Why do I believe this?” and “How do I
know it?” (Arons 1983).

Episode 2: Student initiation of questioning. One of the students accepted her teacher’s
invitation. She sought identification of the force by wondering about the reasons
leading to that identification:

Sue: Well, what force is it? I don’t understand why that would be a force.

T: Why that would be a force?

Sue: Why … what force would it be?

T: The force exerted by the table? [repeating what some of the other
students had expressed]

64 National Science Teachers Association

 CHAPTER 5

Sue: Other than just keeping it up?

The question marks in the last two utterances indicate a rising intonation that we
interpret as a question. These contingent responses may be direct echoes or clarifica-
tions as in this episode, or they may involve a more complex paraphrase of the student’s
thought. Such contingent responses seem to help students continue to elaborate their
thinking.

Episode 3: Teacher use of questioning to prompt student comparison of beliefs.

Minstrell then solicited comments from another student:

T: What do others of you think? What do you think? Mary?

Mary: Well, I think it’s got to be exerting some sort of force, maybe it’s just
a force, kinda resistance, ’cause, I mean, if it wasn’t there, it would fall to the
ground. So there’s got to be something [keeping it up].

Sue: How can a table have a force? I mean I understand it’s holding the
book up, but I don’t think there’s any force that’s ….

Dan: It’s just matter.

The teacher uses another question to probe for and clarify the present meaning of
force to these students.

T: How are you conceiving of force that’s different from the way some of the
other people are conceiving of force?

Sue: I guess, well, I don’t know, a force is when, well, I would think, I mean,
like pushing a book, that’s a force, but actually the table’s not pushing the
book up. I mean, it’s just there. It’s keeping it from falling down.

T: OK.

Episode 4: Negotiation of meaning between student and teacher. The student con-
tinued with a question that offered a rationale for doubting the conception stated by
another student and reiterated by the teacher:

Sue: It’s not … how can it have a force if it’s not even doing anything?

T: All right. So the table’s not doing something.

Everyday Assessment in the Science Classroom 65

CHAPTER 5

Sue: It’s holding it up.

T: It’s just there.

Sue: It’s not like pushing it up. It’s just there, so it’s not falling.

T: OK.

Notice that the teacher is assessing and reflecting students’ ideas and not evalu-
ating them. By saying “OK,” Minstrell is acknowledging that he thinks he is under-
standing what the student means. Then, another student offered an assessment of
what makes the concept of force difficult:

Ann: Like Sue said, a lot of people’s conception of a force and mine too is
something active, like pushing or pulling, or something that can be proven
like gravity, or something like that, but it’s hard to conceive of something just
kind of being there as exerting a force. I mean it’s hard to, like when people
think of forces, like lifting or something, but the table’s not lifting it, but you
know there has to be something there, else it will fall to the ground.

Episode 5: Class poll to assess current thinking.

T: OK. So how many of you, at this point, are really struggling with or
questioning the idea of whether or not we want to think of something like the
table as exerting an upward force? [Nearly half of the students raise their
hands.] OK. So several of you are still wrestling with that.

By polling the class periodically during the discussion, Minstrell structured class-
room events in a way that explicitly recognized the process of learning. Polling the
class prompted students to acknowledge their current point of view and to monitor
changes in their conceptions during the discussion.

Episode 6: Group demonstration. Minstrell used a demonstration to establish an

“anchor” (Clement, Brown, and Zeitsman 1989; Camp and Clement 1994) in a con-
text that students could understand, an upward force on the book by the hand. He
then used this anchor to build a bridge toward a new understanding:

T: Bill, would you hold your hand out like that please? [puts book on
student’s outstretched hand] How ’bout in that situation? [students laugh as

66 National Science Teachers Association

 CHAPTER 5

teacher brings a high stack of books and places them on the student’s
outstretched hand] What about that? Do you think, is Bill exerting a force?

In a next sequence not included here, the teacher places the book on a spring that
gets compressed. The students then point out that while the spring is not alive it is
flexible. They ask, “What would happen with a solid, really stiff thing supporting the
book?” The book is put on the demonstration table and a light beam is aimed at a low
glancing angle off the tabletop to the far wall. The spot of light moves slightly when
the heavy book is placed on or taken off the table. Again the teacher is anchoring
from the spring to the table so that the table acts as a “very stiff spring.” All through
this sequence the teacher is posing situations that prompt questions from the stu-
dents. Then the teacher presents situations that guide students to answer their own
questions. The sequence of questions and demonstrations proceeds along a line of
questioning that forms a context in which the teacher can now summarize the devel-
opment of ideas, why scientists say that even passive objects like tables can exert
forces.

Episode 7: Summarizing mini-lecture. At this point, Minstrell inserted a mini-

lecture, a brief exposition by the teacher that summarized the thinking so far and
emphasized the bridging connection that Minstrell hoped the students would make
between the various situations.

T: The question is whether we should also say that there is a force exerted
by the table. Now the physicist wanting to come up with a logical sort of
definition for force that makes rational sense as you go across different
situations says, “Look, I see something in common in these situations. This
book [on the hand] is at rest. So, if I want to say that here the book is at rest
and what keeps it at rest is a down force exerted by the Earth and an up force
exerted by the hand, and those two balance each other, well, shoot, I guess
that means that I better think of force as something that the table can do as
well. But I want to think of that as sort of a passive support that the table
does rather than as something really active muscular wise. If I want to be
logical about it, then I want the same kind of explanation here [points to
diagram of book on table] as what I have over here [points to diagram of
book on hand].”

In a follow-up interview, Minstrell explained what he was trying to do through

the set of episodes.

Part of what we’re trying to do in this activity is to focus on the pattern

recognition side of problem-solving or situation-solving…. So in this set of
episodes, we were essentially looking at the book on the table and bringing up

Everyday Assessment in the Science Classroom 67

CHAPTER 5

lots of issues out of that, and then we looked at the book on the hand and that
one was easy to understand for us, that the hand was exerting an upward force,
but can tables do that? Here, the important piece is that students then begin
looking at the patterns and saying, “Well, wait a minute … just like the hand
supports the book, something’s got to be supporting the book when it’s on the
table, so the table’s supporting the book, although passively, so we end up
concluding that our definition of force might best include passive sorts of
action.” The “at rest” becomes an important pattern—it’s “at rest” on the
table or it’s “at rest” on the hand or it’s “at rest” on a spring; these are all “at
rest” and “at rest” becomes the pattern that we’re recognizing … not whether
it’s on the hand or on the table but the fact that it’s “at rest” wherever it
happens to be. Understanding the pattern of balanced forces in an “at rest”
situation prepares students for distinguishing later between two frequently
confused situations: an object “at rest” and a tossed ball at the top of its arc,
which is not “at rest” even though its velocity is zero at one instant.

We use the term reflective discourse to represent the above type of classroom
interaction (van Zee and Minstrell 1997; van Zee et al. 2001). Reflective discourse
describes classroom interaction in which students as well as teachers are actively
engaged in monitoring their own and others’ thinking. We have defined reflective
discourse to be classroom talk in which (1) students typically express their own
thoughts in comments and questions rather than reciting textbook knowledge, (2) the
teacher and an individual student often engage in an extended series of questioning
exchanges, and (3) student-student exchanges sometimes involve one student trying
to understand the thinking of another. As discussed below, we have identified several
kinds of questions that occur during reflective discourse. “In addition to optimal
wait-time, it [questioning strategy] requires a solid understanding of the subject matter,
attentive consideration of each student’s remarks, as well as skillful crafting of fur-
ther leading questions” (Atkin, Black, and Coffey 2001, 35).

Purposes and Kinds of Questions Used in Learning Science

The central purpose of questioning is to ask or promote the asking of questions that
foster deep understanding rather than questions that ask for repetition of memorized
information and conclusions. For each sort of question described below, we will give
an example related to the discussion above and follow with a few other typical ex-
amples focused on that questioning purpose. The readers are encouraged to think of
other questions they have used that are consistent with each sort of question listed.

1. Ask or promote questions to open an inquiry and elicit students’ initial under-
standing and reasoning. The elicitation question just prior to Episode 1 in the first
part of this chapter asked students about the forces on a book on the table, even
before they had studied that situation or topic. “Even though we have not yet studied

68 National Science Teachers Association

 CHAPTER 5

this topic formally, I want to know what ideas you have at this point in time. You will
not be graded on whether you are right or wrong. That is not the point of this exer-
cise. Use words and diagrams to answer the question ‘What force or forces would be
acting on the book while it is at rest on the table?’ Briefly explain how you decided.”
2. Ask or promote questions to interpret and make sense of data in order to
generate new knowledge and understanding. In Episode 7 the teacher is modeling
the sort of thinking and questioning of oneself that he wants students to do as they
attempt to make sense of the results of their investigations. Examples of these sorts
of questions include the following: “In what ways are all these situations the same?
How might I explain all of them the same way?” “Look at the data you generated.
What patterns or relationships do you see in the data?” “What can you conclude
from your observations?”
3. Ask or promote questions to clarify or elaborate on observations and infer-
ences. If students are not encouraged to be clear in their verbal expression of ideas,
there may be a tendency for the teacher to think the student has the same understand-
ing as the teacher or another student and the speaking student may get an invalid
assessment. In Episode 3 the teacher asked for clarification: “How are you conceiv-
ing of force that’s different from the way some of the other people are conceiving of
force?” Another more general form of questioning for clarification or elaboration is,
“Say more—can you be more specific about what you observed? What did you see
in the container as each new chemical was added?” Also, teachers and students need
to clarify terms that have been used—for example, “What do you mean by the term
_____ which you just used?”
4. Ask or promote questions to encourage learners to justify their answers and
conclusions or to explain their reasoning to go beyond a mere stating of an answer.
An example is in Episode 3 when Sue asked, “How can a table have [exert] a force?”
Other examples of asking for justification include the following: “Why do you think
that variable will increase?” “Briefly explain how you decided that would happen?”
“How do you know?” “Why do you believe [that conclusion]?”
5. Ask or promote questions to extend or apply learned ideas. Since the context
above was to develop new understanding, there are no examples of applying what
has been learned, but after Episode 7 such a question could have been asked—for
example, “Given the argument I just summarized, how would you apply those ideas
to the forces acting on a person sleeping on a mattress as compared with a person
sleeping on a cement patio?” Other examples of application questions include the
following: “Given the relationship we just concluded, what would happen to the
output variable if we doubled this input variable?” “What are some other situations
in which you think this idea might apply? Say how it applies.” “What would happen
if …?”
6. Ask or promote questions that help learners monitor their own learning. In
Episode 1 the teacher asked, “Does it make sense to everybody that the table is
exerting an upward force on that book? No? Want to argue against it?” Other ex-

Everyday Assessment in the Science Classroom 69

CHAPTER 5

amples of promoting the monitoring of learning include the following: “So, how
have your ideas changed as a result of our activities of the past class period [past
week]?” “What do you not yet understand? What doesn’t make sense to you yet?”
“What would you need to do to test your idea (or answer your question)?”
Although the sorts of questions above are listed separately, many questions will
have dual purposes. For example, a particular question may ask students to clarify
their ideas, while at the same time asking for justification of knowledge. These are
the sorts of questions that can help teachers assess student’s ability to reason and
learn from experiences as well as assessing their science content knowledge. The
questions could be used either in verbal interaction with individual students or with
whole classes, and most could also be used in written assessments of students, such
as in quizzes or in focused writing in journals.

Using Questions and Answers to Inform Instruction

Listening to or reading students’ responses to questions gives assessment informa-
tion that can inform instruction. This part of the chapter will be in first person by the
first author, whose experience is described here.
Early in my teaching I did not address students’ thinking. For example, I taught
about explaining the “at rest” state of objects by demonstrating with a cart that if I
pushed horizontally in one direction on the cart, the cart would move (accelerate)
unless I also pushed with an equal force on the cart in the other direction with my
other hand. This situation was completely consistent with the students’ belief that
forces can be exerted by active objects. I fear that many of my students never did
learn why we believe that even passive objects can exert force.
Later, when I engaged in action research in my classroom, I became more inter-
ested in and responsive to learners and adjusted my instruction to their needs. Based
on what I learned from research about the problematic aspects of their ideas and
reasoning, I chose or adapted lessons to address their ideas and promote a deeper
understanding. Since there were almost no curricular resources that specifically ad-
dressed student thinking, I needed to adapt lessons, often on the fly. The lesson
described in this chapter is one of those lessons. I will describe here how what I was
learning through assessment was changing what I did in the classroom.
From the pre-instruction elicitation question I knew that about half the class
believed that passive objects didn’t exert force. I wanted to find some situation in
which they believed there would be an upward support force, a situation in which
they might see similarities with the book on the table. I chose the book on the out-
stretched hand. Even in this case students wanted to know if the hand was actually
outstretched or whether the person was sleeping with his or her hand resting on a
table, in which case they didn’t think the person would be exerting a force. So the
person was awake in my supposed situation of the book on the outstretched hand.
Since many of their explanations involved saying that the difference was that the
person was active, that the person was alive, I searched my mind for a situation in

70 National Science Teachers Association

 CHAPTER 5

which there was an obviously active object that was not alive. That was when I
thought of using the spring. I asked about the forces exerted by the spring. The
feedback from students suggested that they believed that while the spring was not
active, one could readily see that the spring was affected by the book. So, they seemed
to give up on the “live” aspect but remained tied to the need for some recognizable
“activity” of the object that was supporting the book. They were convinced that the
rigidity of the table did not reveal any level of “activity.”
Armed with this information about their thinking, I again searched my mind for
a way to show that the table was acting like a spring, that is, that the table could flex
as well. The very sturdy demonstration table I was using to demonstrate the book on
the table did have a shiny surface. Somehow I thought of glancing a light beam off
the tabletop. I hoped to use the tabletop as a “light lever.” I set the light beam at a low
angle so it would reflect off the tabletop to the far wall. I put the book on the table but
the movement was too small to be noticeable by the whole class. So, I sent one
student to the wall to tape a piece of paper on the wall where the light was hitting.
Then after marking where the edge of the light beam hit the paper without the book
on the table, I put the book on the table and the student observer noted that the beam
moved ever so slightly. I then climbed onto the table so the whole class could see the
movement of the beam. It was noticeable to everyone. I was more like the stack of
books.
From this series of situations the students themselves concluded that the table
was acting “like a very stiff spring.” They had moved from believing that only live
objects could exert force to active objects (that show deformation) exerting force to
the view that even rigid objects like tables could deform and exert force.
There were still a few students who continued to object, saying that “if the sup-
port object was very solid [rigid] like concrete or rock, there would be no deforma-
tion and therefore no force.” I again searched my mind for evidence they could see of
concrete or rock deforming under stress. I pulled out a piece of video of the Tacoma
Narrows Bridge Collapse, which has a segment showing the concrete roadway of the
bridge flexing and deforming in undulating wave patterns shortly before it broke
apart and fell into the water below. By now virtually all the students were willing to
hold tentatively true that all support structures, whether active or passive, could exert
force. The structures just showed it differently.
Now this whole sequence of assessment questions, discussion, and activities has
become part of my teaching curriculum and that of teachers with whom I work.
Thus, results of assessment can inform instruction and result in improved learning.
This and similar assessment tasks and related lesson activities can be found at http:/
/www.talariainc.com/facet (click on Diagnoser Tools).

Summary
Assessment can be ongoing with daily class activities such as elicitation activities,
discussions, demonstrations, and lab activities as well as quizzes and tests. The sorts

Everyday Assessment in the Science Classroom 71

CHAPTER 5

of questions that can reveal the most information are questions that ask for deep
thinking, that will elicit explication of understanding and reasoning. Finally, the re-
sults of assessment can inform instructional decisions.
One final point is very important: The use of these sorts of questions for ongoing
assessment and use of the assessment results for redirecting curriculum and instruc-
tion can be stimulating professionally. When in the classroom, I now wear two hats,
one as a teacher and another as a researcher studying my students’ thinking and how
to effect better learning. I can no longer teach without learning about my students’
thinking. The more I learn about my students’ thinking, the more I can tune my
instruction to help students bridge from their initial ideas toward more formal scien-
tific thinking. One of my teacher colleagues put it this way: “I used to think only
about the activities I was assigning to students. I now have to think about what is
going on in my students’ heads and then think about what I need to do as a teacher to
challenge problematic thinking and promote deeper thinking. I go home each day
tired out but energized about the improved effects I am having.” Rather than merely
serving students the activities from the book, we are first using questions to diagnose
their thinking. Then, we choose activities to address their thinking. Thinking in this
way about our work in the interest of improving our practice is part of what it means
to be professional. Teaching never becomes boring—quite the contrary. As teachers,
we can expect to be lifelong learners about our profession.

Acknowledgments
We are indebted to Professor Gerry Philipsen of the Department of Speech Communication at the
University of Washington for suggesting the term reflective discourse to describe the ways of speak-
ing that we recount in this chapter. A special thanks to our teacher colleagues who have allowed us to
work with them, shoulder to shoulder, to better understand teaching and learning.
The research reported in this chapter was supported by grants from the James S. McDonnell Foun-
dation Program in Cognitive Studies for Educational Practice. Preparation of this chapter was par-
tially supported by grants from the National Science Foundation.

References
Arons, A. 1983. Achieving wider scientific literacy. Daedalus 2: 91-122.
Atkin J. M., P. Black, and J. Coffey, eds. 2001. Classroom assessment and the national science educa-
tion standards. Washington, DC: National Academy Press.
Bransford, J. D., A. L. Brown, R. R. Cocking, eds. 1999. How people learn: Brain, mind, experience,
and school. Washington, DC: National Academy Press.
Camp, C., and J. Clement. 1994. Preconceptions in mechanics: Lessons dealing with students’ con-
ceptual difficulties. Dubuque: Kendall Hunt.
Clement, J., D. E. Brown, and A. Zeitsman. 1989. Not all preconceptions are misconceptions: Find-
ing “anchoring conceptions” for grounding instruction on students’ intuitions. Paper presented at
the annual meeting of the American Educational Research Association, San Francisco, CA.
diSessa, A., and J. Minstrell. 1998. Cultivating conceptual change with benchmark lessons. In Think-
ing practices in mathematics and science learning, eds. J. Greeno and S. Goldman. Mahwah, NJ:
Lawrence Erlbaum.

72 National Science Teachers Association

 CHAPTER 5

Minstrell, J. 1982. Explaining the “at rest” condition of an object. Physics Teacher 20(1):10–14.
———. 2001. The role of the teacher in making sense of classroom experiences and effecting better
learning. In Cognition and instruction: 25 years of progress, eds. D. Klahr and S. Carver. Mahwah,
NJ: Lawrence Erlbaum.
National Research Council (NRC). 1996. National science education standards. Washington, DC:
National Academy Press.
Rowe, M. 1974. Relation of wait-time and rewards to the development of language, logic and fate
control. Part one: Wait-time. Journal of Research in Science Teaching 11(2): 8l–94.
van Zee, E., and J. Minstrell. 1997. Reflective discourse: Developing shared understanding in a phys-
ics classroom. International Journal of Science Education 19: 209–28.
van Zee, E., M. Iwasyk, A. Kurose, D. Simpson, and J. Wild. 2001. Student and teacher questioning
during conversations about science. Journal of Research in Science Teaching 38: 159–90.

Everyday Assessment in the Science Classroom 73

 CHAPTER 6

Involving Students in Assessment

Janet E. Coffey

Janet Coffey currently works with middle school science teachers on the Classroom Assessment
Project to Improve Teaching and Learning (CAPITAL), a National Science Foundation–funded
effort that seeks to better understand the assessment practices of teachers and teacher change in
the dimension of assessment. She is particularly interested in students’ roles and perspectives in
assessment activity and the intersections of assessment and learning, both in and out of school
settings. She is a former middle school science teacher, worked at the National Research Coun-
cil as a staff memeber on the development of the National Science Education Standards, and
holds a Ph.D. in science education from Stanford University.

“Does anybody have any suggestions or comments?” asked Allison, a sixth

grader, at the end of a presentation of her investigation of the growth rate of
tomato plants given varying amounts of fertilizer.

“I like the way you described photosynthesis on the poster, and you explained
the data well. I still don’t understand what in the fertilizer affects the growth.
How come more wouldn’t be better?” was one reply from a classmate.
Another student commented on her eye contact with the audience and
suggested she use a louder voice when sharing her observations and data.
Subsequent comments pertained to presentation of data, level of detail in her
observations, and the model she provided of her design setup. Mark
complimented Allison on the manner in which she organized her data and
decided that he could present his own data in a similar manner for his study
on rates of erosion. The teacher asked Allison to explain how she designed
her experiment and if new questions arose in the process of her investigation.

After this class discussion, students completed a written evaluation of the

presentation, and the teacher provided feedback as well. Allison completed a
written reflection on her final product—what she had accomplished, what she
learned, what questions still lingered with her, and what she would do
differently next time. Allison would have ample opportunity to consider the
feedback from her peers and teacher before she completed her next
investigation, project, or presentation.

I n the vignette above, we see students actively engage in assessment-related activi-

ties—deliberating about quality work, reflecting on and revising work, complet-
ing evaluation sheets, and commenting on work done by their peers. In some form or
another, planned or unplanned, students have opportunities to participate in many of
these types of activities every day, yet these types of exchanges and interactions are
often not what first comes to mind when we think of assessment. Assessment is often
thought of as an event or strategy—a test, a performance activity, a portfolio, a re-

75
CHAPTER 6

sponse to a specific item. Whether it comes after teaching, while teaching, or by

teaching, we often think of assessment as something done to students, not with them.
These common conceptions of assessment obscure the roots of the word: “to assess”
derives from the Latin assidere, “to sit with.” Students engaged in and involved with
assessment reflect the term’s origins. As well, they represent a classroom assessment
culture that is supportive of student learning.
This chapter addresses the student in the realm of everyday assessment. After a
brief consideration of the research literature, I discuss a middle school science pro-
gram that prioritized student involvement in assessment. A closer look at the pro-
gram illuminates some of the supports that may help students as they become more
involved in their own assessment and as they learn the skills necessary to incorporate
assessment into their everyday learning. The chapter concludes with suggestions
about opportunities that exist for students to participate meaningfully in assessment.

Locating Students in Assessment Activity

From gauging student understanding to assigning grades, from planning the next set
of questions to communicating to parents, teachers use assessment for a wide range
of purposes, not the least of which is the support of student learning. Teachers
• check assignments and interpret student responses,
• listen closely to students’ questions so that they can gain insight into their stu-
dents’ understandings,
• seek to make explicit the assessment criteria so that all students know how they
will be evaluated,
• try to use what they learn through assessment to inform teaching, plan future
learning activities, and provide relevant feedback,
• constantly gauge trends in class engagement, interests, and understandings, and
• strive to fairly assign grades that accurately reflect what a student knows and is
able to do.
While all of these responsibilities certainly are important dimensions of effective
teaching, ways in which students themselves can participate in assessment are
less clear.
Everyday assessment is a dynamic classroom activity that includes the ongoing
interactions among teachers and students as well as more scheduled events, such as
weekly quizzes and unit tests. Attention to student involvement, then, becomes a
necessary aspect of assessment activity. Considered broadly, assessment encompasses
all activity wherein the quality of work or behavior is discussed, reflected upon,
examined, or established. In any classroom, students participate in an assessment
activity whenever they are involved in the examination of quality of work—whether
or not the teacher guides this activity and whether or not the focus is on actual stu-

76 National Science Teachers Association

 CHAPTER 6

dent work or on examples the teacher provides. The form this participation takes
varies enormously. Class discussions can center on what makes a piece of work
good, what could be improved, and what attributes students can use to improve their
own work. Student assessment also occurs when
• students appropriate an idea they saw in another class that made a presentation
compelling to them.
• they reflect on what they are learning, how it fits into a bigger picture, what they
still do not understand, and where they can seek extra help.
• they consider the thoroughness of their responses to test questions while they are
taking a test and when they review their graded exam as a class.
• they make a case to their teacher about why their response deserves more points
than she initially assigned.
While these types of activities occur with great regularity in many classrooms, the na-
ture and quality of the experience are worthy of deliberate attention and consideration.
One of the many purposes of everyday assessment is to facilitate student learn-
ing, not just measure what students have learned. As we note in the literature, key
features of assessment that improve student learning include explicating clear crite-
ria (Sadler 1989), improving regular questioning (Fairbrother, Dillon, and Gill 1995),
providing quality feedback (Butler 1987; Kluger and DeNisi 1996; Bangert-Drowns
et al. 1991), and encouraging student self-assessment (Sadler 1989; Wolf et al. 1991).
The importance of student involvement in assessment, the fourth feature, needs to be
underscored. While teachers can do much to assist their students, ultimately, the
student must take action (Black and Wiliam 1998). A teacher cannot learn for a stu-
dent. She cannot participate in activities in place of her students and expect her stu-
dents to benefit. When students play a key role in the assessment process they ac-
quire the tools they need to take responsibility for their own learning.
Research literature supports the idea of explicitly engaging students in assess-
ment. In a controlled study conducted by White and Frederiksen (1998), for ex-
ample, four middle school science classes received the same inquiry-based instruc-
tion and were given criteria for expectations and grading. Two of the classes used
time during class to regularly reflect on what they were learning and how they were
learning it (including using evidence from their work to support their evaluations).
The other two classes, the control classes, spent the same amount of time talking
about how the curricular activities could be modified. Although criteria were made
available to all four classes, the two experimental classes that spent time involved in
assessment-related discussions about learning performed better on project work and
on the unit test. Lower-performing students (as designated by their scores on the
California Test of Basic Skills) in the experimental class showed the greatest im-
provement in performance when compared to the control class. In light of the
National Science Education Standards’ (NRC 1996) underlying principle that sci-

Everyday Assessment in the Science Classroom 77

CHAPTER 6

ence teaching and programs should support all students in their quest for high perfor-
mance in science, the improvement seen across all levels in the experimental group
is significant.
Benefits of student participation in assessment are varied and far-reaching and
ultimately can contribute to improved student learning, as the study described above
suggests. Other advantages exist as well, perhaps related to or contributing to these
improved learning outcomes. Participation in assessment provides students with op-
portunities to discuss, contribute to, and develop an explicit understanding of what
constitutes quality work. In the process of such deliberation, students often generate
many of the salient educational goals themselves (Duschl and Gitomer 1997), which
can increase their commitment to achieving them (Dweck 1986). Participating in
assessment can also provide students with opportunities to reflect on what they are
learning in order to make coherent connections within and between subject matters
(Resnick and Resnick 1991). Brown (1994) stresses the strategic element of being
aware of particular strengths and weaknesses: “Effective learners operate best when
they have insight into their own strengths and weaknesses and access to their own
repertoires of strategies for learning” (9). To become a self-directed, lifelong learner,
an aim set forth in the National Science Education Standards (NRC 1996), the abil-
ity to self-assess becomes essential. A closer look at a program in which middle
school students participated in assessment shows how students can become involved
in assessment and benefit from doing so.

