Integrating assessment with learning:

what will it take to make it work?

Dylan Wiliam and Marnie Thompson, ETS

Introduction
Improving education is a priority for all countries. Increasing the level of educational

achievement brings benefits to the individual, such as higher lifetime earnings, and to

society as a whole, both in terms of increased economic growth and lower social costs

such as health care and criminal justice costs (Gritz & MaCurdy, 1992; Hanushek, 2004;

Levin, 1972; Tyler, Murnane, & Willett, 2000). Indeed, the total return on investments in

education can be well over $10 for every $1 invested (Schweinhart, et al., 2005). This

means that even loosely focused investments in education are likely to be cost-effective.

Given public skepticism about such long-term investments, however, and given too the

reluctance of local, state, and federal governments to raise taxes, there is a pressing need

to find the most cost-effective ways of improving student achievement.

Unfortunately, for many years analysis of school effectiveness focused primarily

on easily collected outputs such as average school test scores, and, given the huge

differences in the achievement of students entering different schools, between-school

differences (what we might call the school effect) appeared to be much larger than

within-school differences (such as classroom effects). As disaggregated and longitudinal

datasets have become more available, however, it has become possible to look at how

much progress each student makes (sometimes called “value-added modeling”).

Obviously, the estimates of the relative sizes of school and teacher effects will vary
according to the models used, but it is quite common to find that the variability of the

teacher effect is much greater than the variability of the school effect (although, as Bryk

& Raudenbush, [1988] note, disentangling these two effects is very difficult). In other

words, it seems to matter more which teachers you get in a particular school than which

school you go to. Perhaps more importantly, the effect of having a good teacher appears

to be far greater than that of being in a small school or even of being in a small class.

Hanushek (2004) estimates that students taught by a teacher one standard

deviation better than the average—i.e., at the 84th percentile of effectiveness—will

achieve 0.22 standard deviations higher than those taught by the average teacher. For

example, students taught by a teacher at the 95th percentile of effectiveness will achieve

0.36 standard deviations higher than those taught by average teachers. Because the annual

increase in average achievement found in NAEP studies is approximately one-third of a

standard deviation (see, for example, NAEP, 2006), this means that students taught by

one of the best teachers will learn in six months what students taught by an average

teacher would take a year to learn.

This kind of effect is much greater than is found in most class-size reduction

studies. For example, Jepsen and Rivkin (2002) found that reducing elementary school

class size by ten students would increase the proportion of students passing typical

mathematics and reading tests by 4% and 3% respectively, equivalent to a standardized

effect size of approximately 0.1 standard deviations for mathematics, and 0.075 for

reading. In other words, increasing teacher quality by one standard deviation would

produce between two and three times the increase in student achievement yielded by a

reducing class-size by ten students. The challenge of improving student achievement at

2
reasonable cost therefore effectively reduces to a labor force problem with two possible

solutions: replacing the teachers we have with better teachers, or improving the teachers

we have.

Hanushek (2004) shows that over time even small changes in the way that

teachers are hired can lead to large improvements in the quality of the teaching force.

Keeping the quality of the teaching force as it is requires hiring at the 50th percentile—

this way, the new teachers are just as good as the ones who are leaving. Hiring instead at

the 58th percentile would, in 30 years, increase student achievement by 1 standard

deviation (and, incidentally, add 10% to per capita GDP). There is however, little

evidence that this would be possible. Some have argued that there are large numbers of

prospective teachers who would be effective practitioners but who are deterred by

burdensome requirements for certification and/or inadequate remuneration (see, e.g.,

Hess, Rotherham, & Walsh, 2004). Evidence from recent studies suggests, however, that

teachers admitted via alternate routes are no more effective than those who follow

traditional certification paths (Darling-Hammond, Holtzman, Gatlin, & Vasquez Heilig,

2005), and there is no evidence yet that raising teacher pay attracts better teachers or

results in improvements in student learning. In other words, even if one were motivated

solely by economic considerations, any significant improvement in educational outcomes

will require developing the capability of the existing workforce rather than looking for

ways of replacing it—what might be called the “love the one you’re with” approach.

Fifteen or twenty years ago, this would have resulted in a gloomy prognosis.

There was little if any evidence that the quality of teachers could be improved through

teacher professional development, and certainly not at scale. Indeed, there was a

3
widespread belief that teacher professional development had simply failed to “deliver the

goods.”

“Nothing has promised so much and has been so frustratingly wasteful as the

thousands of workshops and conferences that led to no significant change in practice

when teachers returned to their classrooms” (Fullan, 1991, p. 315).

Within the last few years, however, a clearer picture of the features of effective

teacher professional development has begun to emerge. First, teacher professional

development needs to attend to both process and content elements (Reeves, McCall, &

MacGilchrist, 2001; Wilson, & Berne, 1999). On the process side, professional

development is more effective when it is related to the local circumstances in which the

teachers operate (Cobb, McClain, Lamberg, & Dean, 2003), takes place over a period of

time rather than being in the form of one-day workshops (Cohen & Hill, 1998), and

involves the teacher in active, collective participation (Garet, Birman, Porter, Desimone,

& Herman, 1999). It is important to note, however, that some foci for teacher

professional development are more productive than others. In particular, professional

development is more effective when it has a focus on deepening teachers’ knowledge of

the content they are to teach, the possible responses of students, and strategies that can be

utilized to build on these (Supovitz, 2001). There are many approaches that satisfy both

these process and content considerations, such as lesson study (Fernandez & Yoshida,

2004), but in this chapter we will argue that a focus on the use of assessment in support of

learning, developed through teacher learning communities, promises not only the largest

4
potential gains in student achievement, but also provides a model for teacher professional

development that can be implemented effectively at scale.

