1

section

Assessment and
Learning
■ Steven Katz

T he relationship between classroom assessment and learning is undoubtedly complicated.

Fortunately, ongoing work has served to disentangle some of the complexities. We know, for
example, that classroom assessment is imbued with a multi-purpose character. At times, it must
certify proficiency of students; at other times, it must direct teachers in the next steps of their
instructional sequences; and, at still other times, it must cultivate the capacity within students to
take charge of their own learning.
In Rethinking Classroom Assessment with Purpose in Mind (2006), Lorna Earl and I outline
these three distinct purposes of classroom assessment as:
1. assessment of learning
2. assessment for learning, and
3. assessment as learning

ASSESSMENT AND LEARNING:

Why, What, and How?

■ Dylan Wiliam

DYLAN WILIAM sees that teacher quality is the wealth, and a longer and healthier life (OECD, 2001 p. 33).
single most important variable in student progress For societies, increasing levels of educational achievement
result in lower criminal justice costs, lower costs of state-
and he highlights assessment for learning as the provided health-care (Schweinhart et al., 2005), and increased
practice whereby teachers use evidence about economic growth. Quantifying the amount of extra economic
student achievement to adapt instruction to better growth inevitably involves some pretty heroic assumptions.
meet learning needs. Nevertheless, Hanushek (2004) has calculated that if we could
increase student achievement by one standard deviation—
equivalent to about 15 percentage points on a typical test where

E ducation matters. It matters for individuals and it matters for

societies. For the individual, greater levels of education are
associated with greater control over one’s life, greater economic
the scores range from around 20 per cent to around 80 per
cent—then in 30 years, the increased growth in the economy
would more than pay for the entire cost of K–12 education.

o r b i t , Vol 36, No 2, 2006 C a l l 4 1 6 9 2 3 - 6 6 4 1 , e x t . 2 0 7 7 t o o r d e r O r b i t o r v i s i t w w w. o r b i t m a g a z i n e . c a

2
We also note the importance and validity of each both the learning of self-regulation
in the education enterprise and argue for the need as well as in the self-regulation of
to be able to differentiate among the purposes of learning. Laveault argues that a
classroom assessment in the service of high- better understanding of the role of
leverage, intentional practice. Indeed, even the regulation can contribute to improved
Assessment Reform Group in the U.K., the contextualization and differentiation
collaborative synonymous with promoting the in formative assessment practices
assessment for learning agenda, has recently by, for example, taking students’
worked on the differentiation challenge by developmental levels into account.
explicitly considering the role of teachers in ■ Carmel Crévola, Peter Hill, and Michael
assessment of learning. Fullan extend this latter point by noting
Each of the articles in Section 1 considers the the place of assessment for learning within
relationship between assessment and learning: Critical Learning Instructional Paths
■ Dylan Wiliam sees that teacher quality is the (CLIPs). An essential part of developing a
single most important variable in student critical learning path, they explain, is
progress and he highlights assessment for being able to make a direct link
learning as the practice whereby teachers use between a student’s results on a
evidence about student achievement to adapt set of assessments, their stage of
instruction to better meet learning needs. development, and specific teaching
■ Dany Laveault connects the psychological strategies relevant to that stage of
notion of “regulation” to formative assessment. development.
He explains the role that assessment can play in

Where’s the Solution? teacher was the most significant variable. If you get one of the
best teachers, you will learn in six months what an average
Around the world, many approaches to school improvement have teacher will teach you in a year, and if you get one of the worst
been proposed, but most have been relatively ineffective because teachers, the same amount of progress will take you two years—
they have failed to embrace a lasting and inevitable truth: the there is a fourfold difference between the productivity of the
single most important variable in the amount of progress that a best and worst teachers (Hanushek, 2004; Wiliam & Thompson,
student makes at school is the quality of the teacher. in press).
This has not been obvious because for many years we lacked Improving education, is therefore, in effect, a labour-force
the right kinds of data-sets. There is a huge variability in the issue with two solutions. One is to replace existing teachers with
levels of achievement of students in different schools, and better ones. The problem with this is that there is no evidence
simplistic analyses suggested that these differences must be that there are better teachers queuing up to get in, but can’t get
attributable to the schools. However, it rapidly became clear that jobs (Darling-Hammond et al., 2005). Nor is there that much
one of the major reasons for these differences in the “outputs” evidence that better pay would have a substantial impact on
of schools was that the schools served very different teacher quality (there are many good reasons for paying teachers
communities. Analyses that took into account the socio-economic more, but the supply-and-demand argument is not one of them).
status (SES) of students in different schools painted a very The other solution is to invest in the teachers that we have—
different picture—it was SES, rather than school quality that was what my colleague Marnie Thompson calls the “love the one
the major determinant of educational achievement. Still others you’re with” strategy.
attributed the differences to the students themselves; their prior Twenty years ago, this would have resulted in a very gloomy
achievement for example. prognosis, since there was little evidence that it was possible to
However, as more and more data-sets began to consider the improve the quality of a teacher’s practice, except very slowly,
“value added” by the school—the difference between what the and at great cost (Fullan, 1991). Many people therefore concluded
students knew when they started at that school and what they that the situation was hopeless, and that policy makers should,
knew when they left—it became clear that the quality of the instead, focus on solutions that did not require quality teaching