Learning from Connections

The scene of Allison and her classmates that opens this chapter represents a typical
scene in Connections,1 a middle school program that sought to engage students in
assessment as a meaningful part of their everyday classroom experiences. Assess-
ment of work was an explicit part of students’ classroom activity—doing it, talking
about it, learning how to use it to improve their work. Assessment-relevant conversa-
tions about public displays of student work, student reflections, peer evaluations,
goal setting, teacher and peer observations, and tests were everyday occurrences in
the program (Coffey 2002).
Through the students’ explicit participation in all aspects of assessment activity,
they arrived at shared meanings of quality work. Teachers and students used assess-
ment to construct the bigger picture of an area of study, concept, or subject matter

1
Connections is a semi-contained sixth-, seventh-, and eighth-grade program that sits within a large
public middle school in the San Francisco Bay Area. The students take many of their core academic
classes within Connections and join their peers from the larger school for electives, such as music,
physical education, art, and foreign languages. The eighth-grade students also take math and science
with their peers from the larger school. The author conducted extensive research in Connections
classrooms and is greatly indebted to the Institute for Research on Learning, formerly in Menlo Park,
California, for funding the study.

78 National Science Teachers Association

 CHAPTER 6

area. Student participation in assessment also enabled students to take greater re-
sponsibility and direction for their own learning.

Shared Meanings of Quality Work

One of the benefits of student participation in assessment is the emergence of a
shared understanding among all of the students of what constitutes quality work.
Rather than remain as a tacit understanding of teachers, or of savvy students who
have figured out expectations, when criteria become explicit, all students understand
what it takes to do “good” work. In any classroom, criteria for good work come from
somewhere. In Connections, they were informed by standards in the field and work-
place, by district guidelines, by years of the teachers’ experiences, as well as by
student experience, observations, and reflections about what looks and sounds good
in the work they produce and share in the classroom. With roots in all of these,
students and teachers negotiated what became high-quality work in these classrooms—
sometimes explicitly but often implicitly.
The conversation that Allison and her classmates had following her presentation
on the effects of fertilizer and the other activities that ensued highlight some of the
ways that this idea of quality became established. Let’s take a closer look at the
vignette that starts this chapter.
Allison’s presentation allowed her classmates to observe her organization, rep-
resentation, and explanation of data, and it allowed her to share with her classmates
what she had learned about plant growth. She also discussed her research design and
possible areas for further study. Following her presentation, students offered ideas
about what they liked (e.g., her description of photosynthesis) and what they thought
she could have done differently (e.g., speak in a louder voice). They also asked for
clarification, such as regarding the role that fertilizer plays in plant growth. Com-
ments about what her classmates thought was good—and their explanations of why
they thought this way—were heard and evaluated by everyone in the class. Likewise,
the teacher’s comments and questions focused attention on a certain area or set of
issues. Allison received feedback, as did the other students who observed her work
and followed the conversation.
All of the observations were not made public. Students could note something
they liked, or even didn’t like, and address those aspects in their own work. Mark
made note of her effective organization and display of data, which gave him ideas for
his own work. Written evaluations afforded students opportunity for further reflec-
tion on the quality of work they had just observed and provided Allison with addi-
tional feedback. In this case, Allison developed the written evaluation sheet the
students completed for her work as part of the assignment. In the process of con-
structing the evaluation sheet, she reviewed key areas that were important for feed-
back. She thought about questions to ask with regard to key ideas and, reflecting on
her explanations for the results, realized she still needed to consider other factors
that could have influenced growth besides the amount of fertilizer. The teacher’s

Everyday Assessment in the Science Classroom 79

CHAPTER 6

comments addressed the key areas Allison included on her sheet as well as ones the
teacher thought were worthwhile. She focused most of her remarks on content, an
aspect of the presentation in which students were not as confident in their comments.
All of the feedback could be considered and, if appropriate, incorporated into Allison’s
next piece of work. As work in this class improved, in part due to deeper understand-
ings of what makes something high-quality work, the starting point for these conver-
sations shifted.
These activities were not unique to this assignment. From the start of the year,
Allison, her classmates, and the teacher engaged in discussions about work, about
what makes something good, about what they liked, about what they thought could
be improved. In September, reflective conversations about work were established as
a class norm. Furthermore, students had numerous opportunities to observe and re-
flect on work prior to Allison’s presentation, and they understood that they would
have numerous opportunities to engage in similar assignments in the future.
No single activity led to shared meanings of what constitutes quality work. These
meanings emerged in a number of ways, including from
• students generating their own evaluation sheets,
• conversations in which students and teachers shared ideas about what consti-
tuted a salient scientific response, or a good presentation, lab investigation, or
project,
• discussions of an actual piece of student work,
• students’ reflections on their own work or a community exemplar, and
• students’ decision making as they completed a project.
These meanings were reinforced and revised as students evaluated their own and
others’ projects, used their new understanding to inform a piece of work, and adopted
a common way of talking about work with peers.
Detailed and in-depth understandings of quality did not appear immediately, nor
were they static. With frequent reflection and discussion, Connections students’ ideas
of what constituted a good project became more refined over time; ideas among
students began to converge. Conversations and comments evolved along with under-
standings of the criteria. Initially, students’ reflections and comments on their own
and their peers’ work focused primarily on process and presentation-related aspects
of work. We see some of these types of comments in the opening example, when
students respond to Allison’s request for feedback with remarks about her eye con-
tact and suggestions that she speak louder. Eventually, students improved in these
areas and less explicit attention was directed to them.
Gradually, with direction from the teacher, issues related to science content moved
to the foreground of reflection and discussion. Here, too, ideas evolved. While most
students know content is an important dimension of good work, particular aspects of

80 National Science Teachers Association

 CHAPTER 6

content that are important can take time to develop. In Connections, students began to
remark more frequently on content issues, such as clarity of information, organization
of information, source of information, and even accuracy of information. In the open-
ing vignette with Allison, we see evidence of this shift. One student’s remark, for
example, indicated that she did not understand the relationship between fertilizer and
plant growth. This remark also suggested that while students were listening to their
peers, they were reflecting on their understanding. This had not always been the case in
September. Another student’s comment accented the level of detail in Allison’s obser-
vations. Over the course of the year, students learned to ask questions about evidence
and explanation. Inclusion of “key ideas” and major concepts as goals for what content
to include and convey gradually replaced “a lot of facts.”
A shared idea of what makes good work developed in Connections through ex-
plicit attention to issues of quality in all stages of assessment and work. Even when
the teacher introduced criteria from the outside, such as in the form of a rubric,
criteria needed to be discussed and negotiated in order to become understood by and
operational for all students. A rubric the teacher distributed for a library research
assignment on erosion, for example, indicated that the top “score” of “5” required
that students include at least five facts about soil types. Many students spent time
counting the number of facts, without a real sense of why five facts were better than
four facts for this particular assignment or how to make the facts hang together for a
more convincing argument. On a related assignment, a rubric indicated to students
that a “5” required them to write a “clear hypothesis” for an investigation on absorp-
tion rates for soil samples. Many of the students did not know what constituted “clear”
for a hypothesis and only developed a notion of “clear” after class discussions and
numerous examples. In short, providing the students guidelines or criteria for evalu-
ation did not replace the need for an iterative process of meaning making that hap-
pened by explicitly attending to what makes good work—during day-to-day work
and before, during, and after project development and presentation.
Developing understandings of quality takes time and is an iterative and collabora-
tive process. Discussions do not end with a rubric or grading scale; they are ongoing and
evolving, and become increasingly sophisticated with time and practice. When a class
has shared definitions and standards for good work, students can more effectively par-
ticipate in an activity, as well as in the assessment of their own work and that of others.

Assessment as a Means to Connect to a Bigger Picture

Relationships and interrelationships between and among ideas, concepts, and activi-
ties are not always clear, especially early in a unit or at the beginning stages of an
investigation. Yet, an aspect of understanding a subject area or concept involves coming
to understand the relationships between what may appear as seemingly disparate
practices, activities, and information. For students in Connections, participation in
assessment helped forge these connections and establish a broader framework for
what they were learning.

Everyday Assessment in the Science Classroom 81

CHAPTER 6

Take, for example, a test review for the seventh-graders’ unit on the cell. Part of
the test review entailed a discussion comparing the parts of a cell, and the functions
of those parts, to different parts of a submarine. For some students, this culminating
exercise helped to establish the relationship between the parts and the whole. Or-
ganelles and their functions were considered together and in relationship to each
other, and discussion included the whole class. Although the teacher had attempted
to build in this broader framework throughout the unit, some students needed the
final review in order to truly make sense of what they had been learning. Another
aspect of the test review entailed small groups of students writing test questions,
some of which the teacher would use on the actual exam. These small-group discus-
sions served as further review (and helped to demystify the origins of test questions).
After the test was returned, the class went over their responses as a class, and stu-
dents had an opportunity to do test corrections for a homework assignment. In this
case, the activities that occurred before and after the test were as important for learn-
ing as the actual test itself. Conversations about the students’ presentations of their
investigations, of which Allison’s is just one in a series, provided other opportunities
to do much more than simply hear what classmates have been up to. The class dis-
cussed fundamental aspects of science as a practice, such as communication of data,
information, and conclusions. Students came to see that communication is an intrin-
sic aspect of scientific endeavors.
Opportunities to reflect, both as individuals and as a group, helped improve stu-
dents’ understandings and made clearer how what they were doing connected to a
broader topic or idea and, at times, to a larger world of science. Coupled with reflec-
tion, assessment-relevant conversation and activities gave students the opportunity
to synthesize information and experiences; the revision process helped them to clarify
criteria and to strengthen connections and analysis. Establishing a bigger picture
was a collaborative endeavor: Class discussions and shared reflections, for example,
allowed students access to connections others had made and provided data for teach-
ers about student learning. Making these connections explicit and the bigger picture
more apparent is particularly important in terms of reaching all students.

Assessment as a Vehicle to Facilitate Lifelong Learning

One aspect of becoming more self-directed in an activity involves identifying what
is not well understood. Learning is no exception. An important feature of participa-
tion is knowing when and where to go to for additional, and perhaps more expert,
help. The relationship between learning and assessment, particularly the timing of
assessment, changed in Connections, as students became intimately involved in the
assessment process. The role assessment played in the actual doing of the project
work changed, too. By the end of the school year, for example, seventh graders used
assessment as a way of seeing whether they needed to gather more data, whether
they had provided an adequate explanation of their data, and whether their conclu-
sions mapped back to their hypotheses. In the process of making her evaluation sheet,

82 National Science Teachers Association

 CHAPTER 6

and reflecting on key questions to ask, Allison recognized that she needed to revise
her own explanations about what happened. As students read material, they paused
to reflect on the main points, and as they listened to the presentation of material, they
reflected on their understanding, as was the case with the student who asked Allison
about fertilizer. As they planned investigations, they made sure their initial questions
were clear. Assessment shifted from something that largely followed learning—mea-
suring what had been learned—to something that occurred simultaneously with learn-
ing. This came to occur with such fluidity that the two became inextricable and
difficult to tease apart. For many students, assessment became a vehicle for asking
questions that would help clarify their understanding and help guide production of
good work. Students became more self-directed in their assessment as they learned
what questions to ask and what questions they should be able to answer.
Reflection became incorporated into the students’ everyday classroom activities,
whether prompted by the teacher, a peer, or themselves. In the busy lives of students
and schools, carving out the time for quiet reflection can be somewhat difficult, but
students grew to appreciate its value. As one seventh grader said, “When you think
about what you learn, you actually learn it. It soaks in more.”

Relationship among Students, Content, and Assessment

In response to a teacher’s request to reflect on certain subject matter, a student may
ask, “How can I assess what I don’t know?” Students are not the only ones who hold
this view. When teachers design ways to involve students in assessments, they may
overlook involvement in the content dimensions of work on the same premise—that
students will not be able to evaluate the accuracy of specific information. Relation-
ships among content, assessment, and students become critical issues as teachers
shift some of the responsibility for assessment to students. A trade-off, and some
would say tension, exists in involving students in the assessment of their work in the
realm of content.
With time and an increasingly clear idea about what is important to consider
with respect to content, students can become more self-directed with respect to as-
sessing content. Questions about evidence and sources can guide work, reflection,
and revision. Students can learn to look for, and ask themselves and each other about,
main points and key concepts. If students ask themselves questions such as, “What is
the evidence?”, “What do we know?”, “How do we know it?”, and “Where does this
information come from?”—and if they pay attention to their subsequent responses—
they are becoming familiar with their own understanding of the content, and they
can move on to ask questions about their peers’ understanding. They can develop a
better sense of when they do not really understand something and, importantly, when
to look for information as well as where and to whom to go for that information.
Although students may learn to consider content, and develop tools for doing so,
they still may not be able to evaluate the soundness or accuracy of an explanation on,
say, photosynthesis. Teachers play an important role in involving students actively in

Everyday Assessment in the Science Classroom 83

CHAPTER 6

assessment, but the challenges associated with involvement are not obstacles that
cannot be overcome. Assessment-relevant conversations related to content help cul-
tivate within students the types of questions they should be asking when dealing with
content information. In addition to commenting on content in terms of accuracy and
soundness, teachers can facilitate conversations related to content, highlighting as-
pects of content and asking questions about it in their own remarks.

Creating Meaningful Opportunities for Assessment

If students are to take greater responsibility for their own learning and assessment,
they must be afforded opportunities to participate meaningfully in assessment. The
Connections program offered lessons in six areas about how to organize and support
student participation.
1. Time. Involving students in assessment activity takes time. Teachers in the
Connections program prioritized assessment as an important activity by providing
time and opportunities for students to participate. They traded time from other teach-
ing activities so students could do assessment. They guided and provided scaffolding
for student involvement. They contributed their own feedback and determined when
their remarks were most relevant and when to back off to let student contributions
carry more weight. Students, likewise, invested time and energy into providing feed-
back to peers, commenting on work, contributing to assessment conversations, and
reflecting on work.
2. Use of Traditional Assessment. Many of the assessment opportunities in Con-
nections arose in exchanges among students and teachers and between and around
the more visible assessment events. A student’s presentation or a test, and episodes
preceding and following these events, provided opportunities for the class to discuss
and establish aspects of quality work, to establish the role that assessment would
play in the classroom, to generate the norms and guiding assessment principles that
operate in the classroom, and to use criteria to inform work in process.
Teacher and students leveraged test review as an opportunity to return to the
bigger picture of what they had been studying. The class talked about what was
going to be covered on the test or quiz so that all students knew what to expect. For
some, these discussions provided opportunities to make connections, if they had not
been made before. Discussions about tests also served as springboards to talk about
the meaning of assessment. The test process also encompassed graded responses
after the test, and students would often do test corrections after going over the test.
On occasion students would write test questions and grade their own work.
3. Public Displays of Work. During the initial weeks and months of school, teach-
ers used public displays of student work to catalyze conversations about good work,
to prompt student reflection, and to promote peer feedback. At times, work done by
a student from a previous year or work done by a teacher or another adult served as
the basis of these conversations. Immediately following presentations of work, teachers
asked students to consider what was successful and effective and what could be

84 National Science Teachers Association

 CHAPTER 6

improved and how. Such discussions about quality of work and suggestions for im-
provement were established as a classroom norm. Gradually teachers asked students
to provide written evaluations on work. These written evaluations augmented the
discussions and provided additional opportunities to cultivate and support reflection,
as well as additional avenues for feedback.
4. Reflection.Teachers asked students to reflect on dimensions that should be
included in upcoming work. They asked for volunteers to share thoughts publicly.
These conversations often addressed general features to consider in a piece of work
or presentation; they also offered students the opportunity to think systematically
about quality and about their own work with respect to quality. The class reflected,
both publicly and privately, after doing their work. Gradually, students were encour-
aged to reflect during work as well. The integration of reflection into daily practice
helped students grasp “the big picture.”
5. Revision. Students had opportunities (and expectations) to revise their work.
They did this with peers and teachers as they went along, submitting proposals and
meeting to discuss their progress. Often the teacher would request that a student re-
do something, taking into account the feedback that he or she had received.
6. Goal Setting. Another dimension of the assessment activities involved goal
setting. Students set short- and long-term goals and specific and broad goals. Through-
out the year, they reflected on progress made toward these goals and set new goals.
For project work, daily and weekly goals helped students to prioritize big ideas and
to keep themselves focused during work time. By reflecting on progress toward their
goals, students considered their current work in light of where they were headed.
With time and attention, the teacher could help students increase the specificity of
their goals in different subject areas. Longer-term goals served as the centerpiece of
student-led student-teacher-parent conferences.

Learning to Assess and Learning How to Use Assessment

At times, especially early in the school year, opportunities to participate in self-
assessment were not immediately embraced. Students would ask questions like, “Isn’t
this your job?” when asked to reflect on their progress. Students resisted requests for
written reflection on their work and struggled to come up with evidence for accom-
plishments or progress. And when they did participate in class discussions, student
contributions tended to be incomplete and at times even superficial. At the start of
the year, students had difficulty identifying key components of a good lab report,
essay, or response.
The quality of student participation in assessment improved in Connections as
students learned and practiced the necessary skills to participate meaningfully. With
a teacher’s help, students learned how to reflect on work, both for a single occasion
and on work over time; they learned how to comment constructively on peers’ work;
they learned what constituted quality work; they learned how to use assessment in
ways that informed their work in process. Students became better at assessing as

Everyday Assessment in the Science Classroom 85

CHAPTER 6

they had opportunities to assess. Despite initial resistance, as students learned as-
sessment-related skills, demarcations between roles and responsibilities with respect
to assessment blurred. They learned to take on responsibilities and many even appro-
priated ongoing assessment into their regular habits and repertoires.

Conclusion
To create a culture of assessment that promotes powerful and productive student
learning, the roles students play in the assessment process need to be re-examined.
Their essential role in the process—as partners, not just as recipients—needs to be
recognized. Numerous opportunities exist for students to participate, but they need
to be supported in their efforts to do so.
With support and guidance from the teacher, student involvement can grow and
build over the course of the year. As students learn how to participate, their involve-
ment shifts to become more integrated and self-directed. High levels of student in-
volvement take time and investment, on the part of both teacher and students, and
will not happen overnight. Any single assessment event or episode cannot be iso-
lated from the whole of the activity. Student engagement with and participation in
any one event seeps into and informs the others. Student-developed evaluation sheets,
as well as responses, for example, are informed by ongoing deliberations about what
constitutes quality work.
Lessons from Connections and the research literature show that cultivating a
productive assessment culture becomes compelling. As students learn to engage in
and with assessment, assessment can be incorporated into everyday work habits, not
just at the urging of the teacher, but as a productive and inextricable dimension of
learning, and these habits can extend well beyond the school walls.

References
Bangert-Drowns, R. L., C-L Kulik, J. A. Kulik, and M. T. Morgan. 1991. The instructional effect of
feedback in test-like events. Review of Educational Research 61 (2): 213–38.
Black, P., and D. Wiliam. 1998. Inside the black box: Raising standards through classroom assess-
ment. London: King’s College. Also published in Phi Delta Kappan (1998) 80 (Oct.): 139–48.
Brown, A. L. 1994. The advancement of learning. Educational Researcher 23 (8): 4–12.
Butler, R. 1987. Task-involving and ego-involving properties on evaluation: Effects of different feed-
back conditions on motivational perceptions, interests, and performance. Journal of Educational
Psychology 79 (4): 474–82.
Coffey, J. E. 2002. Making connections: Students engaging in and with assessment. Doctoral diss.,
Stanford University.
Duschl, R. D., and D. H. Gitomer. 1997. Strategies and challenges to changing the focus of assess-
ment and instruction in science classrooms. Educational Assessment 4 (1): 37–73.
Dweck, C. S. 1986. Motivational processes affecting learning. American Psychologist, Special issue:
Psychological science and education 41 (10): 1040–48.
Fairbrother, B., J. Dillon, and G. Gill. 1995. Assessment at Key Stage 3: Teachers’ attitudes and prac-
tices. British Journal of Curriculum and Assessment 5 (3): 25–31.

86 National Science Teachers Association

 CHAPTER 6

Kluger, A. N., and A. DeNisi. 1996. The effects of feedback interventions on performance: A historical
review, a meta-analysis, and a preliminary feedback intervention theory. Psychological Bulletin
119 (2): 254–84.
National Research Council (NRC). 1996. National science education standards. Washington, DC:
National Academy Press.
Resnick, L. B, and D. P. Resnick. 1991. Assessing the thinking curriculum: New tools for educa-
tional reform. In Changing assessments: Alternative views of aptitude, achievement and instruc-
tion, ed. B. Gifford. Boston, MA: Kluwer.
Sadler, R. 1989. Formative assessment and the design of instructional systems. Instructional Science
18: 119–44.
White, B. Y., and J. R. Frederiksen. 1998. Inquiry, modeling and meta-cognition: Making science
accessible to all students. Cognition and Instruction 16 (1): 3–118.
Wolf, D., J. Bixby, J. Glen III, and H. Gardner. 1991. To use their minds well: Investigating new forms
of student assessment. Review of Research in Education 17: 31–74.

Everyday Assessment in the Science Classroom 87

 CHAPTER 7

Reporting Progress to Parents and Others:

Beyond Grades
Mark Wilson and Kathleen Scalise

Mark Wilson is a professor of education at the University of California, Berkeley, specializing in

educational assessment, educational evaluation, and applied statistics. He holds a B.Sc. (math-
ematics), teaching certificate (elementary education), and master’s degree (research methodology)
from the University of Melbourne. He completed a Ph.D. in measurement from the University of
Chicago. Extensively published in professional journals and books, his work involves (a) the devel-
opment of new psychometric models for analyzing student performance in theory-rich contexts
such as complex performance assessments and intelligent tutoring systems, (b) the application of
current measurement theory to practical problems of testing and evaluation, and (c) the develop-
ment of assessment resources that are useful to classroom teachers, policy makers, and other edu-
cation professionals.

Kathleen Scalise is a Ph.D. student in quantitative measurement at the University of California,

Berkeley. She was the science writer for the high school physics, chemistry, biology/life sciences,
and Earth sciences chapters of the 2002 California Science Framework for K–12 Public Schools,
produced by the Curriculum Frameworks and Instructional Resources Division of the California
Department of Education. She holds a bachelor’s degree in biochemistry from the University of
California, Berkeley, and holds teaching credentials in both physical and life sciences.

W hen it comes to grades and grading practices in the United States, there is one
thing parents often don’t understand and students understand all too well: Get-
ting an A, B, or C, even passing or failing a class, is often based more on a teacher’s
perceptions of a student’s effort than on accomplishment.
This “secret of current classroom grading,” said one middle school teacher during
a recent online gripe session, “frustrates many teachers as they think about holding
students to high standards. It’s hard to be tough about standards when a student says to
you, ‘But I tried so hard.’ But we’ve got to help teachers get past this, because we all
deal with the repercussions when we judge students mostly on effort” (Berry 1997).
What are the repercussions when such grades, and only such grades, are re-
ported to parents and students as the final summation of success in a class? Students
think, as would be expected, that they have satisfied course objectives to the extent
of the grade and so do their parents, assuming the grade is based on mastery of
content, strategies, or other learning goals that may not have been met at all. Later
when students encounter more challenging work for which they are not prepared,
failure sits on their shoulders as a lack of ability, talent, or some other hard to quan-
tify but definitely personal characteristic; surely it cannot be lack of preparation
when they have done so well in prerequisites.
A single score in the form of a grade can do little to inform mastery of today’s
complex science standards, consisting of multiple strands and numerous objectives.
Add to this the confounding effect of incorporating effort, work ethics, and other con-
cerns into a grade, on a basis that differs from class to class, and such a score can be all

89
CHAPTER 7

but meaningless in terms of the “rigorous and wise diagnostic information” (Wolf et al.
1991, 31) called for by educators. In the end, when students are trained to work hard
but not effectively, and standards are taught but not met, damage can be inflicted on
students’ opportunities and on their perceptions of themselves and their abilities.
Grades, however, have a long tradition of being awarded on the basis of many
considerations and not outcomes alone. Surely the final value judgment of a student’s
performance in a class should consider many dimensions, including, but not limited
to, outcomes. Teachers as experts in their disciplines have long had jurisdiction to set
a suitable balance of factors in awarding grades.
And so we see our grading practices have left us perched in a difficult position:
What is to be valued and what isn’t, and how does any of this relate to standards-
based education or any type of education in which learning goals are considered
important to assess and communicate?
We argue here that both sides of this grading dilemma can be satisfied. If teach-
ers and students were given robust but easy-to-use tools for generating solid diag-
nostic information and relating this information to standards or learning goals, they
then would have the power to better understand progress toward achievement. Teachers
could retain the freedom to decide how to value this information toward assigning a
grade. Parents and students could share in diagnostic information while also getting
feedback on the larger classroom picture in the form of a grade.
In this chapter, we will discuss the principles of constructing classroom-based
assessment tools that can generate high-quality and interpretable evidence for teach-
ers, students, and parents, and we show how these tools can be built into curriculum
materials. A new “embedded assessment” system designed at the University of Cali-
fornia, Berkeley, called the BEAR Assessment System (Wilson and Sloane 2000),
will be introduced as a generic approach that can be customized to the needs of many
courses. The system was named for its origin at the Berkeley Evaluation and Assess-
ment Research (BEAR) Center and is a comprehensive, integrated system for as-
sessing, interpreting, and monitoring student performance. It provides a set of tools
for teachers and students to use to
• reliably assess performance on central concepts and skills in a curriculum,
• set standards of performance,
• validly track progress over the year on the central concepts, and
• provide feedback on student progress and effectiveness of instructional materi-
als and approaches for teachers, students, parents, administrators, and other au-
diences.