In the following sections, we outline the research on the use of assessment to

support learning, sometimes termed formative assessment or assessment for learning, and

how the key ideas of formative assessment can be integrated within the broader

theoretical framework of the regulation of learning processes. We then summarize briefly

the research on teacher learning communities (TLCs) as a powerful mechanism for

teacher change, and show how TLCs are perhaps uniquely suited to improving teachers’

capabilities in using assessment in the service of learning.

Formative assessment
In 1967, Michael Scriven proposed the use of the terms “formative” and “summative” to

distinguish between different roles that evaluation might play. On the one hand, he

pointed out that evaluation “may have a role in the on-going improvement of the

curriculum” (Scriven, 1967, p. 41) while on the other, evaluation “may serve to enable

administrators to decide whether the entire finished curriculum, refined by use of the

evaluation process in its first role, represents a sufficiently significant advance on the

available alternatives to justify the expense of adoption by a school system” (pp. 41-42).

He then proposed “to use the terms ‘formative’ and ‘summative' evaluation to qualify

evaluation in these roles” (p.43).

Two years later, Benjamin Bloom (1969, p. 48) applied the same distinction to

classroom tests:

5
 Quite in contrast is the use of "formative evaluation" to provide feedback and correctives at each stage

in the teaching-learning process. By formative evaluation we mean evaluation by brief tests used by

teachers and students as aids in the learning process. While such tests may be graded and used as part of

the judging and classificatory function of evaluation, we see much more effective use of formative

evaluation if it is separated from the grading process and used primarily as an aid to teaching.

And there, for a while, things rested. Everyone agreed that formative assessment was “a

good thing”; people occasionally used the terms “formative” and “summative” to denote

different kinds of assessments, and, when asked, teachers typically said that they did

indeed use the results of their assessments to make instructional decisions. The kinds of

decisions that teachers made, and the kinds of evidence that informed them were rarely

examined however. When the evidence about teachers’ practice did begin to emerge, it

was clear that the way that teachers actually used assessment evidence to inform

instruction was rarely in the way that Bloom and others had envisaged.

Stiggins and Bridgeford (1985) found that although many teachers created their

own assessments, “in at least a third of the structured performance assessment created by

these teachers, important assessment procedures appeared not to be followed” (p. 282)

and “in an average of 40% of the structured performance assessments, teachers rely on

mental record-keeping” (p. 283). A few years later, two substantial review articles, one

by Natriello (1987) and the other by Crooks (1988), provided clear evidence that

classroom evaluation practices had substantial impact on students and their learning,

although the impact was rarely beneficial. Natriello’s review used a model of the

assessment cycle, beginning with purposes; and moving on to the setting of tasks, criteria,

and standards; evaluating performance and providing feedback and then discussing the

6
impact of these evaluation processes on students. His most significant point was that the

vast majority of the research he cited was largely irrelevant because of weak theorization,

which resulted in the conflation of key distinctions (e.g., the quality and quantity of

feedback).

Crooks’ paper had a narrower focus—the impact of evaluation practices on

students. He concluded that the summative function of assessment has been too dominant

and that more emphasis should be given to the potential of classroom assessments to

assist learning. Most importantly, assessments should emphasize the skills, knowledge,

and attitudes regarded as most important, not just those that are easy to assess. The

difficulty of reviewing relevant research in this area was highlighted by Black and

Wiliam (1998a), in their synthesis of research published since the Natriello and Crooks

reviews. The two earlier papers had cited 91 and 241 references respectively, and yet

only 9 references were common to both papers. In their own research, Black and Wiliam

found that electronic searches based on keywords either generated far too many irrelevant

sources, or omitted key papers. In the end, they resorted to manual searches of each issue

between 1987 and 1997 of 76 of the journals considered most likely to contain relevant

research. Black and Wiliam’s review (which cited 250 studies) found that effective use of

classroom assessment yielded improvements in student achievement between 0.4 and 0.7

standard deviations. A more recent review focusing on studies in higher education

(Nyquist, 2003) found similar results.

7
 Thirty-five years ago, Bloom had suggested that “evaluation in relation to the

process of learning and teaching can have strong positive effects on the actual learning of

students as well as on their motivation for the learning and their self-concept in relation to

school learning. … [E]valuation which is directly related to the teaching-learning process

as it unfolds can have highly beneficial effects on the learning of students, the

instructional process of teachers, and the use of instructional materials by teachers and

learners” (Bloom, 1969, p.50). At the time, Bloom cited no evidence in support of this

claim, but it is probably safe to conclude that the question has now been settled: attention

to classroom assessment practices can indeed have a substantial impact on student

achievement. What is less clear is what exactly constitutes effective classroom

assessment, and how the gains in student achievement that the research shows are

possible can be achieved at scale. These two issues are the main focus of this chapter.

The nature and purpose of assessments

Educational assessments are conducted in a variety of ways and their outcomes can be

used for a variety of purposes. There are differences in who decides what is to be

assessed, who carries out the assessment, where the assessment takes place, how the

resulting responses made by students are scored and interpreted, and what happens as a

result. In particular, each of these can be the responsibility of those who teach the

students. At the other extreme, all can be carried out by an external agency. Cutting

across these differences, there are also differences in the purposes that assessments serve.