Ma k in g Cl a s s r o o m a s s e s s me n t w o r k f o r s t u d e n t l e a r n in g
3
(such as producing “teacher-proof” textbooks, so that it didn’t
matter how good the teacher was).
Building Capacity for
However, the problem with this argument is that we can Assessment for Learning
conclude very little from the failure of years of teacher
professional development to improve teaching, because for the KEY PARAMETERS
vast majority of teachers, the kind of PD they received flew in the 1. Accountability
face of what the research says about what makes for effective PD.
Write out an action plan; give monthly reports
That research says that PD needs to be continuing, rather than in
the form of “one-shot” inputs (Cohen & Hill, 1998); it needs to be on what was tried, and the results.
tailored to the local circumstances in which teachers work (Cobb 2. Support
et al., 2003); it needs to be directly related to the content of the Agree on an agenda for peer observation; what
subjects they teach (Supovitz, 2001); and it needs to involve
teachers in active and collective participation (Garet, Birman,
should count as evidence with reference to the
Porter, Desimone, & Herman, 1999). action plan.
So much for the “process” of teacher PD. What about the 3. Choice
“content” of teacher PD? Does it matter what teachers focus on, Use techniques within a framework of
as long as they are engaged in the processes listed above? It turns
out that it does. It turns out that teachers can spend a lot of time accountability,
developing their thinking, without changing their practice very always being able to answer “What’s formative
much, and from the research it appears that one particular about that?”
focus—the use of assessment for learning—has a bigger impact
4. Flexibility
on student achievement than any other.
Modify techniques as needed, keeping the
Assessment for Learning research evidence in mind.
5. Gradualism
For over 20 years, Paul Black and I have been researching the role
that assessment can play in raising levels of student achievement.
Proceed slowly to effect faster change; implement
In particular, we wanted to find out if using assessment to no more than three changes at one time into
support learning, rather than just to measure its results, could your practice.
improve students’ achievement, even when such achievement is
measured in the form of state-mandated tests. In reviewing 250
studies from around the world, published between 1987 and
1998, we found that a focus by teachers on assessment for There is ample evidence that focusing on any one of these five
learning, as opposed to assessment of learning, produced strategies is highly effective in raising student achievement
substantial increases in students’ achievement—typically (Wiliam, in press). Together they provide the highest leverage
doubling the rate of learning (Black & Wiliam, 1998). Since the focus that we know for improving schools, classrooms, and the
studies also revealed that day-to-day classroom assessment was achievement of students. The problem is that knowing what
relatively rare, we felt that considerable improvements would needs to be done is one thing. Doing it is quite another.
result from supporting teachers in developing this aspect of their
practice. Putting It into Practice
One of the things that we have struggled with is what, exactly,
is assessment for learning? Many people have proposed We know that “one size fits all” does not work in teacher
definitions, all of which differ slightly, but it seems that the professional development. What might be exactly the right thing
“big idea” is that evidence about student achievement is used to do in one situation might be exactly the wrong thing to do in a
to adapt instruction to better meet learning needs (see Wiliam & similar, but different situation. Expert teachers do not use
Thompson, in press). The key strategies involved in this are: general, all-purpose approaches to solve problems, but rather
■ clarifying and sharing learning intentions and success criteria generate solutions that take advantage of specific details in the
with learners; challenges they face (Berliner, 1994). That is why “what works”
■ engineering effective classroom discussions, activities, is not the right question in education; everything works
and tasks that elicit evidence of student achievement; somewhere, and nothing works everywhere. The right question is
■ providing feedback that moves learners on; “under what conditions does this work?” That is why if we are
■ activating students as owners of their own learning; and serious about developing assessment for learning, we must help
■ activating students as instructional resources for one each teacher find her or his own way of doing this.
another. The opposite of one size fits all is to allow each teacher to

o r b i t , Vol 36, No 2, 2006 C a l l 4 1 6 9 2 3 - 6 6 4 1 , e x t . 2 0 7 7 t o o r d e r O r b i t o r v i s i t w w w. o r b i t m a g a z i n e . c a

