Dylan

Wiliam
Wiliam | Assessment: The
Bridge
between Teaching and Learning
page

Assessment: The Bridge between

Teaching and Learning

ur students do not learn what

we teach. It is this simple and
profound reality that means
that assessment is perhaps the central
process in effective instruction. If our
students learned what we taught, we
would never need to assess. We could
simply catalog all the learning experiences
we had organized for them, certain in
the knowledge that this is what they had
learned.
But of course, anyone who has spent more than a
few hours in a classroom knows this hardly ever
happens. No matter how carefully we design and
implement the instruction, what our students
learn cannot be predicted with any certainty. It
is only through assessment that we can discover
whether the instructional activities in which we
engaged our students resulted in the intended
learning. Assessment really is the bridge between
teaching and learning.

Formative Assessment
Of course, the idea that assessment can help
learning is not new, but what is new is a growing
body of evidence that suggests that attention to
what is sometimes called formative assessment, or
assessment for learning, is one of the most powerful ways of improving student achievement.
Different people have different views about what
exactly counts as formative assessment. Some
think it should be applied only to the minute-tominute and day-to-day interactions between stu-

15

dents and teachers, while others also see interim,

or benchmark, tests administered every six to ten
weeks as formative. For my part, I believe that
any assessment can, potentially, be formative,
which is why I suggest that to describe an assessment as formative is what Gilbert Ryle (1949) described as a category mistake (p. 16; ascribing
to something a property it cannot have).
The term formative should apply not to the
assessment but to the function that the evidence
generated by the assessment actually serves. For
example, a seventh-grade teacher had given her
students an English language arts test, under test
conditions, and collected the students test responses. Most teachers would then try to grade
the students responses, add helpful feedback,
and return the graded papers to the students the
following day. On this occasion, however, the
teacher did not grade the papers. She quickly
read through them and decided that the following day each student would receive back her or
his own paper; in addition, groups of four students would be formed, and each group would
be given one blank response sheet, so that they
could, as a group, produce the best composite paper. When the groups had done this, the teacher
led a plenary discussion in which groups reported
back their agreed responses. What is interesting
about the example is that the assessment being
used had been designed entirely for summative
purposes, but the teacher had found a way of using it formatively.
If we accept that any assessment can be used
formatively, we need some way of defining formative assessment in a way that is useful for classroom practice. The way that I have found most
useful is to think of three key processes in learning:

Copyright 2013 by the National Council of Teachers of English. All rights reserved.
Voices from the Middle, Volume 21 Number 2, December 2013

e15-20-Dec13-VM.indd 15

10/31/13 5:13 PM

Wiliam | Assessment: The Bridge between Teaching and Learning

page

16

1. Where the learner is right now

2. Where the learner needs to be
3. How to get there
Its also vital to consider the respective roles
of teachers, students, and their peers. Regarding
the processes and the roles as independent (so
that teachers, students, and peers have a role in
each) suggests that formative assessment can be
thought of as comprising five key strategies as
shown in Figure 1.1 Each of the five strategies is
discussed in further detail in the sections that follow.

Key Strategies of Formative

Assessment
Learning Intentions
The idea that teachers should share with their
students what it is intended that they learn from
a given instructional activity seems obvious, but
it is only within the past 20 years or so that this
has been routine in English language arts classrooms. While this is a welcome development, it
is also important to note that in many schools,
well-intentioned attempts to communicate learning intentions to students have made writing a
mechanistic process of checklist management.
It is true that rubrics can identify important elements of progression in writing, but they can too
easily become a straitjacket. To be sure, where
we can, with fidelity, specify what makes writing
Where the
Where the learner is How to get there
learner is going right now

Teacher

Peer

Clarifying,
sharing, and
understanding
learning intentions

Student

Engineering effective
Feedback that
discussions, activities,
moves learning
and tasks that elicit
forward
evidence of learning
Activating students as learning resources
for one another
Activating students as owners of their
own learning

Figure 1. The five key strategies of formative assessment

good, we should do so, but we should also remember Albert Einsteins advice: Make things
as simple as possible, but no simpler. Sometimes, we should accept that the best we can do
is help our students develop what Guy Claxton
has called a nose for quality (1995, p. 339). Indeed, some writers, such as Royce Sadler (1989),
have argued that this is an essential precondition
for learning:
The indispensable conditions for improvement are
that the student comes to hold a concept of quality
roughly similar to that held by the teacher, is able
to monitor continuously the quality of what is being
produced during the act of production itself, and has
a repertoire of alternative moves or strategies from
which to draw at any given point. (p. 121)

In recent years, it has been common for writers to advocate the co-construction of rubrics
with students. The idea is that rather than having
the teacher present the students with a rubric as
tablets of stone, the rubric is developed with the
students. A common method for doing this is for
the teacher to provide the students with a number of samples of work of varying quality (e.g.,
anonymous samples from a previous years class);
the students then rank them and begin to identify features that distinguish the stronger work
from the weaker work. This can be a very powerful process, but the end point of such a process
must reflect the teachers concept of quality. The
teacher knows what quality writing looks like;
students generally do not. Of course, the teachers views of quality work may shift in discussion
with a class, but the teacher is already immersed
in the discipline of English language arts and the
students are just beginning that journey.