Broadly, educational assessments serve three functions:

• supporting learning (formative)

8
 • certifying the achievements, or potential of individuals (summative)

• evaluating the quality of educational institutions or programs (evaluative)

Through a series of historical contingencies, we have arrived at a situation in many

countries in which the circumstances of the assessments have become conflated with the

purposes of the assessment (Black & Wiliam, 2004a). So, for example, it is often widely

assumed that the role of classroom assessment should be limited to supporting learning

and that all assessments with which we can hold educational institutions to account must

be conducted by an external agency, even though in some countries, this is not the case

(Black & Wiliam, 2005a).

In broad terms, moving from formative through summative to evaluative

functions of assessment requires data at increasing levels of aggregation, from the

individual to the institution; and from specifics of particular skills and weaknesses to

generalities about overall levels of performance. (Although, of course, evaluative data

may still be disaggregated in order to identify specific sub-groups in the population that

are not making progress, or to identify particular weaknesses in students’ performance in

specific areas, as is the case in France—see Black & Wiliam, 2005a.) It is also clear,

however, that the different functions that assessments may serve are not easy to reconcile.

For example, when information about student performance on standardized tests is used

to hold schools and districts accountable, there is pressure on teachers to raise student

performance even if this is at the expense of a narrowed curriculum. As has been shown

by the work of Linn and others (see, for example, Linn, 1994), increases in test scores

may indicate only that students are better at doing the specific items in the test. In such

9
cases, the test score provides little information about the aspects of achievement not

specifically tested.

For similar reasons, it has been argued that the uses of assessment to support

learning and to certify the achievements of individuals are so fundamentally in tension

that the same assessments cannot serve both functions adequately (Torrance, 1993).

Elsewhere, in a series of papers summarized in Newton (2003) and in Wiliam (2003a),

one of us has sketched out how an assessment system might be designed to serve all three

functions reasonably well. There are, of course, particular difficulties in implementing

such systems where there is a history of students being required to take tests that have

little or no consequences for them, as is the case in the United States. For the purposes of

this chapter, however, the crucial point is that whatever methods are used for assessing

student achievement for summative purposes, assessment still has a role to play in

supporting learning. Whether the assessment of student performance for purposes of

selection and certification, or for evaluation, is conducted through teacher judgment,

external assessments, or some combination of the two, classroom assessment must first

be designed to support learning (see Black & Wiliam, 2004b, for a more detailed

argument on this point). The remainder of this chapter considers further how this might

be done.

What is formative assessment?

In the United States, the term “formative assessment” is often used to describe

assessments that are used to provide information on the likely performance of students on

state-mandated tests—a usage that might better be described as “early-warning

10
summative.” In other contexts, the term is used to describe any feedback given to

students, no matter what use is made of it, such as telling students which items they got

correct and incorrect (sometimes called “knowledge of results”). These kinds of usages

suggest that the distinction between “formative” and “summative” applies to the

assessments themselves, but because the same assessment can be used both formatively

and summatively, as both Scriven and Bloom realized, it suggests that these terms are

more usefully applied to the use to which the information arising from assessments is put.

In some contexts, assessments that are used to support learning are described

under the broad heading “assessment for learning” (in contrast to “assessment of

learning”). This does suggest a process, rather than being a description of the nature of

the assessment itself, but the danger here is that the focus is placed on the intention

behind the use of the assessment, rather than action that actually takes place (Wiliam &

Black, 1996). Many writers use the terms “assessment for learning” and “formative

assessment” interchangeably, but Black, Harrison, Lee, Marshall, and Wiliam (2004, p.

8) distinguish between the two as follows:

Assessment for learning is any assessment for which the first priority in its design and practice is to

serve the purpose of promoting pupils’ learning. It thus differs from assessment designed primarily to

serve the purposes of accountability, or of ranking, or of certifying competence. An assessment activity

can help learning if it provides information to be used as feedback, by teachers, and by their pupils, in

assessing themselves and each other, to modify the teaching and learning activities in which they are

engaged. Such assessment becomes “formative assessment” when the evidence is actually used to adapt

the teaching work to meet learning needs.

11
For the purpose of this chapter, then, the qualifier “formative” will refer not to an

assessment, nor even to the purpose of an assessment, but to the function it actually

serves. An assessment is formative to the extent that information from the assessment is

fed back within the system and actually used to improve the performance of the system in

some way (i.e., the assessment forms the direction of the improvement).

So, for example, if a student is told that she needs to work harder, and does work

harder as a result, and consequently does indeed make improvements in her performance,

this would not be formative. The feedback would be causal, in that it did trigger the

improvement in performance, but not formative, because decisions about how to “work

harder” were left to the student. Telling students to “Give more detail” might be

formative, but only if the students knew what giving more detail meant (which is

unlikely, because if they knew what detail was required, they would probably have

provided it on the first occasion). Similarly, a “formative assessment” that predicts which

students are likely to fail the forthcoming state-mandated test is not formative unless the

information from the test can be used to improve the quality of the learning within the

system. To be formative, feedback needs to contain an implicit or explicit recipe for

future action. Sometimes this recipe will be explicit, for example when the feedback

identifies specific activities the student is to undertake. At other times, the recipe may be

implicit, such as those cases where the teacher has created a culture in the classroom

whereby students know they must incorporate any feedback from the teacher into future

drafts. Where the teacher relies on implicit recipes, it is of course incumbent on the

teacher to check that the students’ understanding of the classroom culture is the same as

the teacher’s.