4
 1
section

ASSESSMENT AND LEARNING

choose what they want to do. This may be attractive, but it is not what counts as evidence, the observer’s own prejudices are
rigorous. Teachers may choose to change things that have no minimized, and the difference between this and supervisory
impact on student achievement. The trick is, therefore, to observation is emphasized.
generate teacher professional development that is open enough
to allow each teacher to adapt the new ideas into their own 3. CHOICE
practice, but structured enough so that the adaptation does not Teachers often describe the process of changing their practice
produce a “lethal mutation” (Pellegrino, 2006) that renders the as “scary.” Implementing assessment for learning in particular
innovation impotent. makes many teachers feel as if they are being asked to “give up
Over the last three years, I and my colleagues at the Learning control” of their classrooms. However, teachers responsible for
and Teaching Research Center at the Educational Testing Service choosing what they will change about their practice feel
in Princeton, NJ, have been working with many groups of empowered, especially when they can choose among techniques
teachers on the best ways to implement assessment for learning. those that appeal to them. This choice lies, however, within a
Some approaches have been extremely successful, and others framework of accountability. While teachers are free to choose
less so, but even from our failures, we have usually been able to what they change, they are accountable for changing something.
learn something (and often more from our own fortunes than They are also accountable for showing their peers how their
from our successes). As a result of reflection on both our innovations are consistent with the principles of assessment for
successes and failures, we have been able to identify five learning. One especially powerful way to support this is for peers
elements that increase the successful implementation of to ask each teacher, “What’s formative about that?” in order to
assessment for learning: accountability, support, choice, emphasize the need for the changes in practice to be focused
flexibility, and gradualism. on using assessment evidence to adapt instruction to meet
student needs.
1. ACCOUNTABILITY
Most professionals involved in teacher development will have 4. FLEXIBILITY
had the experience of generating considerable enthusiasm for, A technique that works for one teacher may not work for another,
and commitment to, change during a summer workshop, only to but may do so after some modification. Teachers need to be
find that all their good intentions seem to be erased by the encouraged to modify techniques to make them work in their
demands of the new school year. That is why we suggest that classrooms, but they also need to understand enough of the
teachers should make a commitment in writing about what they research evidence so that the changes they make do not render
are going to change about their practice (the action plan) and the innovation ineffective.
then to be held accountable by colleagues at monthly meetings
for making those changes in their practice. Each month, every 5. GRADUALISM
teacher describes what he or she tried and how it went. Teachers Asking teachers to make wholesale changes in their practice
have told us that having to face their colleagues and “deliver” on is a little like asking a golfer to change her swing during a
the promises they made the previous month helps them move tournament. Teachers have to maintain the fluency of their
their “change” task to the top of their in-tray. classroom routines, while at the same time disrupting them. The
action plans that teachers develop should specify a small
2. SUPPORT number of changes—ideally two or three—that they will make
The other side of the coin of accountability is support. We have in their teaching. In our experience, teachers who try to change
found that at the monthly meetings teachers are able to offer more than three things in their practice at the same time are
each other advice when the planned changes are not going well, never successful. They try to do too much, and their classroom
and the fact that these meetings are with groups of peers, rather descends into chaos, and as a result they revert to what they
than an expert teaching a novice, appears to be particularly know how to do. In other words, going slower produces faster
beneficial. It must be borne in mind that ,ultimately, real change.
implementing assessment for learning involves changes in day-
to-day, and even minute-to-minute classroom practice. That is Conclusion
why we also recommend that teachers engage in peer
observation. To clearly distinguish these observations from those Our experience is that, equipped with some basic ideas about
routinely carried out to manage performance, these observations assessment for learning, teachers can support each other in
should be done by genuine peers rather than those in a making radical improvements in their students’ learning,
hierarchical relationship. Another important requirement is that provided each teacher is responsible for her or his own
the teacher being observed must set the agenda for the development targets. It is perfectly OK to hold teachers
observation and spell out for the observer what should count as accountable for making changes in their practice, provided that:
evidence, by reference to her or his action plan. By defining the ■ the teachers have a choice in how they put the basic principles

observer’s role, both in terms of what is to be looked for and of assessment for learning into effect;

Ma k in g Cl a s s r o o m a s s e s s me n t w o r k f o r s t u d e n t l e a r n in g
5
■ they are not pushed to make changes faster than they can
incorporate them into their normal practice; and
■ they have the support and accountability that a group of like-
minded group of peers can provide.