A Word about Embedded Assessment

The term embedded assessment means what it says: assessment activities are “em-
bedded,” or become part of, classroom learning activities, indistinguishable from

90 National Science Teachers Association

 CHAPTER 7

day-to-day operations. Teachers have been doing embedded assessment for a long
time: a homework assignment, for example, is embedded assessment, as are a lab
practical, a classroom discussion in which students first jot down their thoughts on
the topics, and a short-answer essay. Any of these and much more can be considered
an embedded assessment activity if a student produces something that can be scored
or rated or if the student can be observed and assessed in some manner. The only
difference between these examples and what we discuss here as more formal embed-
ded assessment is that the latter calls for attention to task design and formal “calibra-
tion” of assessment tasks in relationship to a framework that describes the learning
to take place and that can be used to generate interpretable, valid, and reliable diag-
nostic information on student performance.
We argue that embedded assessment is desirable not only because it reflects
what a student is actually being taught but also because it can be a learning tool in
and of itself. If an assessment task is also a learning activity, then it does not take
unnecessary time away from instruction, and the number of assessment tasks can be
increased in order to improve the results of measurement, diagnostics, and account-
ability (Linn and Baker 1996).
The potential usefulness of embedded assessments can be greatly enhanced
when the framework on which they are based is consistent with that for the more
formal assessments used in accountability assessments, such as school district or
state assessments. This also greatly enhances the value of the more formal assess-
ments (for a discussion of this point, under the topic “assessment nets,” see Wilson
and Adams 1996).

What Is Needed in an Assessment System: Four Principles

With the accountability debate raging, it is hard to imagine principles of assessment
that would be widely agreed upon by all. However, a few central concepts seem to
stand out. Here we first develop and then address four principles that we believe any
assessment system designed to be useful in the classroom must uphold: a develop-
mental perspective, the generating of quality evidence, a match between instruction
and assessment, and management by teachers.

Principle 1: A Developmental Perspective

A developmental perspective of student learning means assessing the development
of student understanding of particular concepts and skills over time, as opposed to,
for instance, making a single measurement at some final or supposedly significant
time point. Traditional classroom assessment strongly supports a developmental per-
spective. Every teacher knows he or she must make repeated measurements of stu-
dent progress, often done with tests and quizzes, as instruction proceeds. Standardized
testing situations and educational research, however, often neglect this perspective.
“One-shot” testing situations and cross-sectional approaches—comparing two sets
of students at different times as if they were one set proceeding from one time to the

Everyday Assessment in the Science Classroom 91

CHAPTER 7

next—abound in education. Here we affirm what is perhaps the obvious: For diag-
nostic information to be diagnostic, it must be collected over time and in relationship
to some set of goals about what is to be learned.
However obvious, it is important to emphasize this perspective because of its logi-
cal implications. To truly understand and interpret the significance of development
requires a model of how to interpret data taken at these repeated intervals. Clear defini-
tions of what students are expected to learn, and a theoretical framework of how that
learning is expected to unfold as the student progresses through the instructional mate-
rial, are necessary elements to a well-grounded developmental perspective.

Principle 2: Quality Evidence

For classroom-based assessment procedures to gain “currency” in the assessment
community, issues of technical quality have to be addressed. Despite criticism of
standardized tests based on issues such as the impoverished nature of questions and
the often poor match between instructional goals and what is assessed, standardized
tests are based on extensive background and technical work on issues such as reli-
ability—the reproducibility of results—and validity—measuring what is intended to
be measured. Standardized exams also are vetted for bad questions; in addition, they
are tested prior to use and their interpretability is enhanced by being normed.
These issues have typically been ignored in classroom-based assessment until
now. Despite the plea of Wolf et al. (1991), the development of practical procedures
for establishing the technical quality of classroom-based assessments lags far behind
high-stakes assessment programs. To ensure comparability of results across time and
context, these issues must be addressed in any serious attempt at a developmental
assessment system for use in the classroom.
Additionally, classroom-based assessments should maintain standards of fair-
ness, such as consistency and absence of bias. If “open-ended” or “performance-
based” tasks are used, for instance, they require procedures to ensure fairness and
consistency of scoring (as well as fairness and consistency in collecting and manag-
ing student work). Responses can no longer be scanned by machine and entered
directly into a database. Raters score the work, and besides fairness this also raises
issues of time and cost.
It is a long-accepted truism that teacher-made (i.e., “classroom-based”) student
assessments have greater “curricular” or instructional validity—that is, they relate
more to actual learning aims—than standardized tests but that they do not have the
strong technical properties of carefully constructed standardized assessments. We
assert that classroom-based diagnostic assessments must have both. While teachers
will and should continue to construct “teacher-made tests” despite not having the
opportunity to establish the comparability or validity of these tests, appropriate class-
room-based assessment procedures can and should be developed on the basis of
educational standards and made available for teacher use and adaptation.

92 National Science Teachers Association

 CHAPTER 7

Principle 3: Match between Instruction and Assessment

The third principle of assessment is establishing a good match between what is taught
and what is assessed. Reports abound of teachers interrupting their regular curricular
materials in order to “teach the material” students will encounter on district- or state-
wide tests. Traditional testing practices—in high-stakes or standardized tests as well
as teacher-made tests—have long been criticized for using too many questions that
assess only memorized facts and ignore more complex levels of understanding. As
Resnick and Resnick (1992) argued, “Assessments must be designed so that when
teachers do the natural thing—that is, prepare their students to perform well—they
will exercise the kinds of abilities and develop the kinds of skill and knowledge that
are the real goals of educational reform” (39). Assessments that don’t meet the goals
of instruction fail this test.

Principle 4: Management by Teachers

This final principle is perhaps most critical: Classroom-based assessment systems
must be under the jurisdiction of teachers. Teachers must be the managers of the
system. They must have the tools to use the system efficiently and to explain result-
ing data effectively and appropriately. Students must also be able to participate in the
assessment process, and they should be encouraged to develop essential metacognitive
skills that will further the learning process.
There are two broad issues involved in this principle. First, it is the teachers and
students who will use the assessment information to inform and guide the teaching
and learning process. By the very nature of turnaround time, assessments conducted
as part of district- or statewide accountability programs—no matter how valid or
appropriate to what is taught in the classroom they may or may not be—cannot
provide the immediate feedback that teachers and students need for instructional
management and monitoring (Haney 1991; Resnick and Resnick 1992). For this
function of assessment, teachers and students must be
• involved in the process of collecting and selecting work for assessment,
• able to score and use the results immediately, not wait months for scores to be
returned,
• able to interpret the results in instructional terms, and
• able to have a creative role in the way assessment is realized in their classrooms.
Second, issues of teacher professionalism and teacher accountability demand
that teachers play a central and active role in collecting and interpreting evidence of
student progress and performance (Tucker 1991). If teachers and students are to be
held accountable for performance, they need a good understanding of what students
are expected to learn and of what counts as adequate evidence of student learning.
Teachers are then in a better position, and a more central and responsible position,
for presenting, explaining, analyzing, and defending their students’ performances

Everyday Assessment in the Science Classroom 93

CHAPTER 7

and outcomes of their instruction. Students are better able to develop their own
metacognitive skills and to bring them to bear in the learning process.
This perspective requires new views of the teaching and learning process, new
roles for and demands on teachers, perhaps even a new “assessment culture” in the
classroom (Brown et al. 1992; Cole 1991; Resnick and Resnick 1992; Torrance 1995a,
1995b; Zessoules and Gardner 1991). Preparing teachers and students to use these
types of assessments in the classroom may be a difficult challenge. But we should
recognize that teacher understanding and belief in assessment often will determine
the ultimate success of change (Airasian 1988; Stake 1991).

The BEAR Approach

An assessment approach that can meet the tests of these principles while remaining
adaptable to the needs of each course in which it may be used is challenging to
design. A number of different approaches might work. Here we present solutions
developed through our work with science curriculum developers and embodied in
the BEAR Assessment System (for more detail, see Wilson and Sloane 2000).

Satisfying Principle 1: A Developmental Perspective

Our strategy to address this issue is to develop a set of “progress variables” (Masters,
Adams, and Wilson 1990; Wilson 1990) that mediate between the level of detail that is
present in the content of specific curricula and the necessarily more vague contents of
state standards and curriculum framework documents. These progress variables are the
core of the generic assessment system that we have developed to help teachers make
the best use of the information contained in assessments. Progress variables define the
intended content of a specific curriculum up to a level of detail that would allow, say,
biweekly tracking of student progress through the curriculum. These variables define
the most important student growth goals of the curriculum; the variables change from
course to course as different areas of knowledge are assessed.
Every instructional unit is seen as contributing in some way to student progress
on at least one of these variables, and every assessment is closely aligned with one or
more of these variables. This alignment allows the creation of a calibrated scale to
map the growth of students, so that teachers can track the progress of individual
students and groups of students as they engage in learning and instruction. This idea
of a “cross-walk between standards and assessments” has also been suggested by
Eva Baker (Land 1997, 6). These variables create a “conceptual basis” for relating
the curriculum to standards documents and to other curricula.
In this approach, the idea of a progress variable is focused on the concept of
progression or growth. Learning is conceptualized not simply as a matter of acquir-
ing quantitatively more knowledge and skills, but as progress toward higher levels of
competence as new knowledge is linked to existing knowledge and deeper under-
standings are developed from and take the place of earlier understandings. The con-
cepts of ordered levels of understanding and direction are fundamental in our approach.

94 National Science Teachers Association

 CHAPTER 7

To use the BEAR Assessment System in any given area it is assumed that learning
can be described and mapped as progress in the direction of qualitatively richer knowl-
edge, higher-order skills, and deeper understandings.
Variables are derived in part from professional opinion about what constitutes
higher and lower levels of performance or competence, but are also informed by
empirical research into how students respond to instruction or perform in practice.
To more clearly understand what a progress variable is, let us consider an example:
progress variables recently designed for a high school chemistry curriculum, “Liv-
ing by Chemistry: Inquiry-Based Modules for High School” (Claesgens et al. 2002).
The Living by Chemistry (LBC) project at the University of California’s Lawrence
Hall of Science is a National Science Foundation–funded, yearlong course to teach
chemistry using real-world contexts familiar and interesting to students. The course
developers were interested in this approach to assessment for at least two reasons.
First, they wanted to reinforce the conceptual understanding and problem-solving
aspects of the course. Traditional “fact-based” chapter tests would not reinforce these
aspects, they believed, and could direct the primary focus of instruction away from
the most important course objectives. Second, the developers wanted to explore learn-
ing trajectories in high school chemistry, gather reliable data on which curricular
approaches best facilitated student learning, and understand for whom these ap-
proaches succeeded and failed, and why.
Following the Developmental Perspective principle, we worked with the LBC
curriculum development team to devise a framework of progress variables, called
“Perspectives of Chemists,” that attempt to embody understanding of chemistry from
a novice to expert level of sophistication and that incorporate the national and Cali-
fornia state science education standards. Three sets of perspectives variables, or
strands, were designed to describe chemistry models and views regarding three “big
ideas” in chemistry: matter, change, and energy. The matter strand is concerned with
describing atomic and molecular views of matter, as well as measurements and model
refinement regarding matter. Change involves kinetic views of change and the con-
servation of matter during chemical change. Energy considers the network of rela-
tionships in conservation and quantization of energy.
Each strand consists of an atomic viewpoint on the topic under consideration and
a macroscopic, or systems, viewpoint.1 The LBC framework is shown in Figure 1.
While too complex to be considered in full here (see Claesgens et al. 2002 for more
details), to illustrate the use of progress variables we will consider assessment data
collected on one portion of the framework: the progress variable representing the atomic

1
While for expert chemists the atomic and systems views are integrated into a single perspective,
conceptual change literature in chemistry suggests that a major source of misconceptions for novices
involves confusion of atomic and systems properties—for instance, considering a molecule of water
to be tear-shaped, like a tiny drop of water, and a molecule of ice to have the geometry of an ice cube
(Griffiths and Preston 1992).

Everyday Assessment in the Science Classroom 95

96
Figure 1. Perspectives of Chemists
CHAPTER 7

6 ATOMIC AND MEASUREMENT AND REARRANGEMENTS OF CONSERVATION OF CONSERVATION OF QUANTIZATION OF

Understanding MOLECULAR VIEWS MODEL REFINEMENT ATOMS MASS ENERGY ENERGY

5 Bonding and Models and Kinetics and Stoichiometry and Particle and Spectroscopy
Integrating relative reactivity evidence changes in bonding equilibrium energy views and structure

4 Phase and Limitations of Products of Amounts of Degrees of Electronic

Predicting composition models reaction products change structure

3 Properties and Measured amounts Change and Amount of reactants Energy transfer Color with
Relating atomic views of matter reaction types and products and change light absorption

2 Matter with Mass with a Change with Change with a Heat and Energies associated
Representing chemical symbols particulate view chemical symbols conservation view temperature with light

1 Properties of Amounts of Attributes of Changes in Measures of Light

Describing matter matter change mass energy

Levels of A B C D E F
success Visualizing matter Measuring matter Characterizing Quantifying Evaluating Quantizing
change change energies energy

Source: Claesgens, J., K. Scalise, K. Draney, M. Wilson, and A. Stacy. 2002. Perspectives of chemists: A framework to promote conceptual under-
standing of chemistry. Paper presented at the annual meeting of the American Educational Research Association, New Orleans.

National Science Teachers Association

 CHAPTER 7

and molecular view of matter, called “Visualizing Matter.” Note that each progress
variable was based first on expert opinion of likely progressions of learning and then
refined and modified with extensive collection of data, which in this case generated
some findings that surprised the developers.
The Visualizing Matter variable incorporates California state science education
chemistry standards 1c, d, g; 2a, e, f, g; and 10b, d, e (see Appendix). It shows how
a student’s view of matter progresses from a continuous, real-world view, to a par-
ticulate view accounting for existence of atoms and molecules, and then builds to
relate this view to more complex behaviors and properties of matter.
Our assessments with pilot studies of this variable show that a student’s atomic
views of matter begin with having no atomic view at all, but simply the ability to
describe some characteristics of matter, such as differentiating between a gas and a
solid on the basis of real-world knowledge of boiling solutions perhaps encountered
in food preparation, for instance, or bringing logic and patterning skills to bear on a
question of why a salt dissolves. This then became the lowest level of the Visualizing
Matter variable in the LBC framework.
At this most novice level of sophistication, students employ no accurate molecu-
lar models of chemistry, but a progression in sophistication can be seen from those
unable or unwilling to make any relevant observation at all during an assessment
task on matter, to those who can make an observation and then follow it with logical
reasoning, to those who can extend this reasoning in an attempt to employ actual
chemistry knowledge, although incorrectly in first attempts. All these behaviors fall
into Level 1, the Describing level of the framework, and are assigned incremental 1-
and 1+ scores, which for simplicity of presentation are not shown in this version of
the framework.
When students begin to make the transition to accurately using simple molecular
chemistry concepts, Level 2 begins, which we call the Representing level. At Level 2
of the Visualizing Matter progress variable, we see students using very one-dimen-
sional models of chemistry: A simple representation, a single definition will be used
broadly to account for and interpret chemical phenomena. Students show little abil-
ity to combine ideas, but begin to grasp and use simple definitional concepts such as
elements, compounds, solutions, and valence electrons.
When students can begin to combine and relate patterns to account for, for in-
stance, the contribution of valence electrons and molecular geometry to dissolving,
they are considered to have moved to Level 3, Relating. Remaining levels of the
framework represent further sophistications, extensions, and refinements on Level 3
and are not expected to be mastered in high school but in university courses.
This example shows how a progress variable can be developed to generate infor-
mation on student mastery. It assumes, however, that data collected are quality data,
an issue we next address.

Everyday Assessment in the Science Classroom 97

CHAPTER 7

Satisfying Principle 2: Quality Evidence

Technical issues of reliability and validity, fairness, consistency, and bias discussed
in Principle 2 can quickly sink any attempt to measure along a progress variable as
described above. To ensure comparability of results across time and context, proce-
dures are needed to (a) examine the coherence of information gathered using differ-
ent formats, (b) map student performances onto the progress variables, (c) describe
the structural elements of the accountability system—tasks and raters—in terms of
the achievement variables, and (d) establish uniform levels of system functioning, in
terms of quality control indices such as reliability. While this type of discussion can
become very technical to consider, it is sufficient to keep in mind that the traditional
elements of test standardization, such as validity and reliability studies and bias and
equity studies, must be carried out to satisfy quality control and ensure that evidence
can be relied upon.
Our approach on this technical end of measurement is to use item response mod-
eling (also known as IRT), as described by Adams and Wilson (1992, 1996). These
are measurement models now well-developed enough for use in classroom-based
assessment in a fairly routine and feasible way. The output from these models can be
used as quality control information to address the concerns above, and to determine
where individual students fall on a progress variable such as Visualizing Matter, or
any other progress variable that might be conceived and validated.
The formal nature of these models and their flexibility allow one to address
technical challenges inherent in the classroom assessment situation, such as the main-
tenance of teacher rating consistency and the maintenance of a meaningful scale
throughout the school year. This puts richer information into the hands of teachers in
the classroom. The central feature is the progress map, which provides a graph of the
progress that students are making through the curriculum. Shown in Figure 2a and
Figure 2b are two maps—one showing one student’s progress along the Visualizing
Matter variable and the other showing where this same student stands on all four of
the LBC high school progress variables. Maps are derived from empirical analyses
of student data collected from classrooms.2
Once constructed, maps can be used to record and track student progress and to
illustrate the skills a student has mastered and those that the student is working on.
By placing students’ performance on the continuum defined by the map, teachers
can demonstrate students’ progress with respect to the goals and expectations of the
course. The maps, therefore, are one tool to provide feedback on how students as a
whole are progressing in the course.

2
These maps were drawn in GradeMap, an experimental software package developed at the Univer-
sity of California, Berkeley (Wilson, Draney, and Kennedy 2002). The analyses for these maps were
performed using the ConQuest software (Wu, Adams, and Wilson 1998), which implements an Esti-
mation-Maximization algorithm for the estimation of multidimensional Rasch-type models. For a
detailed account of the estimation and model-fitting see Draney and Peres (1998).

98 National Science Teachers Association

 CHAPTER 7

Figure 2a. A “Conference Map” for a Single Student

Note: The map helps the student know where he or she measures on each progress variable
and makes suggestions for how the student can improve.

Maps, as graphical representations of student performance on assessment tasks,

can be used by teachers for their own instructional planning and to show students,
administrators, and parents how students are developing on progress variables over the
year. This can then be used to inform instructional planning. For instance, if the class
as a whole has not performed well on a variable following a series of assessments, then
the teacher might go back and readdress those concepts or issues reflected by the as-
sessments. Additionally, during the development stage, unsatisfactory map results can
indicate changes or additions to the curriculum.

Satisfying Principle 3: Match between Instruction and Assessment

The match between the instruction and assessment in the BEAR Assessment System is
established and maintained through two major parts of the system: progress variables,
described above, and assessment tasks, described in this section. The main motivation
for the progress variables so far developed is that they serve as a framework for the
assessments and a method of making IRT measurement possible. However, this third
principle makes clear that the framework for the assessments and the framework for
the curriculum and instruction must be one and the same. This is not to imply that the

Everyday Assessment in the Science Classroom 99

CHAPTER 7

Figure 2b. A “Progress Map” for a Single Student on the Visualizing

Matter Variable in Two Curriculum Modules (Weather and Toxins)

Note: The first timepoint is a pretest score, followed by a first and second assessment of the
same student in the Weather module and a final assessment in the Toxins module. The
horizontal bands represent the score levels of the Visualizing Matter progress variable,
calibrated by item response modeling of data from classroom tests of the assessment system
completed during curriculum development.

needs of assessment must drive the curriculum, but rather that the two, assessment and
instruction, must be in step—they must both be about accomplishing the same thing,
the aims of learning, whatever those aims are determined to be.
Using progress variables to structure both instruction and assessment is one way
to make sure that the two are in alignment, at least at the planning level. In order to
make this alignment concrete, however, the match must also exist at the level of
classroom interaction and that is where the nature of the assessment tasks becomes
crucial. Assessment tasks need to reflect the range and styles of the instructional
practices in the curriculum. They must have a place in the “rhythm” of the instruc-
tion, occurring at places where it makes instructional sense to include them, usually

100 National Science Teachers Association

 CHAPTER 7

where teachers need to see how much progress their students have made on a spe-
cific topic (see Minstrell 1998 for an elaboration of such occasions).
One good way to achieve this is to develop both the instructional materials and the
assessment tasks at the same time—adapting good instructional sequences to produce
assessable responses and developing assessments into full-blown instructional activi-
ties. Doing so brings the richness and vibrancy of curriculum development into assess-
ment, and also brings the discipline and hard-headedness of assessment data into the
design of instruction.
The variety of assessment tasks used by the BEAR Assessment System can range
widely, including individual and group “challenges,” data interpretation questions,
and tasks involving student reading, laboratory, or interactive exercises. In LBC tasks,
all assessment prompts are open-ended, requiring students to fully explain their re-
sponses. For the vast majority of assessment tasks, the student responses are in a
written format.3
An example of an LBC assessment prompt and actual student answers at Levels
1 and 2 are shown in Figure 3. Performance is judged by the validity of the argu-
ments presented, not simply by the conclusions drawn.
By developing assessment tasks as part of curriculum materials, they can be
made directly relevant to instruction. Assessment can become indistinguishable from
instructional activity, without precluding the generation of high-quality, compara-
tive, and defensible assessment data on individual students and classes.
Whatever the form of instruction, if student work is generated or students can be
observed at work and this work can be scored and matched to progress variables,
then it is possible to consider use of an assessment system such as BEAR and to
clearly match the assessments to instruction.

Satisfying Principle 4: Management by Teachers

For information from assessment tasks and BEAR analysis to be useful to teachers
and students, it must be couched in terms that are directly related to the instructional
goals of the progress variables, and open-ended tasks, if used, must be quickly, readily,
and reliably scorable. Our response to these two issues is scoring guides (a.k.a. ru-
brics). Scoring guides define the performance criteria, or characteristics, for each
score level to be awarded. “Living by Chemistry” has one scoring guide used over all
variables, with both open-ended tasks scored by teachers and multiple-choice and
other close-ended questions scored by machine and computer. Another course might
have one scoring guide for each of the progress variables. Thus, decisions on the
structure of tasks and deployment of scoring and guides can be made course by
course, but must reflect the time constraints and needs of teachers as well as mea-
surement.

3
This is not a limitation of the BEAR system, but reflects the only practical way we had available for
teachers to attend to a classroom of student work.

Everyday Assessment in the Science Classroom 101

CHAPTER 7

Figure 3. Example of a Living by Chemistry (LBC) Project Assessment

Prompt with Actual Student Answers
Question: You are given two liquids. One of the solutions is butyric acid with a molecular
formula of C4H8O2. The other solution is ethyl acetate with the molecular formula
C4H8O2. Both of the solutions have the same molecular formulas, but butyric acid
smells bad and putrid while ethyl acetate smells good and sweet. Explain why you
think these two solutions smell differently.
Student Answers at Level 1 of Visualizing Matter progress variable
Response: “I think there could be a lot of different reasons as to why the two solutions smell
differently. One could be that they’re different ages, and one has gone bad or is
older which changed the smell. Another reason could be that one is cold and one is
hot.”
Response: “Using chemistry theories, I don’t have the faintest idea, but using common
knowledge I will say that the producers of the ethyl products add smell to them so
that you can tell them apart.”
Response: “Just because they have the same molecular formula doesn’t mean they are the
same substance. Like different races of people: black people, white people. Maybe
made of the same stuff but look different.”
Analysis: These students use ideas about phenomena they are familiar with from their
experience combined with logic/comparative skills to generate a reasonable answer,
but do not employ molecular chemistry concepts.
Student Answers at Level 2 of Visualizing Matter progress variable
Response: “They smell differently b/c even though they have the same molecular formula, they
have different structural formulas with different arrangements and patterns.”
Response: “Butyric acid smells bad. It’s an acid and even though they have the same molecu-
lar formula but they structure differently.”
Analysis: Both responses appropriately cite the principle that molecules with the same
formula can have different structures, or arrangements of atoms, within the structure
described by the formula. However the first answer shows no attempt and the
second answer shows an incomplete attempt to use such principles to describe the
simple molecules given in the problem setup, which would have advanced the
response to the next level.

Note: To match instruction and assessment, this LBC assessment question followed a laboratory
project in which students explored chemicals that had different smells.

When scoring guides are used, teachers and students need concrete examples—
which we call “exemplars”—of the rating of student work. Exemplars provide con-
crete examples of what a teacher might expect from students at varying levels of
development along each variable. They are also a resource to understand the ration-
ale of the scoring guides. Actual samples of student work, scored and moderated by
teachers who pilot-tested the BEAR Assessment System in a course called “Issues,
Evidence, and You,” are included with the documentation for the course. These samples
illustrate typical responses for each score level for specific assessment activities.
In addition to the scoring guides, the teacher needs a tool to indicate when as-
sessments might take place, and what variables they pertain to. These are called
Assessment Blueprints and are a valuable teacher tool for keeping track of when to
assess students. Assessment tasks are distributed throughout the course at opportune

102 National Science Teachers Association

 CHAPTER 7

points for checking and monitoring student performance, and these are indicated in
the Assessment Blueprints. Teachers can use these blueprints to review and plan
assessment tasks relating to each variable, and to modify the assessments to their
own needs.