12
 Another way of thinking about the distinction being made here is in terms of

monitoring assessment, diagnostic assessment, and formative assessment. An assessment

monitors learning to the extent that it provides information about whether the student,

class, school or system is learning or not; it is diagnostic to the extent that it provides

information about what is going wrong; and it is formative to the extent that it provides

information about what to do about it. A sporting metaphor may be helpful here.

Consider a young fast-pitch softballer who has an earned-run average of 10 (for readers

who know nothing about softball, that’s not good). This is the monitoring assessment.

Analysis of what she is doing shows that she is trying to pitch a rising fastball (i.e., one

that actually rises as it gets near the plate, due to the back-spin applied), but that this pitch

is not rising, and therefore ending up as an ordinary fastball in the middle of the strike

zone, which is very easy for the batter to hit. This is the diagnostic assessment, but it is of

little help to the pitcher, because she already knows that her rising fastball is not rising,

and that’s why she is giving up a lot of runs. If a pitching coach is able to see that she is

not dropping her pitching shoulder sufficiently to allow her to deliver the pitch from

below the knee, then this assessment has the potential to be not just diagnostic, but

formative. It provides the athlete with some concrete actions she can undertake in order to

improve. This use of formative recalls the original meaning of the term. In the same way

that an individual’s formative experiences are the experiences that shape the individual,

formative assessments are those that shape learning. The important point is that not all

diagnoses are instructionally tractable—an assessment can accurately diagnose what

needs attention without indicating what needs to be done to address the issue.

13
 This was implicit in Ramaprasad’s (1983) definition of feedback. According to

Ramaprasad, a defining feature of feedback was that it had an impact on the performance

on the system. Information that did not have the capacity to improve the performance of

the system was not feedback.

“Feedback is information about the gap between the actual level and the

reference level of a system parameter which is used to alter the gap in some way”

(Ramaprasad, 1983, p. 4).

In this view, formative assessments (and feedback, in Ramaprasad’s terminology)

cannot be separated from their instructional consequences, and assessments are formative

only to the extent that they impact learning (for an extended discussion on consequences

as the key part of the validity of formative assessments, see Wiliam & Black, 1996). The

other important feature of Ramaprasad’s definition is that it draws attention to three key

instructional processes:

• Establishing where the learners are in their learning

• Establishing where they are going

• Establishing what needs to be done to get them there

Traditionally, this may have been seen as the teacher’s job, but we need also to take

account of the role that the learners themselves, and their peers, play in these processes.

The framework shown in Table 1 provides a way of thinking about the key strategies

involved in formative assessment. Table 1 could be extended to include schools, districts,

or systems. Because the stance taken in this chapter is that ultimately, assessment must

14
feed into actions in the classroom in order to affect learning, this simplification seems

reasonable.

This framework suggests that formative assessment, or assessment for learning

(AfL) can be conceptualized as consisting of five key strategies and one “big idea.” The

five key strategies are

1. Clarifying and sharing learning intentions and criteria for success;

2. Engineering effective classroom discussions, questions, and learning tasks;

3. Providing feedback that moves learners forward;

4. Activating students as instructional resources for one another; and

5. Activating students as the owners of their own learning.

The “big idea” is that evidence about student learning is used to adjust instruction to

better meet student needs—in other words that teaching is adaptive to the student’s

learning needs.

Table 1

The five key strategies of formative assessment

Where the learner is going Where the learner is right now How to get there

Teacher Clarifying learning Engineering effective Providing feedback that

intentions and criteria for classroom discussions, moves learners forward

success questions, and learning tasks

that elicit evidence of

learning

15
Peer Understanding learning

intentions and criteria for

Activating students as instructional resources for one
success
another

Learner Understanding learning

intentions and criteria for

Activating students as the owners of their own learning
success

Details of how teachers have used these strategies to implement assessment for learning

in their classrooms can be found in Leahy, Lyon, Thompson, and Wiliam (2005). In the

remainder of this chapter we discuss how formative assessment may be integrated

theoretically with instructional design, and implemented in classrooms.

Formative assessment and the regulation of learning

Although the starting point for work on formative assessment was the relatively simple

idea of feedback, the formulation above presents rather a complex picture of formative

assessment, and the ways in which the elements within Table 1 relate to each other are

not straightforward. All the elements in Table 1 can, however, be integrated within the

more general theoretical framework of the regulation of learning processes as suggested

by Perrenoud (1991, 1998). The word “regulation” has an unfortunate connotation in

English, stemming from the idea of “rules and regulations.” In French, there are two ways

to translate the word regulation—règlement and régulation. The former has the

16
connotation of rules and regulations, while the latter connotes adjustment, for example in

the way that a thermostat regulates the temperature of a room. It is the latter sense that the

word is used in the idea of the regulation of learning, and although the term regulation is

not ideal to describe this sense in English, in this chapter we will continue its use, not

least because of the absence of a suitable alternative.

Within such a framework, the actions of the teacher, the learners, and the context

of the classroom are all evaluated with respect to guiding the learning towards the

intended goal. In this context, it is important to note that teachers do not create learning;

only learners can create learning. In the past, this has resulted in calls for a shift in the

role of the teacher from the “sage on the stage” to the “guide on the side.” The danger

with such a characterization is that it is often interpreted as relieving the teacher of

responsibility for ensuring that learning takes place. What we propose here is that the

teacher be regarded as responsible for “engineering” a learning environment, both in its

design and its operation.