Eliciting Evidence
Although feedback is considered by many to be
the heart of formative assessment, it turns out that
the quality of the feedback hinges on the quality of evidence that is elicited in the first place.
Knowing that a student has scored only 30% on
a test says nothing about that students learning
needs, other than that he or she has apparently
failed to learn most of what was expected. The

Voices from the Middle, Volume 21 Number 2, December 2013

e15-20-Dec13-VM.indd 16

10/31/13 5:13 PM

Wiliam | Assessment: The Bridge between Teaching and Learning

page

point is, effective feedback requires asking the

right questions. This may be obvious, but what is
less obvious is that effective feedback requires a
plan of action about what to do with the evidence
before it is collected.
Many schools and districts espouse a commitment to data-driven decision making, but too
often, this entails the collection of large bodies of
data just in case they come in useful at some later
point. This is a particular problem when teachers
administer common formative assessments to all
students in a grade and then meet to discuss what
to do. By the time all the common assessments
have been graded and a meeting to discuss the
implications has been scheduled, the data are well
past their sell-by date; the teaching has moved
on. And even if the data were available in a timely
fashion, unless time has already been scheduled
for any additional instruction shown to be necessary by the assessments, nothing useful can
happen. That is why data-driven decision making is not a particularly helpful approach. What
is needed instead is a commitment to decisiondriven data collection. For example, rather than
an end of unit test, the teacher could schedule a three-fourths of the way through the unit
test. Rather than grading the papers, the teacher
could use the information gleaned from the test
to decide which aspects of the unit need to be
re-taught or, if the students have all done well,
provide some extension material.

17
On an even shorter time-scale, a fifth-grade
teacher had been introducing students to five
kinds of figurative language: alliteration, hyperbole, onomatopoeia,
personification, and By the time all the common
simile. Five minutes
assessments have been graded
before the end of the
lesson, she listed the and a meeting to discuss
five kinds of figurathe implications has been
tive language on the
whiteboard. She then scheduled, the data are well
read out a series of
past their sell-by date; the
sentences, asking the
students to use fin- teaching has moved on.
ger voting to indicate what kinds of figurative language they had
heard (e.g., hold up one finger if you hear alliteration, five fingers if you hear a simile, and so on).
These are the sentences she read out:
A. He was like a bull in a china shop.
B. This backpack weighs a ton.
C. The sweetly smiling sunshine warmed
the grass.
D. He honked his horn at the cyclist.
E. He was as tall as a house.
Most of the students responded correctly to
the first two, but most of them chose to hold up
either one finger or four fingers for the third.
The teacher pointed out to the class that a few

connections from readwritethink

Knowledge into Action: Figurative Language
One way to assess students understanding is to put their knowledge into action. The article shares an example using
the study of figurative language. The ReadWriteThink.org lesson plan Figurative Language Awards Ceremony invites
students to explore books rich in figurative language and nominate their favorite examples of similes, metaphors, and
personification for a figurative language award. Once nominations are in, the class votes, selecting a winning example
in each category. Finally, students are challenged to write an acceptance speech for one of the winners, using as many
literary devices (simile, metaphor, personification) as they can in their speech.
http://www.readwritethink.org/classroom-resources/lesson-plans/figurative-language-awards-ceremony-115.html
Lisa Fink
www.readwritethink.org

Voices from the Middle, Volume 21 Number 2, December 2013

e15-20-Dec13-VM.indd 17

10/31/13 5:13 PM

Wiliam | Assessment: The Bridge between Teaching and Learning

page

18

students had held up one finger on one hand and

four on the other, because the sentence was an
example of both alliteration and personification,
while most students had assumed that a sentence
could only have one kind of figurative language.
With this misconception cleared up, most of the
students realized that the fourth statement was
both alliteration and onomatopoeia while the last
was both a simile and hyperbole.
The important point about this example is
that the teacher planned to collect the data only
once she had decided how she was going to use
itin this case managing to administer, grade,
and take follow-up reFirst, knowing the students medial action in less
than three minutes.