Bringing It All Together

The four principles of the BEAR system are not designed to operate in isolation.
Each of the principles provides a unifying “thread” throughout the system, but their
interrelationships also make the system more integrated. For example, the progress
variables provide an initial unity to the curriculum materials and define not only the
content of student learning but also the paths over which student learning develops
throughout the year. The implication is that each assessment, then, has a designated
place in the instructional flow, reflecting the type of learning that students are ex-
pected to demonstrate at that point in time. Hence, scores assigned to student work
can then be linked back to the developmental perspective and used both to diagnose
an individual’s progress with respect to a given variable and also to “map” student
learning over time.
Adherence to each of the principles across each of the phases of the assessment
process produces a coherence or “internal consistency” to the system. Adherence to
each of the principles within each phase of the assessment process produces a well-
integrated system that addresses the complexity of the classroom and desired link-
ages among curriculum, instruction, and assessment.
Proper operation of the BEAR Assessment System requires that teachers and
students “take control” of essential parts of the assessment system, including the
scoring process, and also that teachers and students sometimes face demands that
they grow in order to master the system. We have devised the “assessment modera-
tion meeting” as part of our staff and student development strategy to accomplish
these goals. In these meetings, teachers discuss the scoring, interpretation, and use
of student work, and make decisions regarding standards of performance and meth-
ods for reliably judging student work related to those standards. Moderation sessions
also provide the opportunity for teachers to discuss implications of the assessment
for their instruction, for example, by discussing ways to address common student
mistakes or difficult concepts in their subsequent instruction. The moderation pro-
cess gives teachers the responsibility of interpreting the scores given to students’
work and allows them to set the standards for acceptable work. Teachers use mod-
eration to adapt their judgments to local conditions. Upon reaching consensus on the
interpretations of score levels, teachers can then adjust their individual scores to
better reflect the teacher-adapted standards. The use of moderation allows teachers
to make judgments about students’ scores in a public way with respect to public
standards and improves the fairness and consistency of the scores.
Moderation also can take place with student groups, so students can better grasp
and determine for themselves what the teacher and course are valuing in terms of

Everyday Assessment in the Science Classroom 103

CHAPTER 7

student learning. Students can score actual class work, if that is deemed appropriate,
or can score work provided as examples in the curriculum materials. They can map
scores against progress variables and see more concretely the paths toward mastery
of learning aims.

The Bottom Line

An important bottom-line question still remains—does an assessment system make
a difference to students’ knowledge and mastery of course material and aims? We
next consider evidence from a different application of the BEAR Assessment Sys-
tem: the Science Education for Public Understanding Project (SEPUP) at the Lawrence
Hall of Science. SEPUP developers created a yearlong, issues-oriented science course
for the middle school and junior high grades entitled “Issues, Evidence, and You”
(SEPUP 1995). Societal decision-making is a central focus of “Issues, Evidence,
and You”(IEY). As part of the course, students are regularly required to recognize
scientific evidence and weigh it against other concerns with the goal of making in-
formed choices about relevant contemporary issues or problems.
The “Issues, Evidence, and You” field test took place in the 1994–1995 school
year at six centers around the country known as Assessment Development Centers
(ADCs). Teachers at these centers were required to use the BEAR Assessment Sys-
tem and to participate in moderation meetings throughout the year; in each of their
classrooms, data were collected from the assessment activities. In addition, the study
included Professional Development Centers (PDCs), where teachers taught the IEY
curriculum and were provided with the same assessment materials as the ADC group
but were not required to use the assessment system. PDC teachers did, however,
participate in professional development activities of similar duration as the modera-
tion activities, related to the curriculum rather than to the assessment system.
In order to examine the effectiveness of the assessment system, the results of the
pretests and posttests for the PDC and comparison groups, and the pretests, posttests,
and various link tests for the ADC group, were compared. Both groups used the
same curriculum. The same instructional and assessment materials were provided to
both groups. The only difference in treatment between the two groups was that teachers
in the ADC group paid specific attention to the assessment system.
Figure 4 shows that ADC students made substantial gains during the school year,
especially as compared with the PDC and comparison students. While the average
gain for the PDC and comparison groups was similar (difference not statistically
significant at the p = 0.05 level), the average gain for the ADC group was signifi-
cantly greater (3.46 times greater, p = 0.05 level).
This gain can also be put in educational terms: The maximum gain for the ADC
group is, for the average student, approximately a difference between a 2 and a 3 on
the 0–4 scaled scoring guide. Without specific attention to the scoring system, stu-
dents achieved an average 2 while those under the assessment system evidenced
mastery at level 3. This is an educationally significant change, marking a difference

104 National Science Teachers Association

 CHAPTER 7

Figure 4. Student Gains in the “Issues, Evidence, and You” Curriculum

When Taught by Teachers at Asssessment Development Centers (ADCs), at
Professional Development Centers (PDCs), and in a Comparison Group

1490

1470

1450 ADC
PDC
Comparison

1430

1410

1390

ADC gain

PDC gain

Comparison gain
1370

Pretest/ Link Link Posttest/

Preanchor test 1 test 2 Postanchor

1350
Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun

Note: Teachers at ADCs were required to use the BEAR Assessment System and to participate in
“moderation meetings” throughout the year. Teachers at PDCs were given the same assessment
materials as the ADC group but were not required to use the assessment system. PDC teachers did
participate in professional development activities, but unlike the ADC group, the activities were
related to the curriculum rather than to the assessment system. Teachers in the comparison group had
the embedded items available to them in the student booklet and teacher’s guide, but there was no
explicit request to use them, or even any encouragement to do so. They did not have the link tests
available to them.

between achievement of partial but incomplete success toward the course standards
and satisfactory completion.
These results reinforce the common-sense idea that considerable potential gains
could be realized by (a) closer attention to assessment concerns at the classroom
level and (b) a more systematic approach to the gathering and interpretation of as-
sessment information.

Everyday Assessment in the Science Classroom 105

CHAPTER 7

Conclusion
The work reported here takes only a few short steps along some of the paths that
could be considered with regard to classroom-based assessment and the generation
of high-quality data. It seems clear that going beyond grades to map individual tra-
jectories of learning at the classroom level is feasible, especially as computers and
data collection devices become readily available and curriculum is restructured to
better accommodate assessment.
In practice, the benefit to students, teachers, and parents is promising. Further-
more, how our theories of learning and theories of instruction may change in science,
for instance, as a result of better data out of classrooms and a clearer understanding
of learning trajectories is fascinating to consider. We hope to see more formal em-
bedded assessment efforts unfold in coming years, especially in science education as
a forerunner in this area, and invite those interested to, of course, consider using the
BEAR Assessment System. But, moreover, it is important for educators to ponder
the principles behind the system, and that is why we have taken the trouble to lay
them out here. It is in satisfying these principles that the argument for classroom-
based assessment lies.
“Grades are part of the system and will always be there,” said one high school
teacher using the BEAR system. “But this kind of analysis gives me more than just a
grade. I can diagnose a problem and move forward with a greater number of stu-
dents. I can see the amount of time it takes for my students to learn, and find out how
much of something they know, or how well they know it, not just whether they have
a fact in their heads or they don’t. It lets me value even wrong answers, because it
shows me what in each answer I can value and support and work with. To me, it’s a
whole different way to truly value student thinking.”

Appendix
Science content standards for California public schools covered by the “Living by Chemistry: In-
quiry Based Modules for High School” (LBC) Visualizing Matter progress variable include those
shown below. Other LBC variables cover the remaining standards.
Grades 9–12—Chemistry
1c. Students know how to use the periodic table to identify alkali metals, alkaline earth metals and
transition metals, trends in ionization energy, electronegativity, and the relative sizes of ions and
atoms.
1d. Students know how to use the periodic table to determine the number of electrons available for
bonding.
1g. Students know how to relate the position of an element in the periodic table to its quantum elec-
tron configuration and to its reactivity with other elements in the table.

2a. Students know atoms combine to form molecules by sharing electrons to form covalent or metallic
bonds or by exchanging electrons to form ionic bonds.
2e. Students know how to draw Lewis dot structures.

106 National Science Teachers Association

 CHAPTER 7

2f. Students know to predict the shape of simple molecules and their polarity from Lewis dot struc-
tures.
2g. Students know how electronegativity and ionization energy relate to bond formation.

10b. Students know the bonding characteristics of carbon that result in the formation of a large vari-
ety of structures ranging from simple hydrocarbons to complex polymers and biological mol-
ecules.
10d. Students know the system for naming the ten simplest linear hydrocarbons and isomers that
contain single bonds, simple hydrocarbons with double and triple bonds, and simple molecules
that contain a benzene ring.
10e. Students know how to identify the functional groups that form the basis of alcohols, ketones,
ethers, amines, esters, aldehydes, and organic acids.

Acknowledgments
The work of the first author was supported by a National Science Foundation (NSF) grant to the
CAESL Center, WestEd. The work of the second author and the ChemQuery/Living by Chemistry
examples cited in the text were produced and supported by an NSF grant to Professor Angelica Stacy,
Department of Chemistry, University of California, Berkeley.

References
Adams, R. J., and M. Wilson. 1992. A random coefficients multinomial logit: Generalizing Rasch
models. Paper presented at the annual meeting of the American Educational Research Association,
San Francisco.
———. 1996. Formulating the Rasch model as a mixed coefficients multinomial logit. In Objective
measurement III: Theory into practice, eds. G. Engelhard and M. Wilson. Norwood, NJ: Ablex.
Airasian, P. W. 1988. Measurement-driven instruction: A closer look. Educational Measurement: Is-
sues and Practice 7(4): 6–11.
Berry, B. 1997. New ways of testing and grading can help students learn and teachers teach. Reform-
ing Middle Schools and School Systems 1(2). http://www.middleweb.com/CSLB2testing.html.
Brown, A. L., J. C. Campione, L. S. Webber, and K. McGilly. 1992. Interactive learning environ-
ments: A new look at assessment and instruction. In Changing assessments, eds. B. R. Gifford and
M. C. O’Connor, 121–212. Boston: Kluwer.
Claesgens, J., K. Scalise, K. Draney, M. Wilson, and A. Stacy. 2002. Perspectives of chemists: A
framework to promote conceptual understanding of chemistry. Paper presented at the annual meeting
of the American Educational Research Association, New Orleans.
Cole, N. 1991. The impact of science assessment on classroom practice. In Science assessment in the
service of reform, eds. G. Kulm and S. Malcom, 97–106. Washington, DC: American Association
for the Advancement of Science.
Draney, K. D., and D. Peres. 1998. Unidimensional and multidimensional modeling of complex sci-
ence assessment data. BEAR Research Report SA-98-1. Berkeley: University of California.
Griffiths, A. K., and K. R. Preston. 1992. Grade-12 students’ misconceptions relating to fundamental
characteristics of atoms and molecules. Journal of Research in Science Teaching 29(6): 611–28.
Haney, W. 1991. We must take care: Fitting assessments to functions. In Expanding student assess-
ment, ed. V. Perrone, 142–63. Alexandria, VA: Association for Supervision and Curriculum De-
velopment.
Land, R. 1997. Moving up to complex assessment systems. Evaluation Comment 7(1): 1–21.

Everyday Assessment in the Science Classroom 107

CHAPTER 7

Linn, R., and E. Baker. 1996. Can performance-based student assessments be psychometrically sound?
In Performance-based student assessment: Challenges and possibilities. Ninety-fifth yearbook of
the National Society for the Study of Education, eds. J. B. Baron and D. P. Wolf, 84–103. Chicago:
University of Chicago Press.
Masters, G. N., R. A. Adams, and M. Wilson. 1990. Charting student progress. In International ency-
clopedia of education: Research and studies. Supplementary volume 2, eds. T. Husen and T. N.
Postlethwaite, 628–34. Oxford: Pergamon Press.
Minstrell, J. 1998. Student thinking and related instruction: Creating a facet-based learning environ-
ment. Paper presented at the meeting of the Committee on Foundations of Assessment, Woods
Hole, MA (Oct.).
Resnick, L. B., and D. P. Resnick. 1992. Assessing the thinking curriculum: New tools for educational
reform. In Changing assessments, eds. B. R. Gifford and M. C. O’Connor, 37–76. Boston: Kluwer.
SEPUP. 1995. Issues, evidence, and you: Teacher’s guide. Berkeley, CA: Lawrence Hall of Science.
Stake, R. 1991. Advances in program evaluation: Volume 1, Part A: Using assessment policy to re-
form education. Greenwich, CT: JAI Press.
Torrance, H. 1995a. The role of assessment in educational reform. In Evaluating authentic assess-
ment, ed. H. Torrance, 144–56. Philadelphia: Open University Press.
———.1995b. Teacher involvement in new approaches to assessment. In Evaluating authentic as-
sessment, ed. H. Torrance, 44–56. Philadelphia: Open University Press.
Tucker, M. 1991. Why assessment is now issue number one. In Science assessment in the service of
reform, eds. G. Kulm and S. Malcom, 3–16. Washington, DC: American Association for the Ad-
vancement of Science.
Wilson, M. 1990. Measurement of developmental levels. In International encyclopedia of education:
Research and studies. Supplementary volume 2, eds. T. Husen and T. N. Postlethwaite. Oxford:
Pergamon Press.
Wilson, M., and R. J. Adams. 1996. Evaluating progress with alternative assessments: A model for
Chapter 1. In Implementing performance assessment: Promise, problems and challenges, ed. M.
B. Kane. Hillsdale, NJ: Lawrence Erlbaum.
Wilson, M., K. Draney, and C. Kennedy. 2002. GradeMap (Version 2.0) [computer program]. Berke-
ley: University of California, BEAR Center.
Wilson, M., and K. Sloane. 2000. From principles to practice: An embedded assessment system.
Applied Measurement in Education 13(2): 181–208.
Wolf, D., J. Bixby, J. Glenn III, and H. Gardner. 1991. To use their minds well: Investigating new
forms of student assessment. Review of Research in Education 17: 31–74.
Wu, M., R. J. Adams, and M. Wilson. 1998. ACERConQuest [computer program]. Melbourne, Aus-
tralia: ACER Press.
Zessoules, R., and H. Gardner. 1991. Authentic assessment: Beyond the buzzword and into the class-
room. In Expanding student assessment, ed. V. Perrone, 47–71. Alexandria, VA: Association for
Supervision and Curriculum Development.

108 National Science Teachers Association

 CHAPTER 8

Working with Teachers in

Assessment-Related Professional
Development
Mistilina Sato

Mistilina Sato taught middle school science in New Jersey prior to receiving her doctorate in
curriculum and teacher education from Stanford University. She has worked with teachers in a
variety of professional development settings on issues such as inquiry and the nature of science;
classroom-centered assessment; Earth, space, and environmental sciences; action research; and
developing science curriculum. In her current position she works with teachers pursuing Na-
tional Board Certification and is a research assistant with the Classroom Assessment Project to
Improve Teaching and Learning (CAPITAL) at Stanford University. She has been a member of the
National Science Teachers Association since 1992.

I begin with a premise about everyday assessment in the classroom: Assessment is

inseparable from the instructional practices of the teacher (Atkin, Black, and Coffey
2001). Everyday assessment goes beyond the design of performance assessments
administered to students. It is more regularly practiced than the exhibition of culmi-
nating classroom projects. It is more tightly bound to instruction than the use of
regular quizzes to check students’ understanding of discrete concepts. Everyday as-
sessment happens not only every day, but in almost every minute of teacher-student
interactions and student-student interactions.
Class discussions, one-on-one interactions with students, and the varieties of stu-
dent work assigned and collected by the teacher are rife with information about what
students are thinking and how well they are performing. Teachers do not always ex-
plicitly recognize these interactions as assessment opportunities, that is, as opportuni-
ties to use the available information about students’ understanding and performance to
support student learning. Frequently so much information is available to the teacher
that to pay heed to all of it would result in information overload. To use this informa-
tion effectively for assessment purposes requires that the teacher act upon the informa-
tion in some way (Sadler 1989). Teachers may use the information to inform their own
instructional decisions about the direction of the current lesson or the next concept to
address within a curricular unit. Teachers may provide feedback to an individual stu-
dent or to a group of students as to what they need to do next to continue their progress.
Teachers may make note of the information and draw upon it at a later time. Teachers
may report the information to parents as a way of demonstrating student progress or
suggesting ways that parents can support learning at home.
This ongoing identification, collection, and use of information in the classroom
is a complex business. It often requires a shift in the teacher’s priorities in the class-
room from managing activity and behavior to a mind-set of managing learning op-
portunities. Traditionally, assessment in the classroom has been viewed as a means

109
CHAPTER 8

by which teachers reach a judgment about the quality of student performances on

tests, quizzes, and a variety of classroom projects. Heightening the teacher’s awareness
about everyday assessment opportunities in the ongoing instruction shifts attention from
the products of assessment to the interactions among people in the classroom. Chang-
ing assessment practices based on this premise then becomes a matter of change that
must incorporate not only new assessment practices, but also a different way of think-
ing about the role the teacher and students play with regard to assessment. New tech-
niques and strategies for assessing student progress and performance will contribute
to improving teachers’ assessment practices; however, working with teachers on
everyday assessment issues must also address the teachers’ views about the assess-
ment cultures in their classrooms.
This chapter looks broadly at approaches to working with teachers on assessment-
related matters and highlights an approach that places the teacher as an individual and
his or her everyday classroom practices at the center of the professional development
interaction. This approach views assessment not only as a means of determining what
the students have taken away from instructional interactions, but also as a means of
supporting learning as it is taking place. I will not attempt to describe all of the assess-
ment-related professional development work going on around the country. This work
is far-reaching, and an essay of this length would not do justice to the various programs
and projects. Rather, I will discuss an approach that sees the teacher not only as a
professional engaged in learning new ideas or strategies for assessing but also as an
individual who is undergoing personal change.

Approaches to Working with Teachers

Work with teachers in assessment-related professional development commonly seeks
to introduce teachers to strategies for assessing student performance. In this approach,
teachers are taught design principles for creating effective assessments, they exam-
ine lesson design to identify when assessment opportunities arise, or they might
practice strategies developed by other teachers and consider when these strategies
would be appropriate in their own teaching situations.
As other writers in this volume have also pointed out, teachers can examine
student work in order to understand what their students know and to refine their
assessment. Teachers also can examine the questioning strategies they use in their
classrooms and relate the questions posed to the responses offered by students. Dis-
secting the teacher’s practice into discrete elements—such as prompts offered to
students in written work or questions posed to students in discussion—helps teach-
ers focus on targeted aspects of their classroom practices. This kind of focused atten-
tion allows teachers to learn about, and potentially implement, new techniques.
Experts in the field of assessment have also developed programs for introducing
teachers and school organizations to a variety of approaches to classroom assess-
ment and schoolwide systems of assessment. (See, for example, the work of Richard
Stiggins, founder of the Assessment Training Institute, an organization in Portland,

110 National Science Teachers Association

 CHAPTER 8

Oregon, that offers professional development programs and materials on assessment

strategies nationwide [http://www.assessmentinst.com/index.html] and the work of
Grant Wiggins, president and director of programs for Re: Learning by Design, a
not-for-profit educational organization in Pennington, New Jersey, that also consults
nationally in professional development and assessment systems [http://
www.relearning.org/].) In these programs, teachers learn about the technical aspects
of assessment design and consider assessment systems that are aligned across grade
levels and with the curriculum. These programs intend to help teachers become “lit-
erate” in a variety of assessment practices and to become skilled at using them.
In both of these examples, the professional development approach aims to bring
about change in the teachers’ instructional practice by helping them learn about and
implement new assessment strategies, skills, and systems. While there is great need
for teachers to develop a stronger understanding about assessment (Stiggins 2002),
an approach focused solely on teaching teachers about assessment only partially
addresses the issue of bringing about change in the teachers’ assessment practices.
A teacher’s practice comprises more than the technical and strategic moves he or
she makes in the classroom. If the classroom is viewed as a place in which teachers
deploy a diverse variety of assumptions, theories, and practices, simply introducing
teachers to new instructional practices without spending time developing an under-
standing of the reasons behind the change—and how these changes jibe with the
teacher’s mind-set about how to operate his or her classroom—may result in the
superficial enactment of those practices.
I am reminded of the illustration that David Cohen (1990) offered in his account
of “Mrs. Oublier,” a second-grade teacher caught up in the reforms in mathematics
instruction. In the ongoing efforts to introduce teachers to models of instruction that
aligned with the National Council of Teachers of Mathematics’ Curriculum and Evalu-
ation Standards for School Mathematics (1989), school districts, universities, and
standards-based curriculum programs offered professional development opportuni-
ties for teachers to help them learn how to teach math using the reform-minded
techniques. Mrs. Oublier was observed implementing many of these techniques in
her classroom. She was doing what she had learned in the workshops—using
manipulatives, engaging the children in multiple modalities for learning, and attempt-
ing to draw out deeper meaning about mathematical patterns rather than teaching
rote memorization of algorithms. However, based on Cohen’s observations, Mrs.
Oublier’s teaching demonstrated a complex blend of her traditional practices and the
new strategies for teaching mathematics. (Cohen represents Mrs. Oublier’s practices
in an unflattering way, which can be argued to be unfair to the teacher who was
actually observed). Based on conversations with Mrs. Oublier, it was clear that she
thought that her instructional practices had changed dramatically, shifting toward
the reform-minded strategies she had been taught. However, Mrs. Oublier imple-
mented the new techniques filtered through her own traditional teaching experiences
and beliefs about what constitutes appropriate responses to mathematics problems,

Everyday Assessment in the Science Classroom 111

CHAPTER 8

sometimes undermining the desired outcomes of the reform approaches. One can
surmise from this example that teachers’ experiences and preferences have a great
influence on the change process. If change in instructional practices is a desired
outcome of professional development, then teaching teachers new techniques and
skills is necessary, but not sufficient.
Teachers can be taught a wide variety of new ideas and skills as adult, profes-
sional learners. However, teachers do not enter professional development programs
as blank slates. The interpretations that teachers bring to new ideas and new roles are
bound up with their values, their personal backgrounds, their prior experiences with
reform, and their beliefs. Thus, the individual is an inextricable part of how these
new ideas and roles are put into action in the workplace. In the next section, I discuss
an approach to working with teachers that recognizes the active part that the indi-
vidual can play in his or her professional development.

A Personal Approach to Working with Teachers

Many factors shape a teacher’s pedagogy and, thus, the assessment practices embed-
ded within that pedagogy. The curriculum adopted by a school has embedded as-
sumptions about what should be assessed and, in some cases, how to assess student
progress and learning. The conception of the subject matter and its foundational
epistemology influence how the teacher selects learning goals, scaffolds lessons aimed
at helping students attain those goals, and determines student progress toward those
goals. The conception of learning that the teacher holds influences how the teacher
designs lessons and how the teacher views the role of student motivation in the learn-
ing process. A teacher’s practice is also influenced by the established practices within
his or her professional community. Finally, teaching can be viewed as a personal act
that is guided and shaped by the teacher’s prior experiences, preferences, and be-
liefs.
Teachers’ practices comprise a broad range of interactions and decision-making.
In the classroom with students, teachers make countless decisions—who to call on
in a discussion, when to summarize the main ideas, when to allow a digression.
Sanders and McCutcheon (1986) suggest that teachers’ decision-making is guided
by their practical theories, defined as

conceptual structures and visions that provide teachers with reasons for
acting as they do, and for choosing the teaching activities and curriculum
materials they choose in order to be effective. They are the principles or
propositions that undergird and guide teachers’ appreciations, decisions, and
actions. (54–55)

Practical theories held by the teachers provide the “reasons why teachers do
what they do” (56), and the actions taken by the teacher are the manifestations of
these theories. The source of the theories is eclectic, drawing on research-based

112 National Science Teachers Association

 CHAPTER 8

information, experience, and the particulars of the circumstances in which the theo-
ries are employed.
When engaging in a process of change with teachers, whether it be about new
curriculum, new instructional strategies, or assessment practices, it is important to un-
derstand who the teacher is and what his or her priorities and assumptions are. This
approach to professional development places emphasis on existing mind-sets in addi-
tion to introducing new ideas, strategies, or skills. Thus, a personal approach to work-
ing with teachers on assessment-related issues promotes the examination and articula-
tion of the teacher’s current assessment practices and considers how the teacher can be
actively engaged in deciding how to make innovations in his or her own classroom.