The key features of an effective learning environment are that it creates student

engagement and that it is well-regulated. As a growing body of research on cognitive

development shows, the level of engagement in cognitively challenging environments

influences not only achievement, but also IQ itself (Dickens & Flynn, 2001; Mercer,

Dawes, Wegerif, & Sams, 2004). As well as creating engagement, effective learning

environments need to be designed so that, as far as possible, they afford or scaffold the

learning that is intended (proactive regulation). In addition, if the intended learning is not

occurring, then this becomes apparent, so that appropriate adjustments may be made

(interactive regulation).

17
 It is important to distinguish between the regulation of the activity in which the

student engages and the regulation of the learning that results. Most teachers appear to be

quite skilled at the former, but have only a hazy idea of the learning that results. For

example, when asked, “What are your learning intentions for this lesson,” many teachers

reply by saying things like, “I’m going to have them describe a friend” conflating the

learning intention with the activity (Clarke, 2003). In a way, this is understandable,

because only the activities can be manipulated directly. Nevertheless, it is clear that in

teachers who have developed their formative assessment practices, there is a strong shift

in emphasis away from regulating the activities in which students engage, and towards

the learning that results (Black, Harrison, Lee, Marshall, & Wiliam, 2003).

Proactive regulation is achieved “upstream” of the lesson itself (i.e., before the

lesson begins), through the setting up of “didactical situations” (Brousseau, 1984). The

regulation can be unmediated within such didactical situations, when, for example, a

teacher “does not intervene in person, but puts in place a ‘metacognitive culture,’ mutual

forms of teaching and the organisation of regulation of learning processes run by

technologies or incorporated into classroom organisation and management” (Perrenoud,

1998, p. 100). For example, a teacher’s decision to use realistic contexts in the

mathematics classroom can provide a source of regulation, because then students can

determine the reasonableness of their answers. If students calculate that the average cost

per slice of pizza (say) is $200, provided they are genuinely engaged in the activity, they

will know that this solution is unreasonable, and so the use of realistic settings provides a

“self-checking” mechanism. Similarly, if a teacher spends time creating a culture in the

18
classroom in which students are used to consulting and supporting each other in

productive ways, then this contributes to keeping the learning “on track.”

On the other hand, the didactical situation may be set up so that the regulation is

achieved through the mediation of the teacher—interactive regulation—when the teacher,

in planning the lesson, creates questions, prompts or activities that evoke responses from

the students that the teacher can use to determine the progress of the learning, and if

necessary, to make adjustments. Sometimes, these questions or prompts will be open-

ended questions requiring higher-order thinking—indeed such questions are essential to

creating learning environments that create student engagement. But it also important to

note that closed questions have a role here too. For example, questions like “Is calculus

exact or approximate?”, “What is the pH of 10 molar NaOH?”, or, “Would your mass be

the same on the moon?”are all closed questions, but are valuable because they frequently

reveal student conceptions different from those intended by the teacher. Many students

believe that calculus is approximate because δx approaches zero, but is not allowed

actually to be zero. The pH of 10 molar NaOH is greater than 14, and thus many students

think they must have made an error in their calculations because their use of the standard

pH indicator has led them to believe that pH cannot be above 14. And although many

students know that mass and weight are different, they do not realize that one’s mass

would be exactly the same on the moon, even though one’s weight would be much less.

These questions may not cause thinking, but they do provide the teacher with evidence

that can be used to adapt instruction to better meet the students’ learning needs.

“Upstream” planning of good questions like those above therefore creates,

“downstream”, the possibility that the learning activities may change course in light of

19
the students’ responses. These “moments of contingency”—points in the instructional

sequence when the instruction can proceed in different directions according to the

responses of the students—are at the heart of the regulation of learning.

These moments arise continuously in whole-class teaching, where teachers are

constantly having to make sense of students’ responses, interpreting them in terms of

learning needs, and making appropriate responses. But they also arise when the teacher

circulates around the classroom, looking at individual students’ work, observing the

extent to which the students are “on track.” In most teaching of mathematics and science,

the regulation of learning will be relatively tight, so that the teacher will attempt to “bring

into line” all learners who are not heading towards the particular goal sought by the

teacher—in these subjects, the telos of learning is generally both highly specific and

common to all the students in a class. In contrast, in much teaching in language arts and

social studies, the regulation will be much looser. Rather than a single goal, there is likely

to be a broad horizon of appropriate goals (Marshall, 2004), all of which are acceptable,

and the teacher will intervene to bring the learners “into line” only when the trajectory of

the learner is radically different from that intended by the teacher. Having said this,

where a science class is considering the ethical impact of scientific discoveries, or a math

class is pursuing an open-ended mathematical investigation, then the regulation is likely

to be more like that in the typical language arts classroom. Conversely, where the

language arts teacher is covering the conventions of grammar, the regulation is likely to

be more like the typical mathematics or science lesson.

Finally, it is worth noting that there are significant differences in how such

information is used. In the United States, the teacher will typically intervene with

20
individual students where they appear not to be “on track” whereas in Japan, the teacher

is far more likely to observe all the students carefully, while walking around the class,

and then will select some major issues for discussion with the whole class.