allows the teacher to make

better judgments about

Feedback

In 1996, two researchers in the psychology

and when to back off. department at Rutgers
Second, when students trust University published an
extraordinary review of
the teacher, they are more research studies on the
likely to accept the feedback effects of feedback in
schools, colleges, and
and act on it. workplaces (Kluger &
DeNisi, 1996). They
began by tracking down a copy of every single
published study on feedback they could find,
going back to 1905! They found around 3,000
(2,500 journal articles and 500 technical reports).
They then analyzed the studies to see whether
the conclusions could be trusted. They eliminated those without a control group that was not
given feedback, those for which the effects could
not be attributed only to the feedback given, and
those for which there was not sufficient detail to
quantify the impact of feedback on achievement.
They were surprised to discover that only 131
studies made the cut. Even more surprising was
that while feedback did increase achievement on
average, in 50 of the studies (i.e., 38%), feedback
actually made performance worse!
They concluded that, from a scientific point
of view, most of the studies that had been un-

when to push each student

dertaken were a waste of time because they completely failed to take into account the reactions
of the recipient. The question What kind of
feedback is best? is meaningless, because while
a particular kind of feedback might make one
student work harder, it might cause another student to give up. There can be no simple recipe
for effective feedback; there is just no substitute
for the teacher knowing their students. Why?
First, knowing the students allows the teacher
to make better judgments about when to push
each student and when to back off. Second, when
students trust the teacher, they are more likely
to accept the feedback and act on it. Ultimately,
the only effective feedback is that which is acted
upon, so that feedback should be more work for
the recipient than the donor.

Students as Learning Resources for

One Another
There is a large body of literature on peer tutoring and collaborative learningparticularly
in English language artsthat it is neither possible nor necessary to review here (e.g., Brown &
Campione, 1995 [reciprocal instruction]; Slavin,
Hurley, & Chamberlain, 2003 [general models
of collaborative learning]). However, it is worth
noting that peers can be very effective assessors
of one anothers work, especially when the focus is on improvement rather than grading. One
sixth-grade class was working on suspense stories, and the teacher had co-constructed with the
students a checklist of four key phases that made
a good suspense story: establishment, build-up,
climax, and resolution. The class also decided
that it would be a good idea, just as an exercise if
nothing else, for the story to contain at least two
examples of figurative language.
The students worked on their stories and,
when everyone was done, exchanged their work
with a neighbor and switched roles from author to editor. The editors task was to mark
up the story by using four different colored pencils to indicate the beginning of each phase, with
a fifth color to underline the two examples of
figurative language. With the editors approval,

Voices from the Middle, Volume 21 Number 2, December 2013

e15-20-Dec13-VM.indd 18

10/31/13 5:13 PM

Wiliam | Assessment: The Bridge between Teaching and Learning

page

a story could be submitted to the teacher (the

chief editor). Because each editor was responsible for ensuring that the required elements were
present, students took the role very seriously (not
least because they were accountable to the chief
editor!).
Of course, one could list many more examples of this kind, but what teachers routinely report is that students tend to be much tougher on
one another than most teachers would dare to be.
This is important because it suggests that with
well-structured peer-assessment, one can achieve
better outcomes than would be possible with one
adult for every student.

Students Owning Their Own Learning

As Rick Stiggins (Stiggins, Arter, Chappuis, &
Chappuis, 2004) reminds us, the most important
instructional decisions are not made by teachersthey are made by students. When students
believe they cannot learn, when challenging tasks
are just one more opportunity to find out that
you are not very smart, many students disengage.
And this is perfectly understandable. What students are really doing when they disengage is denying the teacher the opportunity to make any
judgment about what the student can doafter
all, it is better to be thought lazy than dumb. This
is why the most important word in any teachers
vocabulary is yet. When a student says, I cant do
it, the teacher responds with yet. This is more
than just sound psychology. It is actually what
we are learning about the nature of expertise and
where it comes from.
Of course individuals vary in their natural
gifts, but these differences are very small to begin
with. What happens is the small initial advantages of some students quickly become magnified
when the students with these small advantages
work hard, engage, and improve, while those who
are slightly behind avoid challenge, and thus miss
out on the chance to improve. As the title of a
recent book by Geoff Colvin (2010) makes clear,
Talent is overrated. It is practice that creates
expertise. Chess grandmasters dont have higher
IQs than average chess playersthey just prac-

19
tice more. Indeed, in almost all areas of human
expertise, from violin playing to radiography,
expertise is the result of ten years of deliberate
practice (Ericsson, Charness, Feltovich, & Hoffman, 2006).
When students come to believe that smart is
not something you are but something you get,
they seek challenging
work, and in the face In the best classrooms,
of failure, they increase students would not mind
effort. Student athletes
get this. They know that making mistakes, because
to improve, they must mistakes are evidence that
practice things they
cant yet do, rather than the work they are doing is
just simply rehearse the hard enough to make them
things they know how
to do. We need to get smarter.
students to understand
this in the English language arts classroom, too.
Ultimately, we would want students to resent
work that does not challenge them, because they
would understand that easy work doesnt help
them improve. In the best classrooms, students
would not mind making mistakes, because mistakes are evidence that the work they are doing is
hard enough to make them smarter.