An Illustration
The Classroom Assessment Project to Improve Teaching and Learning (CAPITAL),
a National Science Foundation1–funded project, was conducted by Stanford Univer-
sity in partnership with five school districts in the San Francisco Bay Region from
1998 through 2003. The project worked with teachers in a close collaboration to
examine classroom assessment practices in middle school science. The science teach-
ers who worked in the project shared a desire to improve their classroom assessment
practices and engaged in a form of collaborative action research—identifying as-
pects of their everyday assessment practices that they explored through a process of
inquiry, with the intention of making changes in their classrooms.
The CAPITAL staff approached their work with teachers from the perspective
that teachers’ practical theories, their beliefs about assessment, about students, and
about themselves influenced their instructional practices and would also help deter-
mine the changes they chose to make in those practices. Project staff met regularly
with the participating teachers individually and in small groups to discuss classroom
innovations, to address struggles the teachers were facing, and to raise questions
about assessment practices. In these discussions, the teachers reflected on their teach-
ing practices and the modifications they were making in instruction and assessment.
When appropriate, university staff offered suggestions and ideas about assessment.
Generally, the university staff served as facilitators of discussion, raising questions
that would elicit more elaborate descriptions and reasoning behind decisions made
by the teachers.
CAPITAL staff did not begin their work with teachers with preconceived ideas
about desired changes in classroom practice. Rather, they focused on the teachers’
emerging interests and priorities. The nature of the interaction with each teacher and
each group of teachers varied, depending on who the teachers were and what they
identified as their needs. The project team played different roles with the individual
teachers: a mirror reflecting back to the teacher aspects of his or her teaching; a
second pair of eyes or ears in the classroom to document classroom interactions; a

1
National Science Foundation Grant REC-9909370

Everyday Assessment in the Science Classroom 113

CHAPTER 8

videographer who captured classroom interactions for future reflection and analysis;
a critical friend who raised questions for further consideration.
Relationships with teachers took time to develop and cultivate. The joint work
between university staff and classroom teachers was intense, drawing on each mem-
ber of the project both professionally and personally. Delving into questions about
the goals that the teacher had for his or her students, why the teacher made particular
choices in the classroom, or how information would be used required that the teach-
ers trust the university staff and that the university staff trust the teachers. This ap-
proach to working with teachers was personal (Coffey, Sato, and Schneider 2001). It
was work that took place in their classrooms, examined the issues that they deemed
important, and investigated the ways that their personal views about teaching and
learning influenced their teaching.
Through this focused approach with small groups and individual teachers, CAPI-
TAL staff and teachers came to understand how important teachers’ beliefs are in
their carrying out of assessment. The following examples from the project illustrate
this point.
One teacher participating in the project began teaching as a second career after
working for decades in a biology laboratory. This piece of biographical information
is important for understanding how this teacher views everyday assessment. Assess-
ment is firmly grounded in evidence from her classroom. She carefully examines
student work and maintains detailed records of what the students have completed.
Much like a scientist drawing upon empirical evidence when making inferences or
drawing conclusions, this teacher regularly uses her data to make judgments about
students’ progress. This is also the case in her everyday interactions in the class-
room. She listens to students while they work independently and in groups to learn
about what they are thinking, and she weighs what she learns from her observations
with what she expects before moving forward with instruction.
This teacher’s work within CAPITAL centered on the examination of how stu-
dents understand particular concepts about the practice of science. This focus again
stemmed from the teacher’s deep interest—one might call it a passion—in science as
an enterprise practiced by people. Particularly, she was interested in the students’
skills and understanding about data analysis. She designed her science curriculum so
as to return to questions of data analysis iteratively throughout the school year. She
raised questions such as how to support the learning of the idea of error in scientific
measurements through scientific investigation and subsequent data analysis; how to
help students understand the concept of “average” and its relationship to multiple
trials of a scientific investigation; and how to help students understand the accuracy
of scientific instrumentation and the limitations that the instrumentation places on
scientific investigations.
This teacher’s work in investigating assessment practices in her classroom arose
from her own interests in science as a discipline comprising not solely factual infor-
mation, but also concepts of accepted practice and procedure for scientific investiga-

114 National Science Teachers Association

 CHAPTER 8

tion and analysis. Her interests led her to link her explorations of her assessment
practices tightly to the scientific practices that she valued from her own experience
as a scientist. This pursuit was not trivial. She plumbed the depths of scientific con-
cepts and identified nuances in student understanding of these concepts through her
analysis of student work and her careful listening in the classroom. In this case, both
the methods of investigating assessment practices and the substance of the investiga-
tions related to the teacher’s personal experience and who she is as a person.
Another teacher entered teaching as a career because he was determined to try to
make school a positive and productive experience for students who have tradition-
ally been marginalized by the system. For this teacher, social justice and equitable
opportunities for students were foundational to his work as a teacher and what he
believed in deeply as a person. One of his goals in his work with CAPITAL was to
“find lots of possible ways for kids to be successful.” When he first began working
with CAPITAL, he viewed traditional assessment practices as processes that had
damaged students’ self-esteem, that had treated students unfairly, and that were de-
signed to punish students. He was very interested in figuring out how assessment in
his classroom could support students in their endeavors to learn science and how
assessment could encourage students to engage in his science class, rather than dis-
couraging them from doing so.
One of the strategies that this teacher implemented was to allow students to work
on assignments until the assignments met the teacher’s standards. Students could get
written or verbal feedback from the teacher and continue to improve their work until
it received an A. Knowing that they could correct, revise, or polish their work helped
the students view their efforts as a process for engaging in learning rather than as a
process of production. Students who had not been successful in traditional grading
systems were now able to engage in the work without fear of imminent failure.
For this teacher, everyday assessment was not merely about assigning grades.
He saw assessment as a means to encourage participation in science and to nurture
success. In his view, if the students eventually demonstrated that they understood a
concept or could perform a skill, it did not matter how many times they had to revise
their work. To implement this assessment technique, the teacher had to develop strat-
egies of giving meaningful feedback that the students could act on. He struggled
with the amount of time that it took to provide multiple rounds of feedback. How-
ever, the payoff of greater student involvement and higher quality work were mean-
ingful to him and his beliefs about what his students needed to be successful.

A Collaborative Personal Approach

CAPITAL was founded on the principle that teachers’ understanding about their
own teaching is deepened when they discuss their experiences with peers who are
trying to make similar changes (Atkin 1994; Wineburg and Grossman 1998). Not
only do the teachers learn more about their practice, but they also develop a height-
ened awareness of their sense of self-as-teacher while engaged in communities of

Everyday Assessment in the Science Classroom 115

CHAPTER 8

like-minded colleagues (Lave and Wegner 1991). Teachers who work in CAPITAL
devise what they want to explore with regard to assessment. They try innovations in
their classrooms and periodically come together to talk about what they have done
and what they have learned from their forays into assessment in their everyday prac-
tices. Again, an illustration from the project will demonstrate this principle in action.
A group of teachers from the same school district recently reflected together on
their work in CAPITAL. One of the teachers recounted how an expert in the develop-
ment of rubrics for assessing student performance had given a presentation at her
school. She said that she found the presentation interesting and she could see how
she might use some of the ideas, but she did not have the opportunity to try them out
and explore how the rubrics might work in her classroom. Further, she emphasized
the value of having time within CAPITAL to convene with her colleagues and dis-
cuss the strategies and innovations she has since tried in her classroom.
This consultation and deliberation time gave this group of five teachers time to
reflect on their assessment practices, garner new ideas and refinements from other
experienced teachers, and make connections to broader and deeper issues related to
assessment. For example, this group of teachers began their work on everyday as-
sessment with CAPITAL with a desire to reduce the amount of paperwork they were
handling in the process of collecting and grading daily assignments from the more
than 150 students they saw each day. They were feeling overwhelmed and wanted to
work on ways to streamline their grading practices. At first blush, this pursuit might
have seemed self-serving, as one of the teachers described it after a year of working
with the project. Through the discussions about strategies to reduce all the paper-
work, however, the teachers began to raise questions about what it was that they
were fundamentally assessing about their students by collecting homework and
classwork every day. Together, they came to a realization that much of their time was
spent keeping records of completion of assignments without a clear focus on whether
or what the students were learning. The backlog of paper that accumulated some-
times meant that students did not get their assignments back for weeks. Jointly, the
teachers realized that the comments they made on student work may not have the
desired impact, given that the assignment was weeks old when it was returned; fur-
thermore, students felt no responsibility for reading or responding to the comments
the teacher had painstakingly made on their completed work. The teachers began to
explore ideas about the importance of timely feedback, whether verbal or written,
while the students were engaged with the assignment.
As a result of these discussions, the teachers began to look more closely at the
actual assignments they were asking the students to complete. Their desire for greater
efficiency in their grading led them to ask themselves why they assigned particular
projects and what they thought students would learn from the assignment and to
develop strategies that would allow them to give feedback to students while the stu-
dents were working on their assignments. The teachers had begun the discussion
with an understanding of assessment as a process of keeping track of students’ com-

116 National Science Teachers Association

 CHAPTER 8

pleted work. Through their classroom inquiries, deliberations with colleagues pursu-
ing similar lines of investigation, and the periodic input from the university staff, the
teachers began to see assessment as a process to support learning in a timely and
ongoing fashion. Their collaboration prompted deeper thinking about their assess-
ment practices and led to a variety of practical, innovative projects with clear learn-
ing goals and to new processes of giving feedback to students.
One of the teachers described how the process of discussion is critical to under-
standing how new techniques can best be implemented in the classroom: “When we
try to explain what we have done to other teachers, they get excited about our ideas.
But when they don’t have the opportunity to discuss it after they have tried it out,
they don’t get it.” This insight speaks to the importance of not only introducing new
ideas to teachers, but also allowing for opportunities to reflect with others on how to
tailor these ideas to meet the needs of particular teachers.

Implications for Working with Teachers

How does this personal approach to working with teachers fit into a broader view of
professional development? In recent years, the professional development commu-
nity has reached consensus about principles for the design of high-quality profes-
sional development for teachers (Hawley and Valli 1999, 139–44):

Principle one: Professional development should be driven by analyses of the

differences between goals and standards for student learning and student
performance.

Principle two: Professional development should involve learners (such as

teachers) in the identification of what they need to learn and, when possible,
in the development of the learning opportunity and the process to be used.

Principle three: Professional development should be primarily school based

and integral to school operations.

Principle four: Professional development should provide learning

opportunities that relate to individual needs but for the most part are
organized around collaborative problem solving.

Principle five: Professional development should be continuous and ongoing,

involving follow-up and support for further learning, including support from
sources external to the school that can provide necessary resources and an
outside perspective.

Principle six: Professional development should incorporate evaluation of

multiple sources of information on outcomes for students and processes that

Everyday Assessment in the Science Classroom 117

CHAPTER 8

are involved in implementing the lessons learned through professional

development.

Principle seven: Professional development should provide opportunities to

engage in developing a theoretical understanding of the knowledge and skills
to be learned.

Principle eight: Professional development should be integrated with a

comprehensive change process that deals with impediments to facilitators of
student learning.

The personal approach to working with teachers described in this chapter is not
in conflict with any of these principles, and the CAPITAL example demonstrates
how principles two, three, four, five, and seven can be put into action. With appropri-
ate attention to the role that the teacher plays in the design of professional develop-
ment, principles one, six, and eight can also be addressed while maintaining a focus
on individual teachers’ classrooms and personal changes.
I am not suggesting that a personal approach is appropriate for all teachers or all
programs. An individualized and personal approach to professional development is
time, personnel, and resource demanding. Large-scale reform efforts that are fo-
cused on instructional change are usually economically constrained and thus seek
efficient methods by which to introduce teachers to reform ideas and practices. Pro-
fessional development models that emphasize the dissemination of information to
teachers or those that seek to train teachers in particular pedagogical skills typically
do not pay attention to the individual teacher’s interpretation of the information and
skills into his or her practice. As Mrs. Oublier respectfully demonstrated, the transla-
tion to practice does not always carry with it the spirit of the intended change. If
complex change is expected, then alternative models of professional development
must engage teachers in ongoing examination of practice, rethink how teachers’ time
is spent in schools, provide structural and cultural support for teachers to engage in
new practices, and value teacher change as not just a process of implementing new
behaviors but also as a process of personal change.

Summary and Conclusion

Everyday assessment is embedded in the ongoing interactions in the classroom. For
example, teachers make instructional decisions based on how a student responds to a
question posed during class discussion, students reflect on their own confidence in
understanding a concept, and students judge the quality of their own work and the
work of their peers. Professional development aimed specifically at everyday class-
room assessment, then, must approach the realm of assessment with a broader view
of when assessment occurs and how information is used in the classroom. While it is
important for teachers to learn how to design appropriate stand-alone assessments,

118 National Science Teachers Association

 CHAPTER 8

professional development aimed at everyday assessment must focus more broadly

on the teaching and learning interactions and address how the teacher’s approach to
assessment fits into his or her overall instructional practice.
In this chapter, I have discussed how work with teachers in assessment-related
professional development can be approached from a personal perspective. While
introducing new ideas, strategies, techniques, and skills to teachers is an important
aspect of professional learning, this is only part of the process for working with
teachers to bring about new understanding or change in their classrooms. What a
teacher knows and can do is only a partial determinant of how he or she acts in the
classroom. A teacher’s actions are also shaped by his or her contexts, beliefs, values,
and prior experiences. Teachers are often driven to action by their enthusiasm and
their passions. Professional development that helps teachers to identify and articu-
late their needs, interests, and passions taps into the teacher’s individual desires. A
personal approach to working with teachers values what the individual teacher brings
to the professional development experience, recognizes that the teacher’s beliefs and
prior experiences will have great influence on how new practices and innovations
are enacted in the classroom, and allows the teacher to determine his or her priorities
for change.

References
Atkin, J. M. 1994. Teacher research to change policy. In Teacher research and educational reform:
93rd yearbook of the National Society for the Study of Education, Part I, eds. S. Hollingsworth and
H. Sockett. Chicago: University of Chicago Press.
Atkin, J. M., P. Black, and J. Coffey, eds. 2001. Classroom assessment and the national science
education standards. Washington, DC: National Academy Press.
Coffey, J., M. Sato, and B. Schneider. 2001. Classroom assessment—Up close and personal. Paper
presented at the annual meeting of the American Educational Research Association, Seattle, WA
(April).
Cohen, D. K. 1990. A revolution in one classroom: The case of Mrs. Oublier. Educational Evaluation
and Policy Analysis 12(3): 311–29.
Hawley, W. D., and L. Valli. 1999. The essentials of effective professional development: A new con-
sensus. In Teaching as the learning profession: Handbook for policy and practice, eds. L. Darling-
Hammond and G. Sykes. San Francisco: Jossey-Bass.
Lave, J., and E. Wegner. 1991. Situated learning: Legitimate peripheral participation. Cambridge:
Cambridge University Press.
Sadler, R. 1989. Formative assessment and the design of instructional systems. Instructional Science
18: 119–44.
Sanders, D. and G. McCutcheon. 1986. The development of practical theories of teaching. Journal of
Curriculum and Supervision 2: 50–67.
Stiggins, R. J. 2002. Assessment crisis: The absence of assessment FOR learning. Phi Delta Kappan
83(10).
Wenger, E. 1998. Communities of practice. Cambridge: Cambridge University Press.
Wineburg, S., and P. Grossman. 1998. Creating a community of learners among high school teachers.
Phi Delta Kappan 79(5): 350–53.

Everyday Assessment in the Science Classroom 119

 CHAPTER 9

Reconsidering Large-Scale Assessment to

Heighten Its Relevance to Learning
Lorrie A. Shepard

Lorrie Shepard is dean of the School of Education at the University of Colorado at Boulder. She
has served as president of the American Educational Research Association, president of the Na-
tional Council on Measurement in Education, and vice president of the National Academy of
Education. Her research focuses on psychometrics and the use and misuse of tests in educational
settings. Specific studies address standard setting, the influence of tests on instruction, teacher
testing, identification of mild handicaps, and early childhood assessment. Currently, her work
focuses on the use of classroom assessment to support teaching and learning.

M any science teachers have been affected indirectly by high-stakes, account-

ability pressures as they watch attention and resources flow to language arts
and mathematics instruction—because these subjects are tested. Others have experi-
enced firsthand the ways that external science assessments can undermine inquiry-
based curricula and efforts to teach for understanding. Is it possible to counteract
these effects and make external, large-scale assessments more relevant to student
learning? How can large-scale assessments, remote from the classroom, serve in-
structional purposes?
I agreed to write a chapter addressing these questions with some trepidation
because the history of assessment reform has not been pretty. Ideally, evaluation data
should be used to improve instructional programs and thus ensure meaningful learn-
ing opportunities for students. The difficulty with promoting an ideal, however, is
that we have all seen how a lofty goal can be corrupted when pursued on the cheap or
when too many participants hold conflicting ideas about what was intended. A de-
cade ago, standards-based reformers, recognizing the deleterious effects of tradi-
tional, multiple-choice tests on ambitious learning goals, promised to create “authentic
assessments” and “tests worth teaching to.” These promises have not been realized,
however, in part because accountability advocates have pursued the slogan of high
standards without necessarily subscribing to the underlying theory calling for pro-
found changes in curriculum, instruction, and assessment.
The central aim of this chapter is to consider how large-scale assessments could
be redesigned to heighten their contribution to student learning. In this section, which
acts as a preamble, I (1) explain why assessments must be designed and validated
differently for different purposes and the implications of this differentiation for large-
scale assessments and (2) summarize the essential features of effective classroom
assessment. While classroom assessment is not the focus of this chapter, we cannot
consider here how large-scale assessment could be made compatible with and sup-
portive of classroom instruction and assessment without a shared understanding of
effective classroom assessment. In the next, main section of the chapter I address the
important purposes served by large-scale assessment: (1) exemplification of learn-

121
CHAPTER 9

ing goals, (2) program “diagnosis,” and (3) certification or “screening” of individual
student achievement. In addition, large-scale assessments can serve as a site or im-
petus for professional development to enhance the use of learning-centered class-
room assessment. I conclude with an analysis of the impediments to change and
recommendations for addressing these challenges.

Assessments Designed for Different Purposes

To the layperson, a test is a test. So why couldn’t the same test be used to diagnose
student learning needs; to judge the effectiveness of teachers, schools, districts, and
states; and to compare U.S. schools to the schools of other nations? For measure-
ment specialists, however, purpose matters. Purpose shapes test design and alters the
criteria for evaluating the reliability and validity of the test. According to the Stan-
dards for Educational and Psychological Testing (AERA, APA, NCME 1999), “No
test will serve all purposes equally well. Choices in test development and evaluation
that enhance validity for one purpose may diminish validity for other purposes” (145).
Large-scale assessments are used to monitor achievement trends over time and
to hold schools and school districts accountable. In some states, large-scale assess-
ments are also used to make high-stakes decisions about individual teachers and
students—for example, in regard to teacher pay increases, grade-to-grade promo-
tion, or graduation from high school. Because of the significant consequences that
follow from the results, large-scale assessments must be highly reliable. Thus, pur-
pose shapes technical requirements. And, to be fair, large-scale assessment data must
be collected in a standardized way to ensure comparability across schools. It would
be unfair, for example, if one school gave the test a month later than other schools,
explained unfamiliar words to students, or allowed extra time when students hadn’t
finished.1 Because of the cost of ensuring reliability and standardization and because
of the intrusion on instructional time, large-scale assessments are administered only
once per year and must necessarily be broad, “survey” instruments touching lightly
on the many curricular topics and skills taught throughout the year.
In contrast, classroom assessments intended to help students learn must be closely
tied to particular units of instruction and must be used in the particular days and
weeks when students are learning specific concepts. To be truly diagnostic, teacher’s
questions must probe students’ understandings and push to identify extensions where
mastery is incomplete or where misconceptions impede learning. Because formative
assessment in classrooms is intended to help target the next instructional steps, not to
assign official proficiency status, there is much less need for formal assessment pro-
cedures or adherence to strict reliability standards. Mismeasurement of a student’s
knowledge and skills by a teacher one day can be overturned by subsequent assess-
ments in the next day or week.

1
Note that sources of unfairness to students caused by differences in students’ experiences with and
opportunities to learn tested content are not corrected by standardization.

122 National Science Teachers Association

 CHAPTER 9

Knowing What Students Know (Pellegrino, Chudowsky, and Glaser 2001) is a

landmark report published recently by the National Academy of Sciences that brings
together the knowledge bases of cognitive learning theory and measurement science.
Its authors similarly address this link between purpose and assessment requirements,
referring to the inevitability of trade-offs in assessment design: “Ironically, the ques-
tions that are of most use to the state officer are of the least use to the teacher” (224).
One way to help policy makers understand the limitations of an external, once-per-
year test for instructional purposes is to point out that good teachers should already
know so much about their students that they could fill out the test booklet for them.
“I’m sure Maria can do problems 1, 3, and 4. But she will struggle with problems 2
and 5 because she hasn’t mastered those skills yet.” To be effective in supporting
learning, teachers need in-depth assessments targeted to the gray areas where they
don’t know what their students are thinking.
The following distinctions highlight the differences between large-scale and class-
room-level assessments, which imply that notably different assessment strategies
are needed.
• Standardized vs. dynamic
• Uniform date vs. ongoing dates
• Independent performance vs. assisted performance
• Delayed feedback vs. immediate feedback
• Stringent requirements for technical accuracy vs. less stringent requirements
For example, it is appropriate to provide hints or to alter the task while assessing
for classroom purposes because in so doing the teacher learns what a student can do
independently and pinpoints precisely where understanding breaks down.
In contrast to these distinctions, the single most important shared characteristic
of large-scale and classroom assessments should be their alignment with curriculum
standards. Here I do not mean the limited alignment obtained when test publishers
show that all of their multiple-choice items can be matched to the categories of a
state’s content standards. Rather, I am speaking of the more complete and substan-
tive alignment that occurs when the tasks, problems, and projects in which students
are engaged represent the range and depth of what we say we want students to under-
stand and be able to do. Perhaps a better word would be embodiment. Assessments at
either level should embody and fully represent important learning goals. In science,
we can use the National Science Education Standards (NRC 1996) as our learning
targets. Assessments at both the large-scale and classroom levels, then, must embody
the fundamental concepts, principles, and inquiry skills needed to conduct investiga-
tions and evaluate scientific findings as identified by the standards.
For large-scale and classroom assessments to be symbiotic, they must share this
common understanding of what it means to do good work in a discipline and ideally

Everyday Assessment in the Science Classroom 123

CHAPTER 9

hold a common view of how that expertise develops over time (Pellegrino, Chudowsky,
and Glaser 2001). When the conception of curriculum represented by a state’s large-
scale assessment is at odds with content standards and curricular goals, then the ill
effects of teaching to the external, high-stakes test, especially curriculum distortion
and nongeneralizable test score gains, will be exaggerated, and it will be more diffi-
cult for teachers to use classroom assessment strategies that support conceptual un-
derstanding while at the same time working to improve student performance on the
state test.

A Model of Classroom Assessment in Support of Learning

Classroom assessment is both formal—involving quizzes, exams, laboratory assign-
ments, and projects—and informal—involving journal entries, observations, and oral
questioning. It also serves both formative and summative purposes, depending on
whether assessment insights are used to help students take the next steps in learning
or to report on the level of achievement attained to date. A rich research literature
shows us the dramatic achievement gains that can occur when formative assessment
is used (Black and Wiliam 1998; Shepard 2000). Most surveys of practice, however,
find that assessment is more often used for grading than for learning.
An ideal model of classroom assessment must address both content and process
considerations. The activities in which students engage and the work we ask them to
produce determine the real targets of learning regardless of what goals might be
stated in curriculum guidelines or lesson plans. Therefore, it is essential that the
content of instructional activities capture the big ideas and inquiry skills of the Na-
tional Science Education Standards. Formative assessments are then embedded within
these instructional activities. A student’s ability to communicate scientific informa-
tion might be assessed, for example, when presenting a group’s findings to the rest of
the class. As implementation of standards-based reform progresses, a bigger chal-
lenge is to ensure that summative classroom measures also mirror the standards. Too
often, classroom tests measure what is easiest to measure—vocabulary definitions
and restatement of laws and principles—rather than, say, the ability to use principles
and laws to make a prediction or explain a result. As suggested in the science stan-
dards document, improving the content of assessment means “assessing what is most
highly valued, assessing rich, well-structured knowledge, and assessing scientific
understanding and reasoning” (NRC 1996, 100). Effective classroom assessment
also departs from traditional practice in the way assessment is used, becoming much
more interactive and a part of the learning process. As documented in research stud-
ies (Black and Wiliam 1998; Pellegrino, Chudowsky, and Glaser 2001; Shepard 2000),
effective assessment
• activates and builds on prior knowledge,
• makes students’ thinking visible and explicit,

124 National Science Teachers Association

 CHAPTER 9

• engages students in self-monitoring of their own learning,

• makes the features of good work understandable and accessible to students, and
• provides feedback specifically targeted toward improvement.
These elements can be made a part of everyday instructional routines, using a
definition of formative assessment developed by Sadler (1989) and another recent
National Research Council report, Classroom Assessment and the National Science
Education Standards (Atkin, Black, and Coffey 2001). For assessment to be forma-
tive in the sense of moving learning forward, three questions are asked: (1) Where
are you trying to go? (2) Where are you now? (3) How can you get there? It is
because of the explicitness of these steps and the focused effort to close the gap
between 1 and 2 that assessment actually contributes to learning. Elsewhere I have
also argued that effective use of these strategies requires a cultural shift in class-
rooms so that students are less concerned about grades and hiding what they don’t
know and are more focused on using feedback and support from teachers and class-
mates to learn—that is, to solve a problem, improve a piece of writing, or figure out
why an answer is correct.
Finally, to be effective, classroom assessment will need to find ways to address
the many negative effects of grading on student motivation. Cognitive studies have
shown us that making criteria explicit will improve student-learning outcomes
(Fredericksen and Collins 1989). But motivational psychologists have found that
traditional grading practices may negatively affect students’ intrinsic motivation, their
sense of self-efficacy, and their willingness to expend effort or tackle difficult prob-
lems. Therefore, merely sharing grading criteria will not automatically eliminate the
negative effects of grading.
Unlike the extensive amount of work on formative assessment in recent years,
there has been much less attention, outside of the motivational literature, to the type
of grading policies that would improve rather than decrease motivation. Self-assess-
ment is one example of a change in classroom practice that could serve both cogni-
tive and motivational ends. Self-assessment makes the features of excellent work
explicit and helps students internalize these criteria (thus serving cognitive purposes).
At the same time, asking students to self-assess according to well-defined criteria
establishes a mastery rather than normative definition of success, conveys develop-
ing competence, and illustrates how effort could lead to improvement, all of which
enhance motivation (Stipek 1996). More work needs to be done to relate formative
and summative assessment within classrooms. Perhaps all formative assessment should
be reserved exclusively for learning purposes, not for grading—even while eventual
summative criteria are used formatively. Note that pursuit of this idea would run
against the highly litigious point systems that many teachers currently use to track
every assignment and to justify grades.

Everyday Assessment in the Science Classroom 125

CHAPTER 9

Purposes Served By Large-Scale Assessment

Large-scale assessments such as the Third International Math and Science Survey
(TIMSS), the National Assessment of Education Progress (NAEP), and various state-
and district-level assessment programs are used to measure student achievement for
aggregate units (nations, states, districts, schools), to track changes in achievement
for these units over time, and sometimes to measure the performance of individual
students. If the content of a large-scale assessment adequately represents ambitious
curricular goals—as called for in the science standards, for example—then large-
scale assessment can become an integral part of curricular reform and instructional
improvement efforts. Such an assessment program could be used to: exemplify im-
portant learning goals; diagnose program strengths and weaknesses; report on the
proficiency status of individual students; and, through associated professional devel-
opment opportunities, improve teachers’ abilities to teach to the standards and at the
same time become more adept in using formative assessment. These purposes would
not be served, however, by traditional, multiple-choice-only tests that do not ad-
equately embody the National Science Education Standards.