One of the features that makes a lesson “formative,” then, is that the lesson can

change course in the light of evidence about the progress of learning. This is in stark

contrast to the “traditional” pattern of classroom interaction, exemplified by the following

extract:

“Yesterday we talked about triangles, and we had a special name for triangles

with three sides the same. Anyone remember what it was? … Begins with E … equi-…”

In terms of formative assessment, there are two salient points about such an

exchange. First, little is contingent on the responses of the students, except how long it

takes to get on to the next part of the teacher’s “script,” so there is little scope for

“downstream” regulation. The teacher is interested only in getting to the word

“equilateral” in order that she can move on, and so all incorrect answers are treated as

equivalent. The teacher treats all incorrect responses as equivalent in terms of information

content; all the teacher learns is that the students didn’t “get it.”

The second point is that the situation that the teacher set up in the first place—the

question she chose to ask—has little potential for providing the teacher with useful

information about the students’ thinking, except, possibly, whether the students can recall

the word “equilateral.” This is typical in situations where the questions that the teacher

21
uses in whole-class interaction have not been prepared in advance (in other words, when

there is little or no proactive regulation).

Similar considerations apply when the teacher collects the students’ notebooks

and attempts to give helpful feedback to the students in the form of comments on how to

improve rather than grades or percentage scores. If sufficient attention has not been given

“upstream” to the design of the tasks given to the students so that they elicit conceptual

thinking on the part of students, then the teacher may find that she has nothing useful to

say to the students. Ideally, from examining the students’ responses to the task, the

teacher would be able to judge how to (a) help the learners learn better and (b) what she

might do to improve the teaching of this topic to a future class, this providing a third form

of regulation—retroactive regulation. In this way, the assessment could be formative for

the students, through the feedback she provides, and formative for the teacher herself, in

that appropriate analysis of the students’ responses might suggest how the lesson could be

improved for other students.

A common misconception about assessment—I held by both teachers and

administrators—is that testing more frequently makes that testing formative. However,

when we think of formative assessment within the framework of regulation of learning,

we can see that frequency alone does not guarantee that the information is used to

regulate teaching or learning. Frequent assessment can identify students who are not

making as much progress as expected (whether this expectation is based on some notion

of “ability,” prior achievement, or external demands made by the state). But frequent

summative testing—we might call this micro-summative—is not formative unless the

22
information that the tests yield is used in some way to modify instruction (see next

section).

System responsiveness and time-frames

Although the examples given above have focused on the classroom, it is important to note

that assessments can also be formative at the level of the school, district, and state

provided the assessments help to regulate learning. A key issue in the design of

assessment systems, if they are to function formatively as well as summatively, is the

extent to which the system can respond in a timely manner to the information made

available. Feedback loops need to be designed taking account of the responsiveness of the

system to the actions that can be used to improve its performance. The less responsive the

system, the longer the feedback loops need to be for the system to be able to react

appropriately.

For example, analysis of the patterns of student responses on a “trial run” of a

state-mandated test in a given school district might indicate that the responses made by

students in seventh grade on items involving (say) probability were lower than would be

expected given the students’ scores on the other items, and lower than the scores of

comparable students in other districts. One response to this could be a program of

professional development on teaching probability for the seventh grade mathematics

teachers in that district. Since this would take some weeks to arrange, and even longer for

it to have an effect, the “trial run” would need to be held some months before the state-

mandated test in order to provide time for the system to interpret the data in terms of the

system’s needs. The “trial run” would be formative for the district if, and only if, the

23
information generated were used to improve the performance of the system—and if the

data from the assessment actually helped to form the direction of the action taken.

For an individual teacher, the feedback loops can be considerably shorter. A

teacher might look through the same students’ responses to a “trial run” of a state test and

re-plan the topics that she is going to teach in the time remaining until the test. Such a test

would be useful as little as a week or two before the state-mandated test, as long as there

is time to use the information to re-direct the teaching. Again this assessment would be

formative as long as the information from the test was actually used to adapt the teaching,

and in particular, not only telling the teacher which topics need to be re-taught, but also to

suggest what kinds of re-teaching might produce better results.

The building-in of time for responses is a central feature of much elementary and

middle school teaching in Japan. A teaching unit is typically allocated 14 lessons, but the

content usually occupies only 10 or 11 of the lessons, allowing time for a short test to be

given in the 12th lesson, and for the teacher to use lessons 13 and 14 to re-teach aspects

of the unit that were not well-understood.

Another example, on an even shorter time-scale, is the use of “exit passes” from a

lesson. The idea here is that before leaving a classroom, each student must compose an

answer to a question that goes to the heart of the concept being taught at the end of the

lesson. On a lesson on probability for example, such a question might be, “Why can’t a

probability be greater than one?” Once the students have left, the teacher can look at the

students’ responses, and make appropriate adjustments in the plan for the next period of

instruction.

24
 The shortest feedback loops are those involved in the day-to-day classroom

practices of teachers, where teachers adjust their teaching in light of students’ responses

to questions or other prompts in “real time.” The key point in all this that the length of the

feedback loop should be tailored according to the ability of the system to react to the

feedback.

This does not mean, however, that the responsiveness of the system cannot be

changed. Through appropriate proactive regulation, responsiveness can be enhanced

considerably. Where teachers have collaborated to anticipate the responses that students

might make to a question and what misconceptions would lead to particular incorrect

responses, for example, through the process of Lesson Study practiced in Japan (Lewis,

2002), teachers are able to adapt their instruction much more quickly, even to the extent

of having alternative instructional episodes ready. In this way, feedback to the teacher

that, in the normal course of things, might need at least a day to be used to modify

instruction, could affect instruction immediately.