References
Berliner, D.C. (1994). Expertise: The wonder of exemplary performances. In Classroom resource
J.N. Mangieri & C.C. Block (Eds.), Creating powerful thinking in teachers
Teaching Materials
and students: Diverse perspectives (pp. 161–186). Fort Worth, TX: Harcourt
Brace College. Integrated DVD and CD-ROM
Black, P. J., & Wiliam, D. (1998). Assessment and classroom learning.
Assessment in Education: Principles Policy and Practice, 5(1), 7–73. Lesson Plans
Cobb, P., McClain, K., Lamberg, T.d.S., & Dean, C. (2003). Situating teachers’ Evaluation Rubrics
instructional practices in the institutional setting of the school and
district. Educational Researcher, 32(6), 13–24.
Cohen, D.K., & Hill, H.C. (1998). State policy and classroom performance:
Mathematics reform in California. Philadelphia, PA: University of The Ontario Secondary School
Pennsylvania Consortium for Policy Research in Education. Teachers’ Federation has created
Darling-Hammond, L., Holtzman, D.J., Gatlin, S.J., & Vasquez Heilig, J. (2005). high quality, classroom-ready
Does teacher preparation matter? Evidence about teacher certification,
resources on combatting HIV/AIDS
teach for America, and teacher effectiveness. Education Policy Analysis
Archives, 13(42). in South Africa and Canada. It is
Fullan, M. (1991). The new meaning of educational change. London: Cassell. multimedia in nature, engages
Garet, M.S., Birman, B.F., Porter, A.C., Desimone, L., & Herman, R. (1999). students and instructors alike and
Designing effective professional development: Lessons from the is designed for a variety of subjects,
Eisenhower Program. Washington, DC: US Department of Education.
including: civics, science, health
Hanushek, E.A. (2004). Some simple analytics of school quality (NBER
working paper no. W10229). Washington, DC: National Bureau of and social sciences.
Economic Research.
Pellegrino, J.W. (2006, Mar 17). Personal communication.
Schweinhart, L.J., Montie, J., Xiang, Z., Barnett, W.S., Belfield, C.R., & Nores, M. CONTENTS:
(2005). Lifetime effects: The High/Scope Perry preschool study through age
40. Ypsilanti, MI: High/Scope Educational Research Foundation.
program DVD, 200+ pages of hands-
Supovitz, J.A. (2001). Translating teaching practice into improved student on lesson modules, supporting print
achievement. In S.H. Fuhrman (Ed.), From the capitol to the classroom: materials and resources, CD-ROM
Standards-based reform in the States (Vol. Part 2, pp. 81–98). Chicago, IL: and interactive website access.
University of Chicago Press.
Wiliam, D. (in press). Keeping learning on track: formative assessment and
the regulation of learning. In F.K. Lester, Jr. (Ed.), Second handbook of
mathematics teaching and learning. Greenwich, CT: Information Age PRICE:
Publishing. $50 — Complete package.
Wiliam, D., & Thompson, M. (in press). Integrating assessment with $30 — Program DVD and CD-ROM.
instruction: What will it take to make it work? In C.A. Dwyer (Ed.), The
future of assessment: Shaping teaching and learning. Mahwah, NJ:
Lawrence Erlbaum Associates.
ORDER:
Online at www.osstf.on.ca and
select Common Threads or
DYLAN WILIAM is Deputy Director of the e-mail bellisd@osstf.on.ca.
Institute of Education in London. After Telephone: 416.751.8300 or
teaching in urban schools in London for 1.800.267.7867
seven years, he spent 19 years at King’s
College, London, where he was Dean of the
Canadian Agence
School of Education, and Assistant Principal International
Development
Agency
canadienne de
développement
international

of the College. From 2003–06, he directed the Learning and

Teaching Research Center at the Educational Testing Service
in Princeton, NJ.

o r b i t , Vol 36, No 2, 2006 C a l l 4 1 6 9 2 3 - 6 6 4 1 , e x t . 2 0 7 7 t o o r d e r O r b i t o r v i s i t w w w. o r b i t m a g a z i n e . c a

6