Exemplification of Learning Goals

The science standards developed a vision for science instruction by drawing on best
practices, but for many teachers the standards call for significant changes in prac-
tice—away from vocabulary-laden textbooks and toward more inquiry-based ap-
proaches. For many, these hoped-for changes may seem out of reach either
conceptually or practically. Large-scale assessments can give life to the standards
expectations by illustrating the kinds of skills and conceptual understandings that
students are expected to have mastered. Moreover, because some of the very best
assessment tasks would also qualify as good instructional activities, released assess-
ment items can help to raise awareness about the kinds of instructional opportunities
students need if they are to develop deep understandings and effective inquiry skills.
The performance task illustrated in Figure 1 is taken from A Sampler of Science
Assessment developed by Kathy Comfort and others in the California Department of
Education (1994). The task gives eighth-grade students hands-on experience with
subduction and asks them to generalize their understandings from the physical model
to information about California landmarks. One could reasonably expect that stu-
dents who had had previous instruction on geological processes and plate tectonics
would do well on this task. If, however, students with textbook exposure to these
ideas faltered in providing explanations, the assessment experience might prompt
teachers to consider using more conceptual learning tools in the future, and, in fact,
the investigation shown in Figure 1 is an example of the type of instructional activity
needed.

126 National Science Teachers Association

 CHAPTER 9

Figure 1. Grade-Eight Performance Task Illustrating Hands-On

Instruction and Assessment Focused on “Big Ideas”

(Continues on next page.)

Source: Reprinted, by permission, from A sampler of science assessment, copyright 1994,

California Department of Education, P.O. Box 271, Sacramento, CA 95812-0271.

Everyday Assessment in the Science Classroom 127

CHAPTER 9

(Figure 1. continued)

In some cases a single conceptual question, if used reflectively, can prompt teach-
ers to reconsider the efficacy of their instructional approach. In some sense, Phil
Sadler’s classic films, A Private Universe and Minds of Our Own are each based on
one significant conceptual question. Can you explain what makes the seasons? Can
you use a wire, a bulb, and a battery and make the bulb light? The fact that so many
Harvard graduates struggled with the first question, and MIT graduates with the
second, has prompted many science teachers to think again about what their students
are really understanding when they pass traditional tests. Thus, if a state assessment
reflects the National Science Education Standards it serves both as a model of what’s

128 National Science Teachers Association

 CHAPTER 9

expected for student mastery and also of the kinds of instructional activities that
would enable that mastery.
In preparing to write this chapter, I asked experts in several states to comment on
my outline of large-scale assessment purposes and to provide examples of each ap-
plication where appropriate. Rachel Wood, Education Associate in Science, and Julie
Schmidt, Director of Science, are part of the science leadership team responsible for
the development of Delaware’s Comprehensive Assessment Program. They responded
with a detailed commentary, recounting their experiences in involving science teach-
ers in development of summative assessments for curriculum modules (as part of the
National Science Foundation’s Local Systemic Change Initiative) concurrent with
development of the state’s on-demand test. Here’s what they said about the role of
assessment in leading instructional change.

What was not appreciated early on is that assessment would become the
driver for realizing what it meant to “meet the standards.” Initially
assessment was seen more as an appendage to curriculum. That was due, in
part, to the early recall nature of assessments that contributed minimally in
diagnosing student learning, whereas curriculum laid out a road map to
follow. Later (after the assessments changed dramatically), it was clearer
that assessment indicated whether you reached your destination or not. In
other words, the task of the leadership and its team was building a consensus
around quality student work in science. This consensus had to be founded
upon a different model of student learning than the model most teachers
possessed. (Wood and Schmidt 2002)

Program “Diagnosis”
It is popular these days to talk about making large-scale assessments more diagnostic.
Colorado’s Governor Bill Owens has said that he wants “to turn the annual CSAP
exam from just a snapshot of student performance into a diagnostic tool to bring up a
child’s math, reading, and writing scores” (Whaley 2002). And in the No Child Left
Behind Act of 2001, state plans are required to include assessments that “produce
individual student interpretive, descriptive, and diagnostic reports, … that allow par-
ents, teachers, and principals to understand and address the specific academic needs of
students.” In the next subsection, on individual student “screening,” I explain what
kinds of information a once-per-year test could reasonably provide on individual stu-
dents’ learning. We should be clear that large-scale assessments cannot be diagnostic
of individual learning needs in the same way that classroom assessments can be. What
large-scale assessments can do is “diagnose” program strengths and weaknesses. Typi-
cally we refer to this as the program evaluation purpose of large-scale assessment.
When content frameworks used in test construction have sufficient numbers of
items by content and processes strands, then it is possible to report assessment re-
sults by meaningful subscores. For example, it would be possible for a school to

Everyday Assessment in the Science Classroom 129

CHAPTER 9

know whether its students were doing relatively better (compared to state normative
data) on declarative knowledge items or on problems requiring conceptual under-
standing. It is also possible to report on relative strengths and weaknesses according
to content categories: life science, physical science, Earth and space science, science
and technology, and science in personal and social perspectives. This type of profile
analysis would let a school or district know whether its performance in technology
was falling behind performance in other areas, or whether there were significant
gender effects by content category. For example, we might anticipate that girls would
do better in science programs that emphasize the relevance of science to personal
and social perspectives, while boys might do relatively better in applications of tech-
nology. Results such as these might prompt important instructional conversations
about how to teach to strengths while not presuming that either group was incapable
of mastering material in their traditional area of weakness.
In addition to subtest profiles, particular assessment items can sometimes yield
important program diagnostic information. Wood and Schmidt (2002) provide the fol-
lowing examples of conceptual errors and skill weaknesses revealed by assessment
results that warranted attention in subsequent professional development efforts.

For instance, an eighth-grade weather assessment revealed that students

across the state have over-generalized their knowledge of the movement of all
air masses as having to go from west to east. In the classroom, students are
studying the movement of weather fronts and predicting weather patterns,
many of which do move from west to east. That piece of understanding has
now been applied to the movement of all air masses. They are unable to
explain ocean breezes on the east coast with this model or Bermuda highs
that they experience in their daily lives. This information was not uncovered
through a question about weather patterns in the United States but by using a
question on land and sea breezes. There is now an opportunity to address this
issue in professional development because this suggests that the idea
originates from some connection made in the classroom. This confirms what
we mentioned earlier, that students are indeed constructing knowledge in the
classroom that teachers might not be aware of unless they search for it. Most
teachers are probably delighted that students have the idea that most weather
fronts move from west to east, but were unaware that students would over-
generalize, unless the class has an opportunity to work through the limits of a
“rule” or model.

And a second example:

Analysis of item statistics from the state test reveals major weaknesses that
the leadership can address through professional development. For example,
questions asking students to construct or interpret a simple graph indicate

130 National Science Teachers Association

 CHAPTER 9

that students were not being given enough opportunities to graph data and
analyze the results, compare graphs, or draw conclusions from the kind of
graph that might appear in the newspaper, etc.… One item, for example, with
a P-value of .31 in simple graphing indicated an alarming weakness. A P-
value of .80 was expected. As a result the leadership selected graphing items,
rubrics, and samples of student responses with P-values to focus discussion
on the instructional implications of the student responses…. Some of the lead
teachers participated in the piloting of released items and were stunned that
their own students were performing at a level that confirmed the P-value
found for the whole state.

Because large-scale assessments are broad survey instruments, test makers often
have difficulty providing very detailed feedback for curriculum evaluation beyond
major subtest categories. This is especially true for assessments like TIMSS and
NAEP that cross many jurisdictions and may also be true for state assessments when
each district has its own curriculum. Cross-jurisdictional assessments invariably be-
come more generic in how they assess achievement, using questions that call for
reasoning with basic content (like on the ACT) rather than presenting the type of
question that would be appropriate in a specific course examination. The need for
items to be accessible to all students, regardless of what particular science curricu-
lum they have followed, explains why so many NAEP items, for example, involve
reading data from a table or graph to draw an inference or support a conclusion,
because such items are self-contained and do not presume particular content knowl-
edge. Unfortunately, generic, reasoning items are not very diagnostic nor do they
further the goal of exemplifying standards.
How then could we have more instructionally relevant items, like the earlier
California example? If state assessment frameworks were to stipulate specific in-
depth problem types they intended to use, there would a danger that teachers would
teach to the specific item types instead of the larger domain. Conversely, if different
in-depth problems were used each year representing the full domain, teachers would
be likely to complain about the unpredictability and unfairness of the assessment.
Again, I quote extensively from commentary by Wood and Schmidt (2002). They
have documented the power of released items (accompanied by student papers and
scoring guides) both to exemplify standards and to diagnose gaps in students’ learn-
ing. Here’s how they wrestled with the dilemma of fostering teaching to standards
without encouraging teaching to the test.

Many classroom teachers who haven’t had the opportunity to be directly

engaged in the lead teacher program hold a different view of the test and
items than those involved in the assessment development. For instance,
classroom teachers express frustration at the comprehensive nature of the
standards and not being able to determine “what items” are going to be on

Everyday Assessment in the Science Classroom 131

CHAPTER 9

the next test. They complain that we don’t release entire forms each year for
their students to practice in the classroom in preparation for the next year’s
test. What has been released and is preferable to release are not isolated
items matched to a standard, but an insightful commentary about how and
where the concepts in the released items fit into a larger sequence of student
conceptual understanding. Teachers will revert back to second-guessing the
test items if presented a released item decontextualized from an analysis that
helps explain how and why students are struggling with the concept that the
item is measuring. For example, when many high school students were
unable to construct a simple monohybrid Punnett square and determine the
genotypes of both parents and offspring, teachers could easily have thought,
“I taught them that, they should know it” or “I guess I need to teach more
Punnett squares”—which suggests that it is being taught in a mechanical
approach. But the commentary around the released item attempts to turn
teachers’ attention toward thinking about how students have acquired only a
mechanical sense and don’t understand why you would have a Punnett
square in the first place.

The example in Figure 2 shows how the analysis accompanying the released
item is intended to focus attention on underlying concepts that students might not be
understanding. “This particular item taps both procedural and conceptual knowl-
edge, while most teachers think it is only procedural knowledge” (Wood and Schmidt
2002). Because teachers focus on procedural knowledge, students assume the Punnett
square is an end in itself rather than a tool for reasoning through possible gene com-
binations. Lacking conceptual knowledge, they are likely to stack up illogical num-
bers of alleles in each cell. Wood and Schmidt’s analysis is intended to try and
reconnect the specific test question to a larger instructional domain, which should be
the appropriate target of improvement efforts.

Certification or “Screening” of Individual Student Achievement

Historically, many state assessment programs were designed to imitate the NAEP;
they provided broad content coverage and were used primarily for program evalua-
tion. NAEP does not produce individual student scores. In fact, using the strategy of
matrix sampling, each participating student takes only a small fraction of the items
in the total test pool so as to minimize testing time and ensure a rich representation of
the content domain. In recent years, under pressure to provide more accountability
information, many assessment programs have abandoned their matrix sampling de-
signs and instead give the same test to every student so that individual scores can be
reported. The No Child Left Behind Act requires all states to produce student scores
in reading and mathematics in grades three to eight, with testing in science in certain
grades to begin in 2007–2008. Individual reporting of students’ proficiency status is
a type of certification testing, not unlike a licensure test—with accompanying

132 National Science Teachers Association

 CHAPTER 9

Figure 2. A Released Item from the Delaware State Testing Program

(DSTP) with Scoring Tool and Instructional Analysis

LIFE SCIENCE
In the Life Science section of the DSTP [Delaware State Testing Program] students are required
to figure out the possible gene pairs that come from two parents. Often this type of genetics
word problem will require students to explain how dominant and recessive genes affect the
way traits are inherited. One of the released items from spring 2000 DSTP illustrates a genetics
question students are asked and what is required to earn complete credit.

Analysis:
After analyzing DSTP results from across the State, it appears that many students are struggling
with some of the same genetic concepts. For instance, when expected to construct Punnett
squares, students fail to separate the gene pair (alleles) of the parents. This error tends to
indicate that students are confused as to how meiosis affects the distribution of chromosomes
and subsequently genes. Once the students make this kind of mistake it is impossible for them
to determine all the gene pairs for a given characteristic that could come from a set of parents.
Furthermore, when students end up with gene combinations (inside the squares) that contain
more genetic information than the parents it does not seem to cue them into the fact that they
have done something wrong in setting up the Punnett square.
Students also experience difficulty with genetic problems when they are given phenotypic
patterns of inheritance and asked to derive information about the genotype of an organism (as
in the case of the released problem). Again, if students attempt to construct a Punnett square to
answer the question they must first be able to determine the genotype for each of the parent
organisms and then separate the alleles across the top and down the side of the square. After
completing the simple monohybrid crosses they should then be able to apply their understand-
ing of genetics to explain the relationships between the phenotypes and genotypes of the
parents and offsprings.

Released Item:
In cats, the gene for short hair (A) is dominant over the gene for long hair (a). A short-haired cat
is mated to a long-haired cat, and four kittens are produced, two short-haired and two long-
haired. Explain how the two parents could produce these offsprings.

Scoring Tool:
Response must indicate in words and/or in a correctly constructed Punnett square the appropri-
ate genotypes of both parents and the predicted offspring. For example:
2 points: One parent must be heterozygous and therefore, has a 50% chance of giving
the short-haired gene and a 50% chance of giving the long-haired gene. The
other parent can only give the long-haired gene. Therefore, 50% of the
offspring will be long-haired and 50% short-haired. Note: The words
“heterozygous” and “homozygous” are not required to receive full credit.

OR
a a

A Aa Aa

a aa aa

(Continues on next page.)

Everyday Assessment in the Science Classroom 133

CHAPTER 9

(Figure 2. continued)

OR
Parents: aa x Aa Offspring: 50% Aa 50% aa

1 point: Partially correct response, but some flaws may be included. For example, the
student may explain the parent with the dominant gene is carrying the
recessive allele, but the combinations inside the Punnett square do not reflect
separation of the alleles.

0 points: Incorrect, inappropriate, or incomplete response.

requirements for technical accuracy. When used for high-stakes purposes, tests must
be designed with sufficient reliability to yield a stable total score for each student.
This means that, within a reasonably small margin of error, students would end up in
the same proficiency category if they were retested on the same or closely parallel
test.
Reliability does not ensure validity, however. Especially, reliability cannot make
up for what’s left out of the test or how performance levels might shift if students
were allowed to work with familiar hands-on materials, to work in groups, to consult
resources, or to engage in any other activities that sharply changed the context of
knowledge use. Because no one instrument can be a perfectly valid indicator of stu-
dent achievement, the professional Standards for Educational and Psychological
Testing (AERA, APA, NCME 1999) require that high-stakes decisions “not be made
on the basis of a single test score” (146). While once-per-year state assessments can
be made sufficiently accurate to report to parents about a student’s level of achieve-
ment, they should not be used solely to determine grade-to-grade promotion or high
school graduation.
Can these state proficiency tests also be diagnostic at the level of individual
students? The answer is no, at least not in the same way that classroom assessments
can be diagnostic. Once-per-year survey tests are perhaps better thought of as “screen-
ing” instruments, not unlike the health screenings provided in shopping malls. If one
indicator shows a danger signal, the first thing you should do is see your doctor for a
more complete and accurate assessment.
The same subtest information that is available for program level profiling may
also be useful at the level of individual student profiles. Notice, however, that the
instructional insights provided earlier by Wood and Schmidt (2002) were based on
state patterns for large numbers of students. For individual students, it would be
inappropriate to interpret the results of single items, and even subtest peaks and
valleys are often not reliably different. Unfortunately, the most commonly reported
profiles do not reveal a particular area of weakness, where a student needs more
work. Instead, test results most frequently come back with predictable findings of
“low on everything” or “high on everything.”

134 National Science Teachers Association

 CHAPTER 9

I have explained previously that large-scale tests are too broad to provide (in one
or two hours of testing) much detail on a student’s knowledge of specific content or
skills—such as control of variables, formulating explanations, energy transfer, the
effect of heat on chemical reactions, the structure and function of cells, the relation-
ship of diversity and evolution, and so forth. An additional source of difficulty is the
match or mismatch between the level of a large-scale test and an individual student’s
level of functioning. Some state assessment programs are based on basic-skills tests
with relatively low-level proficiency standards. Low-level, basic-skills tests provide
very little information about the knowledge or knowledge gains of high-performing
students. In contrast, in states that built their tests in keeping with the rhetoric of
world-class standards, there will be few test items designed to measure the knowl-
edge or knowledge gains of below-grade-level students. NAEP, for example, was
designed to measure relatively challenging grade-level content, and therefore yields
unreliable total score estimates for students whose performance is below grade level.
I should also emphasize that the item sampling strategies currently used for fill-
out test frameworks are not designed with an understanding of learning progres-
sions. The authors of Knowing What Students Know (Pellegrino, Chudowsky, and
Glaser 2001) explained that current curriculum standards “emphasize what students
should learn, [but] they do not describe how students learn in ways that are maxi-
mally useful for guiding instruction and assessment” (256). Thus the fourth-grade
NAEP mathematics test is a sample of where students are expected to have gotten by
fourth grade, not how they got here. Models of student progression in learning have
been developed in research settings, but they have not yet been built into large-scale
testing programs. It would be a mistake, therefore, to try to make diagnostic deci-
sions from a fine-grained analysis of test results. Especially, one should not assume
that students should be instructed on the easy items on the test before proceeding to
the difficult items. Such reasoning would tend to reinforce instructional patterns
whereby slower students are assigned rote tasks and more able students are assigned
reasoning tasks. A more appropriate instructional strategy, based on comprehension
research for example, would ask lower-performing students to reason with simpler
material rather than delaying reasoning tasks. The appropriate learning continua
needed to plan instructional interventions cannot be inferred by rank ordering the
item statistics in a traditional test.
Given the inherent limitations of once-per-year, large-scale assessments, there
are only a few ways that large-scale assessments could be made more diagnostic for
individual students. Out-of-level testing is one possibility. This strategy would still
involve a standard test administration, but students would take a test more appropri-
ate to their performance level (such tests are statistically linked across students to
provide an accurate total score for a school even though students are taking different
tests). The state of Wyoming is one of a few states experimenting with a more ambi-
tious effort to make state assessments more instructionally relevant. Director of As-
sessment Scott Marion provided the example in Figure 3 of a curriculum-embedded

Everyday Assessment in the Science Classroom 135

CHAPTER 9

assessment. The state, along with the Wyoming Body of Evidence Activities Consor-
tium, developed 15–18 of these assessments in each of four core areas to be used to
determine if students have met the state’s graduation standards. Districts are free to
use these assessments or to develop their own as long as they meet alignment crite-
ria. The desirable feature of these assessments is that teachers can embed them where
they fit best in the high school curriculum, so long as students have had a fair oppor-
tunity to learn the necessary material and to demonstrate that learning. “Carmaliticus”
could be taken by ninth- or eleventh-grade biology students. Because these tasks
exemplify the science standards and are administered in the context of instruction,
teachers receive much more immediate and targeted information about student per-
formance than they do from more comprehensive large-scale assessments.
The Wyoming example also illustrates one of the inevitable trade-offs if state
assessments were to be made more diagnostic of individual student’s learning. More
diagnosis means more testing—so as to gather sufficient data in each skill and con-
tent area. More testing can perhaps be justified when it is closely tied to specific
units of study. But one could not defend the notion of 5–15 hours of testing for a
state-level science assessment. A reasonable principle to govern the design of exter-
nal tests would be the following: either large-scale assessments should be minimally
intrusive because they are being administered for program-level data, or large-scale
assessments must be able to demonstrate direct benefit to student learning for addi-
tional time spent. For policy makers who want more individual pupil diagnosis, this
principle leads to the idea of curriculum-embedded assessments administered at vari-
able times so that results can be used in the context of instruction. The only other
alternative is for states to develop curriculum materials with sample assessment tasks

Figure 3. A Curriculum-Embedded Assessment

Science Assessment Activity #7: Science Assessed:
Carmaliticus • Knowledge of
Introduction: To describe classification
evolutionary change and systems and
classification systems, scientists evolutionary change
use phylogenetic trees. Pictured • Ability to organize
[at left] is an example of the organisms into a
organization of a phylogenetic phylogenetic tree
tree into branches. according to
observable
In this activity, you will take on the role of a scientist developing a characteristics
phylogenetic tree to represent the evolutionary changes and
classification of an imaginary organism called a Carmaliticus.
Attached are the 66 imaginary organisms, called Carmaliticus. They are organized according to
Eras, indicated in the table below. The organisms and the Eras are not related to Earth’s
geologic time periods or the conditions within earth’s time periods.
(Continues on next page.)

Source: Property of the Wyoming Body of Evidence Activities Consortium and the Wyoming
Department of Education. Reprinted with permission.

136 National Science Teachers Association

 CHAPTER 9

Time in Millions
Eras Organism # of Years Ago

Era A 66 245–209

Era B 64–65 208–145

Era C 60–64 144–67

Era D 53–59 66–58

Era E 43–52 57–37

Era F 29–42 36–24

Era G 15–28 23–6

Era H 8–14 5–2

Era I 4–7 1–.1

Recent– 1–3 Present

Still Living

Part I – Phylogenetic Tree: Organize the Carmalitici into a phylogenetic tree according to Eras
and characteristics of the Carmaliticus. On the tree, link each organism to only one organism
from the previous Era, with a line; and indicate the extinction of a branch, with a labeled line.

Part II –Written Explanation: Provide a written report with your phylogenetic tree that
includes the following:
1) The reasoning you used to make decisions regarding placement of the Carmaliticus and
their branches;
2) For two branches with seven or more Carmaliticus, describe how one organism evolved to
another—based on identifiable characteristics of the organisms;
3) Possible environments of four Eras, supported with characteristics of the organisms that
would justify your decisions;
4) A comparison of your phylogenetic tree to one other tree produced by a classmate. In your
comparison, you are to identify at least two significant differences between your tree and the
other tree, including a description about the difference in the organization and characteristics
of all of the organisms within at least one branch and a comparison of the branches.

NOTE - Important considerations as you develop your phylogenetic tree:

a) Consider the organization of the entire tree before attaching the Carmaliticus.
b) Neatness and spacing will make a difference when you have to examine and explain the
individual characteristics and the overall trends of the tree.
c) Based upon assumptions you make in the development of your tree, it is unlikely that
you and another classmate will have an identical tree.
d) Each organism should only be tied to one other organism from the previous Era.

(Continues on next page.)

Everyday Assessment in the Science Classroom 137

CHAPTER 9

(Continues on next page.)

138 National Science Teachers Association

 CHAPTER 9

(Continues on next page.)

Everyday Assessment in the Science Classroom 139

CHAPTER 9

140 National Science Teachers Association

 CHAPTER 9

for teachers to use to check on student progress but not to be used in formal data
collection.

Professional Development
Professional development associated with standards-based reforms has tended to focus
on the intention of the standards (Why should students be able to communicate math-
ematically?), and on curriculum materials and instructional strategies to implement
the standards (What does inquiry-based instruction look like?). Assessment activi-
ties tied to standards have the potential to deepen teachers’ understandings of the
meaning of standards as well as to provide the means to improve student learning.
The additional goal of having teachers become more adept at using specific forma-
tive assessment strategies can also be furthered by professional development that
addresses content standards. There are two important reasons for embedding teach-
ers’ learning about assessment in larger professional development efforts—one prac-
tical, the other conceptual. First, teachers’ time is already overburdened. It is very
unlikely that teachers could take time to learn about formative assessment strategies
in a way that is not directly tied to the immediate pressures to raise student achieve-
ment on accountability tests. Second, assessment efforts only make sense if they are
intimately tied to content learning. Therefore, assessment learning can be under-
taken in the context of helping teachers improve performance on a state test, so long
as we clearly understand the difference between teaching to the standards and teach-
ing the test.
Folklore of advanced placement (AP) examinations has it that some teachers
return to Princeton year after year to participate in the scoring of AP exams because
of the learning experience. Not only is it important to see what kinds of questions are
asked, but it makes one a better teacher to engage with student work and to discuss
with one’s colleagues how to interpret criteria in light of specific student perfor-
mances. In this same vein, Wood and Schmidt (2002) describe several different as-
pects of professional development that occurred in Delaware when teachers were
involved in assessment development, pilot testing, and scoring. First and foremost,
“teachers became hooked on student learning.” By focusing on what students were
learning, they moved from being good at delivering inquiry-based instruction to fo-
cusing on what students were actually learning from that instruction. For example,
teachers learned to use double-digit rubrics that produced both a score and the rea-
son for the score, “which completely transformed our thinking.”