In the same way, a school district or state that has thought about how it might use

the information about student performance before the students’ results are available (for

example by the preparation of particular kinds of diagnostic reports—see Wiliam, 1999)

is likely to reduce considerably the time needed to use the information to improve

instruction. As in classroom examples, attention to regulation “upstream” pays dividends

“downstream.”

All this suggests that the conflicting uses of the term “formative assessment” can

be reconciled by recognizing that virtually any assessment can be formative, provided it

25
is used to make instructional adjustments and that a crucial difference between different

assessments is the length of the adjustment cycle. A terminology for the different lengths

of cycles is given in table 2.

Table 2

Cycle times for formative assessment

Type Focus Length

Long-cycle Across instructional units, More than four weeks

quarters, semesters, years

Medium-cycle Between lessons or units One day to four weeks

Short-cycle Within a single lesson Five seconds to one hour

Putting it into practice

No matter how elegantly we formulate our ideas about formative assessment, they will be

moot unless we can find ways of supporting teachers in incorporating more attention to

assessment in their own practice. There are, of course, other ways that educational

research can influence practice, such as through the design of curricula and textbooks,

although as Clements (2002) notes, these impacts are generally small. If educational

research is to have any lasting impact on practice, it must be taken up and used by

practitioners. Traditionally, researchers have engaged in a process of “disseminating”

their work to teachers, or engaging in “knowledge transfer.” Both of these metaphors

have some utility, but they suggest that all researchers need to do is to “share the results”

(English, Jones, Lesh, Tirosh, & Bussi, 2002, p. 805) of their research with practitioners

and the findings will somehow be used.

26
 The emerging research on expertise shows, however, that the process of

“knowledge transfer” cannot be one of providing instructions to novices or other non-

expert teachers in the hope that they will get better (see Wiliam, 2003b for more on this

point). This is because, put simply, all research findings are generalizations and as such

are either too general to be useful, or too specific to be universally applicable. For

example, research on feedback, such as the work of Kluger and DeNisi (1996), suggests

that task-involving feedback is to be preferred to ego-involving feedback, but what the

teacher needs to know is, “Can I say, ‘Well done’ to this student, now?” Put crudely, such

generalizations underdetermine action.

At the other end of the expertise continuum, expert teachers can often see that a

particular recipe for action is inappropriate in some circumstances, although because their

reaction is intuitive, they may not be able to discern the reason why. The message

received by the practitioner in such cases is that the findings of educational research are

not a valid guide to action.

The difficulty of “putting research into practice” is the fault neither of the teacher

nor of the researcher. Because our understanding of the theoretical principles underlying

successful classroom action is weak, research cannot tell teachers what to do. Indeed,

given the complexity of classrooms, it seems likely that the positivist dream of an

effective theory of teacher action—which would spell out the “best” course of action

given certain conditions—is not just difficult and a long way off, but impossible in

principle (Wiliam, 2003b).

27
 What is needed instead is an acknowledgement that what teachers do in “taking

on” research is not a more or less passive adoption of some good ideas from someone

else but an active process of knowledge creation:

Teachers will not take up attractive sounding ideas, albeit based on extensive research, if these are

presented as general principles which leave entirely to them the task of translating them into everyday

practice—their classroom lives are too busy and too fragile for this to be possible for all but an

outstanding few. What they need is a variety of living examples of implementation, by teachers with

whom they can identify and from whom they can both derive conviction and confidence that they can

do better, and see concrete examples of what doing better means in practice (Black & Wiliam, 1998b, p.

15).

There are, of course, many professional development structures that would be consistent

with the emerging research base, but professional learning communities, or teacher

learning communities (TLCs), as advocated in the Standards for Staff Development of the

National Staff Development Council (2001), appear to provide the most appropriate

vehicle for this work (Borko, 1997, 2004; Borko, Mayfield, Marion, Flexer, & Cumbo,

1997; Elmore, 2002; Kazemi & Franke, 2003; McLaughlin & Talbert, 1993; Putnam &

Borko, 2000; Sandoval, Deneroff, & Franke, 2002).

There are several reasons that TLCs are particularly appropriate for the

development of teacher expertise in formative assessment. First, formative assessment

depends upon a high level of professional judgment on the part of teachers, so it is

appropriate to build professional development around a teacher-as-local-expert model.

Second, school-embedded TLCs are sustained over time, allowing change to occur

developmentally. Third, TLCs are a non-threatening venue allowing teachers to notice

28
weaknesses in their content knowledge and get help with these deficiencies—in

discussing a formative assessment practice that revolves around specific content (e.g., by

examining student work that reveals student misconceptions), teachers often confront

gaps in their own subject-matter knowledge, which can be remedied in conversations

with their colleagues.

Fourth, TLCs are embedded in the day-to-day realities of teachers’ classrooms

and schools, and thus provide a time and place where teachers can hear real-life stories

from colleagues that show the benefits of adopting these techniques in situations similar

to their own. These stories provide “existence proofs” that these kinds of changes are

feasible with the exact kinds of students that a teacher has in his or her classroom—which

contradicts the common lament, “Well, that’s all well and good for teachers at those

schools, but that won’t work here with the kinds of students we get at this school.”