Conclusion
People often want to know what works in education, but the simple truth is that everything
works somewhere, and nothing works everywhere. Thats why research can never tell teachers what to doclassrooms are far too complex
for any prescription to be possible, and variations in context make what is an effective course
of action in one situation disastrous in another.
Nevertheless, research can highlight for teachers
what kinds of avenues are worth exploring and
which are likely to be dead ends, and this is why
classroom formative assessment appears to be so
promising. Across a range of contexts, attending not to what the teacher is putting into the
instruction but to what the students are getting
out of it has increased both student engagement
and achievement.

Voices from the Middle, Volume 21 Number 2, December 2013

e15-20-Dec13-VM.indd 19

10/31/13 5:13 PM

Wiliam | Assessment: The Bridge between Teaching and Learning

page

20

Different teachers will find different aspects

of classroom formative assessment more effective for their personal styles, their students, and
the contexts in which they workso each teacher must decide how to adapt the ideas outlined
above for use in their practice. Of course, as always, more research is needed, but the breadth
of the available research suggests that if teachers
develop their practice focused on the principles
outlined above, they are unlikely to fail because
of the neglect of subtle or delicate features. There
will never be an optimal model, but as long as
teachers continue to investigate that extraordinarily complex relationship between What did
I do as a teacher? and What did my students
learn? good things are likely to happen.

and classroom practice (pp. 229270). Cambridge,

MA: MIT Press.
Claxton, G. L. (1995). What kind of learning does selfassessment drive? Developing a nose for quality:
Comments on Klenowski. Assessment in Education:
Principles, Policy, and Practice, 2, 339343.
Colvin, G. (2010). Talent is overrated: What really separates world-class performers from everybody else. New
York, NY: Penguin.
Ericsson, K. A., Charness, N., Hoffman, R. R. &
Feltovich, P. J. (Eds.). (2006). The Cambridge
handbook of expertise and expert performance. Cambridge, UK: Cambridge University Press.
Kluger, A. N., & DeNisi, A. (1996). The effects of
feedback interventions on performance: A historical
review, a meta-analysis, and a preliminary feedback
intervention theory. Psychological Bulletin, 119,
254284.
Ryle, G. (1949). The concept of mind. London, UK:
Hutchinson.

Note
1. The representation in Figure 1 was developed by Marnie Thompson and Dylan Wiliam, and was first published as Leahy, S., Lyon, C., Thompson, M., & Wiliam,
D. (2005). Classroom assessment: Minute-by-minute
and day-by-day. Educational Leadership, 63(3), 1824. It is
used here by permission of Educational Testing Service,
Princeton, NJ.

Sadler, D. R. (1989). Formative assessment and the

design of instructional systems. Instructional Science, 18, 119144.

References

Stiggins, R. J., Arter, J. A., Chappuis, J., & Chappuis, S.

(2004). Classroom assessment for student learning:
doing it rightusing it well. Portland, OR: Assessment Training Institute.

Brown, A. L., & Campione, J. C. (1995). Guided discovery in a community of learners. In K. McGilly
(Ed.), Classroom lessons: Integrating cognitive theory

Slavin, R. E., Hurley, E. A., & Chamberlain, A. M.

(2003). Cooperative learning and achievement. In
W. M. Reynolds & G. J. Miller (Eds.), Handbook
of psychology volume 7: Educational psychology (pp.
177-198). Hoboken, NJ: Wiley.

Dylan Wiliam works with schools, districts, and state and national governments all over the world
to improve education. He lives virtually at www.dylanwiliam.net and physically in New Jersey.

NCTE Literacy Education Advocacy Day 2014: February 27

Join NCTE members from across the nation for NCTEs Literacy Education Advocacy Day on Thursday, February 27, 2014. NCTE members attendingAdvocacy Day will learn the latest about literacy education issues at
the federal level and have a chance to interact with people highly involved with those issues. See http://www.ncte
.org/action/advocacyday for details.

Voices from the Middle, Volume 21 Number 2, December 2013

e15-20-Dec13-VM.indd 20

10/31/13 5:13 PM