A single-digit rubric just lumps partially correct responses together and

doesn’t discriminate between the milder and more serious partially correct
or wrong responses. The diagnostic rubrics are ordered so that teachers
score student work and easily flag the most frequent missteps in student
thinking. This kind of diagnostic information is not available from a single-
digit rubric that is so holistic that it fails to identify that students get things

Everyday Assessment in the Science Classroom 141

CHAPTER 9

wrong or right for different reasons. Making explicit an array of student

thinking around a question forces the teachers to think about the implications
for instructional practice. This characterization of student learning resonated
more with teachers at a gut level than the daunting but somewhat intriguing
collection of student thinking documented in the research base. (Wood and
Schmidt 2002)

Dr. Maryellen Harmon, a consultant to Wood and Schmidt’s (2002) project, re-
quired that teachers who were developing summative assessment tasks take time to
write out what each item is measuring and also to write out their own “elegant an-
swers” to each question before developing the scoring criteria and rubric. From these
“academic exercises,” teachers found they could catch flaws in their own thinking
and sometimes reconsidered whether the target was even worth measuring in the
first place. They also became aware of how students would struggle when they them-
selves could not agree on what was being asked or required for a complete response.
During pilot testing, Dr. Harmon also coached teachers to learn from student re-
sponses and not always blame the students when they couldn’t respond. Although
Wood and Schmidt focused on whether learning from the summative assessment
project could be generalized to developing items for the state test, these skills could
as likely be generalized to developing better classroom assessment. As a result of
these experiences, “teachers were much more willing to pilot potential DSTP [Dela-
ware Student Testing Program] test items prior to submitting them and were more
aware of how to interpret student work. Many of these teachers now write out a
“what this test measures …” when they construct an item for the state test. They are
much less likely to blame a student for an unanticipated response and more likely to
reexamine their question and rubric.”
The assessment development process and pilot testing experiences described by
Wood and Schmidt (2002) show us the power of real professional development op-
portunities as compared to merely receiving student scores from a state test. “When
lead teachers had to score student work from a unit that was just taught, teachers had
to evaluate both the extent to which students had acquired certain concepts as well as
reflect on their own teaching strategies for particular lessons.” For example, “teach-
ers had assumed that students could trace the path of electricity in a complete circuit.
When their own data contradicted their assumptions, they realized the need to ad-
dress this learning in another way with their students.” Most tellingly, teachers had
to face the dissonance between what they had taught and what students had learned:

After all, these teachers knew that good science was happening in their
classrooms—they were using NSF materials, had undergone the training on the
modules, and were comfortable with the content knowledge now and employing
inquiry-based strategies. The students were active in their learning and enjoyed
the lessons immensely. Imagine the impact of data that confronts and challenges

142 National Science Teachers Association

 CHAPTER 9

their confidence in knowing what their students know. It was Shavelson (also a
consultant to the project) who encouraged the leadership to let teachers struggle
through this new “problem space” because ultimately that is where all learning
occurs. An opportunity to discuss not only their students’ learning but similarly
situated students’ learning with other teachers using the same units has proven to
be a key ingredient for realizing Fullan’s idea of assessment conversations and is
a more powerful mode of professional development than learning the modules
and inquiry-based teaching without this aspect. (Wood and Schmidt 2002)

To summarize, then, professional development focused on assessment of student

learning can be a powerful tool to help teachers move beyond merely implementing
inquiry activities to an increased awareness of what their students are getting from
the activities. Given the layers of assessment-related demands already faced by teach-
ers, efforts to improve classroom assessment strategies should be woven into stan-
dards-based professional development and curriculum development. Teachers need
better access to materials that model teaching for understanding—with extended
instructional activities, formative assessment tasks, scoring rubrics, and summative
assessments built in. And, as illustrated by Wood and Schmidt’s (2002) experiences,
they need extended support while attempting to use these materials and draw infer-
ences about how to improve instruction.

Conclusion: Impediments and Recommendations

The single most important requirement to increase the likelihood that large-scale as-
sessments will contribute positively to student learning is to improve the substance of
what is assessed. If large-scale assessments were to embody important learning goals—
not only inquiry skills, but also the important big ideas in content areas, geological time
scale, photosynthesis, why electric current is different from “flowing” water, why we
isolate smallpox patients and not AIDS patients—then other aspects of the assessment,
such as program evaluation profiles, released item insights, and professional develop-
ment, can also be used to improve instruction. In Knowing What Students Know,
Pellegrino, Chudowsky, and Glaser (2001) argued that for an assessment system to
support learning, it has to have the feature of coherence. That means that classroom and
external assessments have to share the same or compatible underlying models of stu-
dent learning; otherwise, as in the present-day system, they will work at cross purposes.

While a large-scale assessment might be based on a model of learning that is

coarser than that underlying the assessments used in classrooms, the
conceptual base for the large-scale assessment should be a broader version
of one that makes sense at the finer-grained level (Mislevy 1996). In this way,
the external assessment results will be consistent with the more detailed
understanding of learning underlying classroom instruction and assessment.
(Pellegrino, Chudowsky, and Glaser 2001, 255–56)

Everyday Assessment in the Science Classroom 143

CHAPTER 9

In attempting to pursue this vision of an ideal assessment system, science educa-

tors should be aware of several potential obstacles:
• Cost
• The No Child Left Behind Act’s mandate for testing every pupil (with no out-of-
level testing)
• Technical standards and legal protections
• Curriculum control
• Lack of trust of teachers as evaluators
• Beliefs held by policy makers about standards-based reform
• Mechanical data systems
Substantively ambitious assessments can be developed and scored reliably for
large-scale purposes, but they invariably cost more than machine-scored, multiple-
choice tests. Passage of the No Child Left Behind Act has so markedly increased the
amount of testing required that we are likely to see a continuing decline in the sub-
stantive quality of large-scale tests, because state agencies often cannot afford to do
better. Science educators have the advantage that science will be assessed less fre-
quently than reading and mathematics, and therefore, it is more feasible to advocate
for high-quality science assessments. Technical standards and legal protections also
tend to work against the quality of assessments simply because trivial things are
more easily measured consistently. Therefore, the case will have to be made as to
why better assessments are worth the investment (i.e., why it’s worth it to spend the
extra money to measure important things consistently).
Other obstacles to assessment reform include issues of curriculum control and
lack of trust of teachers as evaluators. Successful implementations of substantively
ambitious assessments, such as the New Standards Project (1997) and the Educa-
tional Testing Service’s Pacesetter program, have moved much closer to curriculum
development than traditional test construction, which merely collected test items. A
problem arises then, when states make the tests and districts control curriculum,
about how to achieve the kind of coherence envisioned by Pellegrino, Chudowsky,
and Glaser (2001). Similarly, teachers gain more from assessments when they are
involved in providing data, and teacher participation makes it more likely that as-
sessments can include extended tasks grounded in classroom work. Therefore, in-
cluding portfolio and project data would increase the validity and meaningfulness of
a large-scale assessment. But because of distrust, which motivates the accountability
movement in the first place, proponents of substantively richer assessments will have
to think of safeguards, such as score moderation schemes that verify the accuracy of
teacher-reported data, to counter the claim that teachers might misrepresent student
achievement.

144 National Science Teachers Association

 CHAPTER 9

Finally, there is the difficulty that policy makers may hold very different beliefs
about standards-based reform than those who originally advocated for conceptually
linked curriculum and assessment reforms. While originators like Smith and O’Day
(1990) and Resnick and Resnick (1992) were clear about the need for what they
called capacity building, including substantial professional development for teach-
ers, many present-day policy makers have adopted an economic incentives model as
their underlying theory of the reform. Those holding the latter view are unlikely to
see the need to invest in curriculum development or professional training. Add to this
picture the fact that “data-driven instruction” is being marketed more aggressively
than are rich assessment and curriculum units. Using data to guide instruction is, of
course, a good thing. Investing in mechanical data systems is a mistake, however, if
they are built on bad tests. There is no point in getting detailed disaggregations of
test data when test content bears little resemblance to valued curriculum. Trying to
make sense of this cacophonous scene will be difficult. What one should advocate
for will clearly be different in each state depending on the quality of the existing
large-scale assessment and likelihood of persuading state-level decision makers to
invest in instructionally relevant curriculum development and professional training.
If science educators want to move toward large-scale assessment that is concep-
tually linked to classroom learning, what should they be for? They should advocate
for a good test that embodies the skills and conceptual understandings called for in
the science standards. A rich and challenging assessment could take the form of
curriculum-embedded assessments or be a combination of state-level, on-demand
assessments and local embedded assessments, projects, and portfolios as in the New
Standards Project (1997). As advocated in Knowing What Students Know (Pellegrino,
Chudowsky, and Glaser 2001), there should be a strong substantive coherence be-
tween what is called for in the state assessment and what is elaborated in local in-
structional units and classroom assessments. To realize the full potential for teacher
learning, professional development should be provided that uses the power of as-
sessment to look at student work and to redesign instruction accordingly. Teachers
should have access to curriculum materials that reflect inquiry-based instruction with
well-conceived assessment tools built in. And they should have supported opportuni-
ties to try out new instructional materials and formative assessment strategies.
What if the state has a bad test? Then the strategies for science educators should
be quite different. In fact, the goal should be to reinvigorate the intended goals for
learning and to be explicit about what would be left out if we focused narrowly on
the curriculum implied by the test. Groups of teachers or curriculum specialists might
want to go through this exercise of mapping the state test to the science standards.
Then they could ask, What support is needed to ensure that instruction focuses on the
standards rather than the test, and what evidence will we provide to parents and
school board members to educate them about important accomplishments not re-
flected in the test?

Everyday Assessment in the Science Classroom 145

CHAPTER 9

Ultimately the goal of any assessment should be to further student learning. Class-
room assessments have the greatest potential for directly improving learning be-
cause they can be located in the midst of instruction and can provide timely feedback
at just the point of a student’s uncertainty or incomplete mastery. Large-scale assess-
ments can also support the learning process, but to do this they must faithfully elicit
the knowledge, skills, and reasoning abilities that we hope for students to develop,
and they must be linked in a well-articulated way to ongoing program evaluation and
professional development.

References
American Educational Research Association (AERA), American Psychological Association (APA),
National Council on Measurement in Education (NCME). 1999. Standards for educational and
psychological testing. Washington, DC: American Educational Research Association.
Atkin, J. M., P. Black, and J. Coffey. 2001. Classroom assessment and the national science education
standards. Washington, DC: National Academy Press.
Black, P., and D. Wiliam.1998. Assessment and classroom learning. Assessment in Education 5(1): 7–
74.
California Department of Education. 1994. A sampler of science assessment. Sacramento: California
Department of Education.
Fredericksen, J. R., and A. Collins. 1989. A systems approach to educational testing. Educational
Researcher 18: 27–32.
Mislevy, R. J. 1996. Test theory reconceived. Journal of Educational Measurement 33(4): 379–416.
National Research Council (NRC). 1996. National science education standards. Washington, DC:
National Academy Press.
New Standards Project. 1997. Performance standards: English language arts, mathematics, science,
applied learning. Vols. 1–3. Washington, DC: National Center for Education Statistics and the
University of Pittsburgh.
Pellegrino, J. W., N. Chudowsky, and R. Glaser. 2001. Knowing what students know: The science and
design of educational assessment. Washington, DC: National Academy Press.
Resnick, L. B., and D. P. Resnick. 1992. Assessing the thinking curriculum: New tools for educational
reform. In Changing assessments: Alternative views of aptitude, achievement, and instruction.
Boston: Kluwer.
Sadler, R. 1989. Formative assessment and the design of instructional systems. Instructional Science
18: 119–44.
Shepard, L. A. 2000. The role of assessment in a learning culture. Educational Researcher 29(7): 4–
14.
Smith, M. S., and J. O’Day. 1990. Systemic school reform. In Politics of education association year-
book 1990, 233–67. London: Taylor and Francis.
Stipek, D. J. 1996. Motivation and instruction. In Handbook of educational psychology, eds. D. C.
Berliner and R. C. Calfee, 85–113. New York: Macmillan.
Whaley, M. 2002. Owens looks to broaden CSAP focus: Governor wants student-performance test to
become tool for individual improvement. Denver Post, 14 March.
Wood, R., and J. Schmidt. 2002. History of the development of Delaware Comprehensive Assessment
Program in Science. Unpublished memorandum.

146 National Science Teachers Association

 CHAPTER 10

Reflections on Assessment
F. James Rutherford

James Rutherford recently retired from the American Association for the Advancement of Sci-
ence, where he was chief education officer and the originator and director of Project 2061.
Leading up to that position, he was a teacher of high school science, president of the National
Science Teachers Association, professor of science education at Harvard University and at New
York University, assistant director of the National Science Foundation, and assistant secretary of
the U.S. Department of Education. He also initiated and directed the Carnegie Science-Humani-
ties Project, Harvard Project Physics, and Project City Science.

P erhaps it is the battle cry for “accountability.” Or perhaps it is the infusion of the
empirical spirit of science into all walks of life. Maybe it is simply the natural
offspring of “standards.” But whatever the cause, assessment has become the educa-
tional issue of the day. Hence this National Science Teachers Association compila-
tion of essays by leaders in the field of educational assessment.
This particular chapter consists of comments and questions triggered by my read-
ing of Chapters l through 9. The first two sections below expand briefly on notions
that I believe provide a larger context for “everyday classroom assessment.” One
notion is that there is now considerable confusion on the assessment scene due to
mounting demands, conflicting expectations, daunting constraints, and lack of agree-
ment on the need for and character of K–12 assessment in science. The other claim I
wish to make is that there simply is no one best method of assessment and that each
must be considered critically in terms of the main purpose it is intended to serve. I
conclude by suggesting some questions for readers to consider as they discuss with
one another the claims and propositions set out in this book.

The Changing Assessment Scene

Assessment—“testing” in simpler times—is not, of course, something new on the
educational scene. Quite the opposite, for it has surely been around in one form or
another for as long as teaching itself. But since the Second World War the circum-
stances bearing on educational assessment have changed dramatically. These cir-
cumstances have included rapid demographic shifts, increasing educational demands,
the accumulation of findings from cognitive research, and the episodic nature of
post-war efforts to improve science education in response to a sequence of national
crises attributed to the failure of American schools.

Demographics
It is difficult enough to determine how each child is progressing in classrooms in
which the students share similar social, economic, language, and cultural backgrounds.
And it is often equally difficult to know what action to take in response to assess-
ment outcomes. Imagine, then, how vastly more difficult assessment (and basing

147
CHAPTER 10

teaching on assessment results) must be when, as in today’s schools, the students

differ greatly from one another in so many ways.
From time to time, the medical metaphor—often disguised linguistically—re-
surfaces in education circles, for surely in principle it makes sense. (Feasibility, how-
ever, is another matter.) Teachers are essentially advised to “diagnose” each student’s
learning condition and then to “prescribe” individual remedies to respond to the
diagnosis. Not surprisingly, in time the idea fades away for a while, since teachers
are not yet really in a position to either diagnose or treat with precision. Assessment
can detect whether or not a student has learned something, but rarely determine the
depth and sophistication of the learning, and certainly not whether the learning is
temporary or lasting. In any case, an assessment showing defective learning is not a
diagnosis, anymore than the detection of high blood pressure is. And treatment de-
pends on a body of dependable clinical research findings and codified experience
that, in education, have yet to reach the level of that found in medical practice.
This is not to counsel despair but rather to urge caution in setting assessment
goals. With greater investment in the development of assessment procedures and
research on learning, substantial advancement can be made over the next few de-
cades, gradually providing teachers with an increasingly scientific basis for their
work. In the meantime, it is well to remember that while it has put new—and not yet
well understood—demands on teaching and assessment, the extreme student diver-
sity now in our schools is valuable precisely because it is a reflection of the demo-
graphics of our society. Any assessment approach that does not make thoughtful
efforts to take the changing demographics explicitly into account should be viewed
with skepticism.

Educational Demands
By the end of the Second World War, the United States was well on the way to
shifting from an economy based largely on agriculture to one based increasingly on
industrial manufacturing and commerce. The schools in time responded to that change.
“Shop” became “industrial arts,” for instance, and the proportion of students taking
the college preparatory curriculum and graduating from high school increased. The
status of science education grew, and as it did, more students took science than be-
fore, and new content was added to the existing science and mathematics courses.
This trend—think of it as expecting more science and mathematics from more
students—was accelerated with the shift to the so-called information-based economy
and rapid advances in science. Still more new content, therefore, was added to exist-
ing science courses (little content ever being removed, subtraction seemingly out of
the question in education), and advanced placement courses proliferated. Thus even
in the face of a K–12 student body that was becoming progressively more diverse—
with a greater percentage of students with learning disabilities and of students (par-
ticularly in rural and inner-city schools) from low socioeconomic families and immi-
grant families from Third World countries—the curriculum became more demanding.

148 National Science Teachers Association

 CHAPTER 10

As we well know, ambitious national education standards were eventually formu-

lated to codify these rising expectations, followed by state standards that are more or
less based on the national standards, followed by legislation requiring more testing
more often of more students. It is interesting to note, by way of contrast, that in
California just prior to the war (at which time, incidentally, California was widely
regarded, whether correctly or not, as having the country’s leading educational sys-
tem), students faced no examinations other than those that individual teachers them-
selves decided to give. It would seem that as confidence in schools and teachers
shrinks, the demand for testing grows—or could it be the other way around?

Learning Studies
The working assumption of teachers through the decades has been that they could
determine accurately how well their students were learning what they were being
taught. That is what all those final examinations are about. But beginning in the late
1970s researchers started looking more closely at learning, and in due course the
research findings started showing up in refereed journals. After such studies had
accumulated for about a decade, Project 2061, the long-range reform project of the
American Association for the Advancement of Science (AAAS), examined all of the
cognitive science research literature (in English, and some in other languages) deal-
ing with science, mathematics, and technology learning at the K–12 ages (AAAS
1993).
The picture that emerged was not reassuring. For one thing, it turned out that the
studies were not well distributed but strongly skewed toward physics and mathematics.
Furthermore, the published papers were uneven: some met rigorous scientific stan-
dards but many—as determined by an external panel of researchers—were flawed
methodologically, and therefore not sound enough to be used in the AAAS analysis.
What the review told us is that we had a long way to go before we had as much trust-
worthy knowledge as we would like, in a scientific sense, about what students actually
learn as a consequence of K–12 science instruction. This has been generally borne out
since then in the cognitive literature on science and mathematics learning (AAAS 1997).
More disturbing, perhaps, the research seems to show that students take away
from school most of the mistaken science notions with which they started. In short,
they seem not to learn what we think we have taught them. Unfortunately, relatively
few teachers are well enough informed on the common misconceptions students
harbor to take them seriously into account in their teaching. This is dramatically
borne out in a series of videos produced by the Harvard-Smithsonian Center for
Astrophysics. For example, the first of these, A Private Universe, demonstrates that
upon graduation even Harvard students, most of whom had attended the nation’s
most prestigious private and public schools, still hold inaccurate ideas about what
causes the seasons and the phases of the moon. Not to be outdone by their crosstown
rivals, new MIT graduates were found to give wrong answers to the question of

Everyday Assessment in the Science Classroom 149

CHAPTER 10

where the mass of a log comes from. (These videos are available from the Annenberg/
CPB Math and Science Collection, P.O. Box 2345, S. Burlington, VT 05407-7373.)
At the very least, learning studies reveal that we must be very careful indeed in
making claims about what students have learned. They also demonstrate just how
difficult it is to determine learning outcomes convincingly. This, in turn, suggests
that we need to develop an array of assessment approaches that may not be as scien-
tifically rigorous as those used in cognitive science but that are explicitly designed to
serve different assessment purposes. And in the bargain, we need to develop ways of
assessing them—which is to say ways of assessing the assessments—rather than
uncritically accepting their validity and reliability. In sum, serious learning studies
have complicated the assessment picture but have clarified both the limits and prom-
ise of assessment.

National Crises
There is always reason enough to evaluate carefully what learning is taking place in
our schools and to report the results honestly to those who have a legitimate interest
in knowing. Ideally, educators should be expected to work continuously, year after
year, at improving the teaching, curriculum, and teaching materials in the nation’s
schools. To accomplish that, they need to work steadily to improve their ability to
assess student progress accurately. But steady, continuous upgrading of assessment
has not been the reality—not, at least, in the United States.
As it turns out, most reform efforts in science education—including those focus-
ing on assessment—seem to be driven by a sense of national crises. Hence serious
attention to testing is episodic rather than continuous and carried out in an atmo-
sphere of urgency and distrust. There has never been a time when American K–12
education was not sharply criticized, the subject usually being the “basics” of read-
ing, writing, and arithmetic or the feud between the advocates of “life adjustment”
and “liberal arts” curricula. For some classics in this genre, see Educational Waste-
lands (Bestor 1953), Quackery in the Public Schools (Lynd 1953), and Education
and Freedom (Rickover 1959). However, in times of national crises, real or imag-
ined, for which education takes the blame, deserved or not, the charges of inad-
equacy become more shrill and ubiquitous.
With the end of the Second World War, the importance of science and mathemat-
ics (as distinct from basic arithmetic) education came into sharp focus. In 1950, the
National Science Foundation came into being, charged by Congress with supporting
the advancement of science education as well as scientific research. The nation was,
it seemed at the time, on its way to the radical improvement of science and math-
ematics education.
But in fact the public was not yet quite ready for science education reform on a
national scale. Some reform projects got underway in the late 1950s, notably the
Physical Science Study Committee in 1956 and the Biological Sciences Curriculum
Committee in 1958 (Randolph 2002). Most of the dozens of “alphabet” reform projects

150 National Science Teachers Association

 CHAPTER 10

(and the summer and academic year institutes for science and math teachers associ-
ated with them) followed a bit later, owing their existence to the shock of Sputnik in
October 1959. That disturbing Soviet Cold War accomplishment quickly became the
basis of a national educational crisis no less than an engineering one. We were fall-
ing behind the Soviet Union, the press and public opinion soon had it, because our
schools were inferior and had let us down by not producing more scientists and
engineers. The response was an unprecedented national effort to boost American
science and mathematics education at every level. The nation was now truly on its
way to science education reform.
Crises, however, have a tendency to go away. Reaching the Moon first had be-
come accepted in the United States as the definitive test of which country would win
the space race. The United States won that race decisively, and, with the sense of
crises gone, the science education reform movement coasted to a halt. There were, to
be sure, other factors leading to the decline of interest in the reform movement.
Among them, for example, was the rancorous debate over “Man, A Course of Study,”
a social studies reform project of the period. The Transformation of the School: Pro-
gressivism in American Education, 1876-1957 (Cremin 1961) provides a historical
perspective preceding the Sputnik episode. Subsequent developments can be found
in Schoolhouse Politics: Lessons from the Sputnik Era (Dow 1991) and The Troubled
Crusade: American Education, 1945–1980 (Ravitch 1983).
But the main point here is that many of the initiatives of the Sputnik-era reform
projects—among them the efforts of the projects to come up with more meaningful
approaches to assessment—did not have time to mature and become entrenched in
the schools before political support for them largely disappeared.
Instances of novel approaches to assessment can be found in the materials of
many of the projects and are best examined in conjunction with the instructional
components of each, especially since some concentrated on assessing performance
rather than acquired knowledge. Two examples can be drawn from the Project Phys-
ics Course developed at Harvard University. In one case, instructors showed students
the same filmed, real-life events at the beginning and end of the course and asked
them to describe in writing what they saw and to offer explanations for what hap-
pened in the film. Conceptual and language changes were to provide evidence of
what physics principles had been learned. This technique showed promise, but the
demise of the reform movement made it impossible at the time to obtain support to
pursue this line of research further. (Needless to say, the development of computer
technology has now made it possible to develop powerful interactive assessment
activities that go far beyond the limited possibilities of 16mm film.)
The Project Physics Course also developed a set of student-choice examinations.
Each of the six units making up the course was accompanied by a booklet containing
four different tests from which students could individually select the one they felt
would best allow them to display what they had learned. The four tests differed from
each other in style and emphasis: One was entirely multiple choice, one was made

Everyday Assessment in the Science Classroom 151

CHAPTER 10

up of problems and essay questions, and two were a combination of objective and
subjective styles, with one emphasizing the historical, philosophical, and social as-
pects of the content and the other the quantitative and experimental. The idea was to
move teachers away from comparing students and toward assessing individual stu-
dents in terms of their progress based on their own expression of what they had
learned.
After a reform lull, another crisis provoked another energetic round of K–12
science education reform. The economies of Japan and Germany seemed to be sur-
passing the U.S. economy. This, too, was attributed largely to the failure of the Ameri-
can public schools. The spirit of the time was fully captured in the influential report
A Nation at Risk (U.S. Department of Education 1983). But the U.S. economy even-
tually reasserted itself (for which, by the way, education received no credit), and
once again the reform movement slackened. Eventually, a new and quite different
crisis came to the rescue: An international study documented how badly we do in
school science and mathematics compared to other nations (Schmidt, McKnight,
and Raizen 1997). Reform was on again—this time an educational crisis provoked
by educational assessment itself. It is still in play, and hence too early to tell how
long it will last and what form it will settle on. It is clear, however, that this on-again-
off-again commitment to K–12 education reform has made it difficult to sustain a
serious effort long enough to accomplish lasting reforms in general, let alone in
science education assessment practices.
A result of all of this—demographic changes, increasing learning demands, the
revelations of cognitive research, and the spasm-like ups and downs of reform—is
assessment in disarray, if not quite chaos. Presidents and governors, federal and state
legislatures, CEOs of great business enterprises, newspaper editors, and syndicated
columnists are again vocal in demanding school reform.
But the voices are not nearly in harmony. Many assert that the only way we can
regain confidence in the schools is for them to provide the nation with hard evidence
that our students surpass those in other countries, becoming, as it was famously put
by the U.S. governors in 1989, the best in the world by the year 2000 (U.S. Depart-
ment of Education 1989). Others take the position that the only way to motivate the
education system to do better is to confront the schools with frequent and demanding
external tests. Then there are those (not too numerous at the moment) who argue for
less external testing on the grounds that the real problem is the quality of assessment
policies and procedures. And in the background, educators themselves are not of a
mind on such science assessment issues as (a) testing for content knowledge versus
testing for inquiry skills and (b) assessment against fixed standards versus against
“the curve.”

Assessment Purposes
The turmoil in the current assessment scene is exacerbated, I believe, by the failure
of many of those who are demanding assessment-based accountability to recognize

152 National Science Teachers Association

 CHAPTER 10

that every form of assessment has its own capabilities, limitations, and side effects
and that no single approach is best for all educational purposes (see Paul Black,
Chapter 1 in this book). To be effective, assessment should be designed with a clear
view of just what service it is intended to perform and for whom. Assessment that is
appropriate for helping teachers improve classroom teaching and learning (the sub-
ject of this book) is significantly different from assessment that is appropriate for
informing parents and certain others persons of student performance or for deter-
mining how well the nation’s schools or those in a given state or local system are
doing (AAAS 2001c).

Improving Classroom Teaching and Learning

The overriding purpose of classroom teaching is to promote student learning of se-
lected understandings and skills. To that end, teachers need to determine how each
student is progressing, as well as how the class as a whole is doing, and then know
what to do with that information in order to improve teaching and learning. That is a
daunting task, but clearly the more direct and immediate the feedback that teachers
are able to obtain on student progress, the better situated they are to adjust their
teaching, if necessary, to get better results. In this sense, assessing student learning is
also assessing teacher performance.
But no less than external testing, classroom assessment can be intrusive and
have unwanted side effects, such as distracting students and wasting time. Its use as
punishment is, alas, not unknown. It can persuade students that the teacher’s role is
more to assess them than to help them learn. At risk is the trust that should exist
between teachers and their students, a trust that is akin to that between physicians
and their patients and between lawyers and their clients. Because thoughtful and
purposive assessment is so important for improving classroom teaching and learn-
ing, and yet so prone to misuse, the professional training of teachers should arm
them with a sophisticated view of assessment and with the knowledge and assess-
ment skills they will need to be effective teachers.
The previous chapters in this book describe aspects of an approach to classroom
assessment that is minimally intrusive and closely connected to the content and skills
being taught; the approach also contributes to the ability of students and teachers to
evaluate their own performance and act on that knowledge. It should be kept in mind,
however, that classroom assessment of the sort discussed in those chapters can, at
best, produce weak estimates of the performance of any larger dimension of the
educational system, such as the school, school district, state, or nation. But of course,
by the same token, externally formulated, high-stakes tests can contribute little to
individual student learning in the science classroom.

Informing Parents and Others

Most of the time, a very special and sensitive relationship exists between teachers
and their students, which is why classroom assessment must be handled with such

Everyday Assessment in the Science Classroom 153

CHAPTER 10

care. But that relationship does not exist in total isolation. Certain other persons
outside the classroom have a need (and a right) to know about individual student
performance. Foremost among these, of course, are parents. In addition, at some
point university admission officers or prospective employers will want the kind of
information about applicants that can be used in making admission and employment
decisions.
Note that for these purposes, detailed information of the sort appropriate for
improving teaching and learning is rarely needed. Parents are generally satisfied
with letter grades and discussions with teachers regarding the progress (and behav-
ior) of their children. As for admission officers and employers, such information as
courses taken, grades, honors, activities, teacher recommendations, attendance pat-
terns, and SAT scores are sufficient. Thus third parties have little need for the inti-
mate details of learning that are so necessary for teachers and students.
On the other hand, one might expect that parents would care greatly about how
their children do on statewide or national tests relative to other children, and indeed
that seems to be the case. Unfortunately, such tests reveal less about individual stu-
dent learning than what well-prepared teachers can provide on the basis of their own
assessments over the year. It may be that teachers and school administrators will
have to become better able to help parents put student performance on external tests
in a broad assessment perspective. Moreover, external tests should not be allowed by
educators or parents to displace careful ongoing assessment by classroom teachers.

Determining System Performance

Just as parents have a right to have informed appraisals of the educational progress
of their children, so, too, does society have a right to have informed appraisals of
how well its schools are performing. For that purpose, national, state, and school
district assessments are appropriate. They must, however, be carefully designed. That
includes covering the explicitly defined learning goals set out by the appropriate
authorities and using valid statistical and sampling techniques. Since the basic pur-
pose of external testing is to find out how schools in some aggregate (national, state,
or district) are doing, and not individual students, there is simply no need to conduct
such assessments in every subject every year, or to test all students, or to have all of
the students who do participate respond to the same set of test items. This seems not
to be well understood by the general public.
Indeed, in addition to the huge and quite unnecessary costs, testing all students
has a severe functional drawback. Testing all students actually reveals less about
institutional performance than testing only some students. Blanket testing can sample
only a small fraction of the subject matter in question and as a practical matter must
be multiple choice and machine scored. By contrast, basing state and national as-
sessments on a stratified random sample of test takers means that the entire sweep of
the targeted matter can be covered, since each student faces only a small portion of

154 National Science Teachers Association

 CHAPTER 10

the test items. And it makes it possible to include some performance testing (which
is prohibitively expensive and difficult to score if taken by all students).

Assessing Curriculum Materials

If classroom or external assessment shows that students are not meeting our learning
expectations, ineffective classroom practices may not be the reason, or at least not
the only reason. The failure may well be due to inadequate curricula. At any one
time, many individuals and institutions (though more at some times than at others,
depending on the national sense of urgency and the level of support from federal
agencies and private foundations) are developing new courses and materials (the
stuff of curricula). Claims made for improved learning resulting from the use of the
products of these efforts should not, however, be taken at face value. They should,
rather, be critically evaluated for their instructional effectiveness.
Since the 1960s most nationally funded curriculum development projects in sci-
ence education have carried out some assessment in the very process of developing
their materials. Feedback from limited field trials enables developers to improve
their products before they are released for general use. Carefully conducted forma-
tive assessment contributes to the credibility of the claims made for new products
and techniques, as well as, usually, to the creation of better products (AAAS 2001a).
Still, such assessment is not entirely sufficient for making informed decisions on the
adoption of those materials. Arms-length assessment in the context of actual use is
also desirable, though difficult to carry out for both technical and practical reasons.
Curriculum assessment is related to but different from assessment for improving
classroom teaching and learning, informing third parties, and judging system effec-
tiveness. The constraints are formidable. Any attempt to evaluate curriculum courses
and materials must be narrowly focused, and yet cannot escape being affected by the
many other variables at play in real-world schools. And of course getting unbiased
test schools and controls is difficult and expensive.
Because of such constraints, few curriculum projects have been able to conduct
rigorous field studies. Indeed, the Project Physics Course developed at Harvard Uni-
versity may be the only reform project of the Sputnik era (or later) whose summative
assessment included selecting participating schools using a stratified random sample
of all high schools in the United States, assigning schools to test and control groups
randomly, and providing similar incentives to the teachers and schools in both the
test and control groups.
The summative evaluation of the Project Physics Course used all of its associ-
ated print and multimedia materials, but the course textbook was the center of atten-
tion (Holton, Rutherford, and Watson 1970). In ordinary circumstances, however,
textbooks and their associated teacher and student handbooks stand in as proxies for
a course and its supplementary materials. Fortunately, there are degrees of materials
assessment that, while short of a completely rigorous assessment based on student

Everyday Assessment in the Science Classroom 155

CHAPTER 10

use, can provide valuable assurance that the claims made for it are reasonable (AAAS
2001b; Kesidou and Roseman 2002).
Regardless of the scale and rigor of the assessment of curricula, what must be
clear from the outset are the intended relationships among learning goals, the cur-
riculum, and assessment (AAAS 2001d). It is a question of priority. Is the intent to
have assessment align with the curriculum (a popular choice), which has been inde-
pendently designed to serve specified learning goals? Or is assessment intended to
align with learning goals and drive the curriculum? Or should curriculum and as-
sessment both derive from goals, rather than from each other? The choice makes a
difference especially in shaping the form and purpose of classroom assessment.

Considering Everyday Assessment in the Science Classroom

As a way of fostering productive discussion of this book, this chapter ends with
questions rather than conclusions. Readers of the book will surely have general ques-
tions such as these: Are the assessment claims in the chapters clearly identified and
explicitly stated? Do the claims made in the separate chapters fit together to present
a coherent view of classroom assessment? Are the validity, feasibility, and policy
consequences of the book’s propositions satisfactorily addressed?
More particularly:
• Is it clear what is being proposed with regard to classroom assessment? Is sci-
ence classroom assessment presented as just a particular context for classroom
assessment or as something rather unique in principle? Is the terminology used
sufficiently precise for conveying unambiguously the meaning of the various
propositions? Is the relationship between the assessment approach being advo-
cated and the science learning goals made clear?
• How persuasive are the claims made in this book? To what extent are the argu-
ments made in support of the various claims anecdotal? To what extent is the
evidence presented quantitative? In each case, was the criterion for success spe-
cifically stated ahead of time, only implied, or left unspecified? If stated, was it
the acquisition by students of specific science knowledge and skills, the acquisi-
tion of general learning skills, interpersonal and social skills, or something else?
Are procedures described in enough detail to enable independent replications?
Are these themselves replications of earlier studies? If so, do the findings sup-
port the earlier studies or throw them into doubt? How implicated in the class-
room events were the authors themselves, and how does that affect the credibility
of the findings?
• How practical are the proposals in this book? Would the approaches described
in this book work with teachers who were not volunteers? Would they work
without the help of university faculty? Without special financial resources be-
yond the standard school operating budget? Does the application of the approaches

156 National Science Teachers Association

 CHAPTER 10

to assessment advocated here require extensive retraining of the current popula-

tion of K–12 teachers? If it is desirable for preservice teacher education to incul-
cate these practices, will it require university schools of education to change
their programs significantly? What incentives would it take for them to do so?
• What policy issues are implicit in the approach to classroom assessment set forth
in this book? If different teachers of the same science course use the approach
described here, can the results be used to compare the aggregate performance of
their respective students? If not, how can school administrators and school boards
decide on its acceptability? Is this approach to science classroom assessment
likely to be at odds with state-mandated assessment? And, finally, do the results
of such classroom assessment provide evidence of how well students, individu-
ally and collectively, are meeting national science content standards?

References
American Association for the Advancement of Science (AAAS). 1993. Benchmarks for science lit-
eracy, 327–77. New York: Oxford University Press.
———. 1997. Resources for science literacy, 43–58. New York: Oxford University Press.
———. 2001a. Designs for science literacy, 129–41. New York: Oxford University Press.
———. 2001b. Designs for science literacy, 195.
———. 2001c. Designs for science literacy, 201–202.
———. 2001d. Designs for science literacy, 203–208.
Bestor, A. E. 1953. Educational wastelands. Urbana: University of Illinois Press.
Cremin, L. A. 1961. The transformation of the school: Progressivism in American education, 1876–
1957. New York: Alfred A. Knopf.
Dow, P. B. 1991. Schoolhouse politics: Lessons from the Sputnik era. Cambridge: Harvard University
Press.
Holton, G., F. J. Rutherford, and F. G. Watson. 1970. Project physics. New York: Holt, Rinehart, and
Winston.
Kesidou, S., and J. Roseman. 2002. How well do middle school science programs measure up? Find-
ings from Project 2061’s curriculum review study. Journal of Research in Science Teaching 39
(6): 522–49.
Lynd, A. 1953. Quackery in the public schools. Boston: Little, Brown.
Randolph, J. L. 2002. Scientists in the classroom: The Cold War reconstruction of American science
education. New York: Palgrave.
Ravitch, D. 1983. The troubled crusade: American education, 1945–1980. New York: Basic Books.
Rickover, H. G. 1959. Education and freedom. New York: Dutton.
Schmidt, W. H., C. C. McKnight, and S. A. Raizen. 1997. A splintered vision: An investigation of U.S.
science and mathematics education. Boston: Kluwer.
U.S. Department of Education. 1983. A nation at risk. Washington, DC: U.S. Government Printing
Office.
———. 1989. The president’s education summit with governors: Joint statement. Washington, DC:
U.S. Government Printing Office.

Everyday Assessment in the Science Classroom 157

 INDEX

Index
Page numbers in boldface type indicate tables or figures.

A purposes of, 1–3, 9–10, 152–156

AAAS (American Association for the Advance- designing assessments for different
ment of Science), 149 purposes, 122–124
Abstract thought, 4 through questioning, 61–72 (See also
Accountability, xi, 1, 2, 91, 93, 121, 147 Questioning)
Acid-base study, 50–51 reliability of, 2, 23, 92, 122, 150
Advanced placement (AP) examinations, 141 reporting progress to parents and others, 89–
“Alternative frameworks” of students, 34 107
American Association for the Advancement of across subjects, 8–9
Science (AAAS), 149 summative, 5, 10, 124, 142
AP (advanced placement) examinations, 141 teachers’ roles in, 76
Assessment validity of, 2–3, 23, 92, 150
“authentic,” 121 Assessment conversation, 54–58, 55
avoiding bias in, 5, 92 “Assessment culture,” 94
broad definition of, 1, 6, 76, 109 Assessment for learning, xi, 2, 4, 6–7, 10, 13–
changing scene for, 147–152 24, 77, 121
demographics, 147–148 elements of, 13–23, 77
educational demands, 148–149 collecting evidence of learning, 20–21
learning studies, 149–150 communicating using evidence of learning,
national crises, 150–152 21–22
context of presentation of, 5 continuing the learning, 23
of curriculum materials, 155–156 providing specific, descriptive feedback,
embedded, 90–91, 118 19
fairness of, 92 revisiting criteria, 17–18
formative, 5, 6–10, 124, 125 self-assessing and goal setting, 16–17
importance of, xi, 1–10 setting and using criteria, 15–16
of inquiry, xi, 51–58 (See also Inquiry) setting goals, 19–20
involving students in, xii, 7–8, 13, 23–24, 75– showing samples, 14–15
86 talking about learning, 14
vs. measurement, xi facilitation of lifelong learning, 82–83
model of effective classroom assessment, redesigning large-scale assessments to
124–125 promote learning, 121–146
opportunities for, xii, 109–110 in science classroom, 23–24
positive influence of, xi Assessment Training Institute, 110–111
principles of, 91–94
developmental perspective, 91–92 B
management by teachers, 93–94 Berkeley Evaluation and Assessment Research
match between instruction and assessment, (BEAR) Assessment System, 90, 94–104, 106
93 application to Living by Chemistry
quality evidence, 92 project, 95–97
for program “diagnosis,” 129–132, 133–134 application to Science Education for Public

159
INDEX

Understanding Project, 104–105, 105 Conceptual domain of inquiry, 62, 46, 51–53
assessment moderation meetings for discus- Conceptual modification, 4
sion of, 103–104 “Constructivist” theories, 3, 54
developmental perspective of, 94–97 Conversation activities for assessment of
progress variables, 94–97, 96 inquiry, 54–58, 55
internal consistency of, 103 Creative problem-solving, 29
interrelationships among components of, 103 Criteria for learning, 15–18
match between instruction and assessment revisiting of, 17–18
in, 99–101 setting and using of, 15–16
assessment tasks, 99–101, 102 Culturally meaningful activities, 31–32
teacher management of, 101–103 Curriculum assessment, 155–156
Assessment Blueprints, 102–103
exemplars, 102 D
scoring guides, 101 Demographic factors, 147–148
technical quality of, 98–99 Determining system performance, 154–155
item response modeling, 98 large-scale assessments for, 129–132, 133–134
progress maps, 98–99, 99, 100 Developmental perspective of student
BGuILE (Biology Guided Inquiry Learning learning, 91–92
Environments), 45
Bias, 5, 92 E
BioLogica, 45 EDA (exploratory data analysis), 53
Biology Guided Inquiry Learning Environments Educational demands, 148–149
(BGuILE), 45 Educational Testing Service’s Pacesetter
Bruner, Jerome, 41 program, 144
EE (evidence-explanation) continuum, 46–49, 47
C Elicitation questions, 62–63, 63, 68–70
Capacity building, 145 Embedded assessment, 90–91, 118, 135–136,
CAPITAL. See Classroom Assessment Project 136–140
to Improve Teaching and Learning English-as-second-language students, 39
Center for Highly Interactive Computing in Epistemic domain of inquiry, 42, 46, 51–53
Education (Hi-CE), 45 Evidence-explanation (EE) continuum, 46–49, 47
Certification assessment, 1 Evidence of learning
Classroom Assessment Project to Improve collection of, 20–21
Teaching and Learning (CAPITAL), 113–118 communicating using, 21–22
collaborative personal approach of, 115–117 teacher evaluation of quality of, 22
Classroom assessments compared with large- Examining students’ work, 27–39
scale assessments, 122–123 case-study approach for, 39
Classroom learning environments, 10, 42–46 by creative problem-solving, 28–29
Colloquiums in science classrooms, 54 cultural context for, 31–32
Communication importance of listening for, 29–31, 46
of evidence of learning, 21–22 instruction and, 38
importance of listening, 29–31, 46 modifying teaching objectives based on, 27–28
inquiry and, 47 professional development programs for, 38–39
for peer-assessment, 75, 78–81 by recognizing difficulty of changing their
speaking together activities for assessment of theories of the world, 34–37
inquiry, 54–58, 55 role of teachers in, 39
Concept maps, 52 to see what they are thinking, 28

160 National Science Teachers Association

 INDEX

for students with English as second conceptual, epistemic, and social domains of,
language, 39 42, 46, 51–52
by studying groups, 32–34 creating learning environment for, 43–46
wide variety of methods for, 37–38 curriculum material for, 45, 48
Experiential learning, 4 definition of, 41
Exploratory data analysis (EDA), 53 designing learning environment for, 43–46
essential features of, 46–47, 47, 52
F evidence-explanation continuum as frame-
Feedback, 7–8, 13, 109 work for, 46–49, 47
evaluative, 19 instructional strategies for promotion of, 49–50
by peers, 4–5, 7, 75, 78–81 limiting number of core concepts for, 48
via questioning, 61–72 NSF-supported full-inquiry science units, 45,
specific and descriptive, 19 46
talking about the learning, 14, 17 Item response modeling (IRT), 98
Forces lessons, 62–68, 63, 68–71
Formative assessment, 5, 6–10, 124, 125 K
Fullan, Michael, 17 Kit-based science investigations, 43, 49
Knowing What Students Know, 123, 135, 143,
G 145
GEMS (Great Explorations in Mathematics and
Science), 37–38 L
Goal setting, 16–17, 19–20 Laboratory activities, student discussions of,
Grades and grading practices, 7–8, 89–90, 125 29–31
Grandy, Richard, 54 Large-scale assessments, 121–146
Gravity lessons, 35–37, 36, 56, 57 compared with classroom assessments, 122–
Great Explorations in Mathematics and Science 123
(GEMS), 37–38 embodiment of curriculum standards, 123–124
Green crab lessons, 14–23 impediments to and recommendations for
promoting student learning, 143–146
H once-yearly administration of, 122
Hi-CE (Center for Highly Interactive Comput- out-of-level testing with, 135
ing in Education), 45 purposes of, 121–122, 126–143
certification or “screening” of individual
I student achievement, 132–141, 136–140
Inquiry, xi, 41–58 exemplification of learning goals, 126–
abilities of young children, 44 129, 127–128
assessment of, 41, 51–58 professional development, 141–143
complexity of, 41, 58 program “diagnosis,” 129–132, 133–134
conceptual domain, 51–53 reliability of, 122
epistemic domain, 52, 53 LBC. See Living by Chemistry project
keys to success in, 58 “Learned helplessness,” 8
social domain, 52 Learning
speaking together activities for, 54–56, 55 assessment for, xi, 2, 4, 6–7, 10, 13–24, 77, 121
storyboards for, 56–58, 57 collecting evidence of, 20–21
changing emphases for promotion of, 43, 44 delivery vs. interactive modes of, 9
communication process and, 47 developmental perspective of, 91–92
experiential, 4

Everyday Assessment in the Science Classroom 161

INDEX

feedback for, 7–8, 13, 19 Classroom Assessment Project to Improve

through inquiry, 41–42 Teaching and Learning, 113–118
large-scale assessments to exemplify goals full-inquiry science units, 45, 46
for, 126–129, 127–128 Living by Chemistry, 95
managing opportunities for, 109 New Standards Project, 144
relevance of, 14 No Child Left Behind Act, 129, 132, 144
research studies on, 149–150 NSF. See National Science Foundation
setting and using criteria for, 15–16
showing samples of student work that O
illustrate, 14–15 One Sky, Many Voices, 45
socially mediated, 5, 14
talking about, 14, 17 P
teachers as mediators of, 5, 7, 24, 42, 53 Parents, communicating evidence of students’
theories of, 3–5 learning to, 21–22, 89–107, 153–154
Learning by Design, 45 Pedagogical content knowledge, 8–9
Learning environments, 10, 42–46 Peer-assessment, 4–5, 7, 75, 78–81
Learning Technologies in Urban Schools Phylogenetic tree activity, 136–140
(LeTUS), 45 Piaget, Jean, 4
Listening to students’ interpretations, 29–31, PISA (Program for International Student
46, 72 Assessment), 1
Living by Chemistry (LBC) project, 95 Practical theories of teachers, 112–113
exemplars for, 101 Principles of assessment, 91–94
progress maps for, 98, 99, 100 in BEAR Assessment System, 90, 94–104
progress variables for, 95–97, 96 (See also Berkeley Evaluation and
science content standards of, 106–107 Assessment Research Assessment System)
scoring guide for, 101 developmental perspective, 91–92
management by teachers, 93–94
M match between instruction and assessment,
Memory, 4 93
Metacognition, 5, 17 quality evidence, 92
Metz, K., 44 Professional development, assessment-related,
Modeling for Understanding in Science 38–39, 109–119
Education (MUSE), 45 approaches to working with teachers, 110–112
Motivation of students, 7–8, 23, 125 a personal approach, 112–113
MUSE (Modeling for Understanding in Science Classroom Assessment Project to Improve
Education), 45 Teaching and Learning, 113–118
collaborative personal approach of, 115–117
N effect of teachers’ experiences and prefer-
National Assessment of Education Progress ences on change process, 111–112
(NAEP), 1, 126, 131, 132 large-scale assessments as tool for, 141–143
National Council of Teachers of Mathematics, 111 principles for design of, 117–118
National crises and science education reform, in Science Education for Public Understand-
150–152 ing Project, 104
National Science Education Standards, xi, xiv, Program for International Student Assessment
41, 43, 44, 48, 61, 78, 123–125, 128 (PISA), 1
National Science Foundation (NSF)–supported Project 2061, 149
projects Project Physics Course, 151–152, 155

162 National Science Teachers Association

 INDEX

Project SEPIA (Science Education through “Scaffolding,” 5, 9

Portfolio Instruction and Assessment), 45, 48, Science Education for Public Understanding
50, 55–56 Project (SEPUP), 104
Purposes of assessment, 1–3, 9–10, 152–156 BEAR Assessment System applied to, 104–
assessing curriculum materials, 155–156 105, 105
designing assessments for different purposes, “Issues, Evidence, and You” curriculum of, 104
122–124 professional development activities of, 104
determining system performance, 154–155 Science Education through Portfolio Instruction
improving classroom teaching and learning, and Assessment (Project SEPIA), 45, 48, 50,
153 55–56
informing parents and others, 153–154 Science—A Process Approach (SAPA), 44
large-scale assessments, 121–122, 126–143 Scientific thinking and reasoning, 41–42
certification or “screening” of individual Self-assessment by students, 4–5, 7, 16–17, 19–
student achievement, 132–141, 136–140 20, 77, 78
exemplification of learning goals, 126– SEPUP. See Science Education for Public
129, 127–128 Understanding Project
professional development, 141–143 Social domain of inquiry, 42, 46, 52
program “diagnosis,” 129–132, 133–134 Socially mediated learning, 5, 14
Speaking together activities for assessment of
Q inquiry, 54–58, 55, 57
Questioning, 61–72 Standardized tests, 29, 91–93. See also Large-
to assess conceptual frameworks, 52–53 scale assessments
elicitation questions, 62–63, 63, 68–70 Standards for Educational and Psychological
episodes of, 64–68 Testing, 122, 134
establishing norms of scientific argumenta- Stiggins, Richard, 110
tion, 61 Storyboards, 56–58, 57
example of ongoing assessment through, 62– Student involvement in assessment process, 7–
68 8, 13, 23–24, 75–86
purposes and kinds of questions, 68–70 benefits of, 78
using questions and answers to inform developing shared understanding of quality
instruction, 70–71 work, 79–81
facilitating lifelong learning, 82–83
R improving performance, 77–78
Re: Learning by Design, 111 understanding relationships to see bigger
Reform in science education, 9 picture, 81–82
Reliability of assessments, 2, 23, 92, 122, 150 creating meaningful opportunities for, 84–85
Reporting students’ progress, 89–107, 153–154 goal setting, 85
assessment principles for, 91–94 public displays of work, 84–85
BEAR approach to, 90, 94–106 (See also reflection, 85
Berkeley Evaluation and Assessment revision, 85
Research System) time, 84
grades and grading practices, 7–8, 89–90 traditional assessment, 84
Rubrics, 18, 101, 141–142 formats for, 77
learning skills for, 85–86
S peer-assessment, 4–5, 7, 75, 78–81
Samples of student work, 14–15 relationship among students, content, and
SAPA (Science—A Process Approach), 44 assessment, 83–84

Everyday Assessment in the Science Classroom 163

INDEX

research support for, 77 learning to teach, 27–28

self-assessment, 4–5, 7, 16–17, 19–20, 77, 78 management of classroom assessment
Students systems by, 93–94
achieving consensus of views of, 56–58 as mediators of learning, 5, 7, 24, 42, 53
“alternative frameworks” of, 34 practical theories of, 112–113
beginning teaching at level of, 5, 27–28 using questions and answers to inform
communicating evidence of learning by, 21–22 instruction, 70–71
culturally meaningful activities for, 31–32 Tests, 20. See also Assessment
diversity of, 147–148 designing for different purposes, 122–124
encouraging creative problem-solving by, 29 formative use of, 7
with English as second language, 39 large-scale, 121–146
examining work of, 27–39 reliability of, 2, 23, 92, 122, 150
large-scale assessments for certification or standardized, 29, 91–93
“screening” of individual student achieve- teacher-made, 92, 93
ment, 132–141, 136–140 “teaching to the test,” 3, 93, 121
learning studies of misconceptions held validity of, 2–3, 23, 92, 150
by, 149–150 Third International Mathematics and Science
listening to interpretations of, 29–31, 46, 72 Study (TIMSS), 1, 126, 131
motivation of, 7–8, 23, 125 Triangulation, 21, 24
recognizing diversity of ideas of, 56
reporting progress of, 89–107 V
Summary feedback, 19 Validity of assessments, 2–3, 23, 92, 150
Summative assessment, 5, 10, 124, 142 Vygotsky, L. S., 5, 54

T W
Teachers Web-based Inquiry Science Environment
assessment-related professional development (WISE), 45
of, 109–119 Wiggins, Grant, 111
assessment roles of, 76 WISE (Web-based Inquiry Science Environ-
directiveness of, 28–29 ment), 45
examination of students’ work by, 27–39
flexibility of, 6, 9 Z
importance of listening by, 29–31, 46 “Zone of proximal development,” 5

164 National Science Teachers Association