Without this kind of local reassurance, furthermore, there is little chance teachers will

risk upsetting the prevailing “classroom contract” (Brousseau, 1997, p. 31). Although

limiting, the old contract at least allows teachers to maintain some form of order and

matches the expectations of most principals and colleagues. As teachers adjust their

practice, they are risking both disorder and less-than-accomplished performance on the

part of their students and themselves. Being a member of a community of teacher-

learners engaged together in a change process provides the support teachers need to take

such risks.

Fifth, and perhaps most importantly, TLCs allow us to address a fundamental

limitation of the formative assessment intervention, which is its (perhaps paradoxical)

generality and specificity. The five formative assessment strategies that we identified

29
above are quite general—we have seen each of them in use in pre-kindergarten classes, in

graduate-level studies, and at every level in between, and across all subjects. Yet

implementing them effectively makes significant demands on subject knowledge.

Teachers need good content knowledge to ask good questions, to interpret the responses

of their students, and to provide appropriate feedback that focuses on what to do to

improve. A less obvious need for subject-matter knowledge is that teachers need a good

overview of the subject matter in order to be clear about what the “big ideas” are in a

particular domain so that these are given greater emphasis. Thus TLCs provide a forum

for supporting teachers in converting the broad formative assessment strategies into

“lived” practices within their classrooms.

Creating the conditions and support mechanisms for establishing and sustaining

TLCs in schools is non-trivial, and, indeed, challenges long-held structures and

assumptions about the nature of teachers’ work, teacher learning, and how time should be

spent in school. It is our contention, though, that getting teachers to integrate assessment

for learning directly into their practice will have such positive effects on student learning,

that it is worth investing research and development efforts in support of developing

scalable TLC models.

Conclusion
In this talk, we have argued that the terms formative and summative apply not to

assessments themselves, but to the functions they serve, and as a result, it is possible for

the same assessment to be both formative and summative. Assessment is formative when

the information arising from the assessment is fed back within the system and is actually

used to improve the performance of the system. Assessment is formative for individuals

30
when they use the feedback from the assessment to improve their learning. Assessment is

formative for teachers when the outcomes from the assessment, appropriately interpreted,

help them improve their teaching, either on specific topics, or generally. Assessments are

formative for schools and districts if the information generated can be interpreted and

acted upon in such a way as to improve the quality of learning within the schools and

districts.

The view of assessment presented here involves a shift from quality control in

learning to quality assurance in learning. Rather than teaching students, and then, at the

end of the teaching, finding out what has been learned, it seems obvious that what we

should do is to assess the progress of learning whilst it is happening, so that we can adjust

the teaching if things are not working. In order to achieve this, the length of the cycle

from evidence to action must be designed taking into account the responsiveness of the

system. Some feedback loops, such as those in the classroom, will be only seconds long,

while others, such as those involving districts or state systems will last months, or even

years.

More generally, we have suggested that formative assessment be considered as a

key component of well-regulated learning environments. From this perspective, the task

of the teacher is to not necessarily to teach, but rather to engineer and regulate situations

in which students learn effectively. One way to do this is to design the environment so

that the regulation is embedded within features of the environment. Alternatively, when

the regulation is undertaken through the teacher’s mediation, it is necessary to build

opportunities for such mediation into the instructional sequence by designing in episodes

31
 that will elicit students’ thinking (proactive regulation) and to use the evidence from these

probes to modify the instruction (interactive regulation).

Work with teachers to date suggests that the development of teachers’ formative

assessment practices through the use of teacher learning communities is manageable and

relatively inexpensive to implement. The changes are slow to take effect, however, and it

is not yet clear how faithfully the model used here can be scaled up. We have begun to

explore what, exactly, changes when teachers develop formative assessment (Black &

Wiliam, 2005b), but much more remains to be done. In particular, we do not know

whether some of the five key strategies identified above have greater leverage than

others, both for promoting professional development and for increasing student

achievement. Nevertheless, we believe that there are reasons to be optimistic. Perhaps

one day we will not talk about “integrating assessment with learning” because the

distinction between the two will be meaningless.

Acknowledgements
We are grateful to Carol Dwyer, Laura Goe, and Siobhan Leahy for comments on an

earlier version of this chapter.

References
Black, P., Harrison, C., Lee, C., Marshall, B., & Wiliam, D. (2003). Assessment for learning:

Putting it into practice. Buckingham, UK: Open University Press.

32
Black, P., Harrison, C., Lee, C., Marshall, B., & Wiliam, D. (2004). Working inside the black

box: Assessment for learning in the classroom. Phi Delta Kappan, 86(1), 8-21.

Black, P. J., & Wiliam, D. (1998a). Assessment and classroom learning. Assessment in

Education: Principles, Policy, and Practice, 5(1), 7-73.

Black, P. J., & Wiliam, D. (1998b). Inside the black box: Raising standards through classroom

assessment. London, UK: King’s College London School of Education.

Black, P. J., & Wiliam, D. (2004a). Classroom assessment is not (necessarily) formative

assessment (and vice-versa). In M. Wilson (Ed.) Towards coherence between classroom

assessment and accountability: 103rd yearbook of the National Society for the Study of

Education (part 2). Chicago: University of Chicago Press.

Black, P. J., & Wiliam, D. (2004b). The formative purpose: Assessment must first promote

learning. In M. Wilson (Ed.) Towards coherence between classroom assessment and

accountability: 103rd yearbook of the National Society for the Study of Education (part

2). Chicago: University of Chicago Press.

Black, P., & Wiliam, D. (2005a). Lessons from around the world: How policies, politics and

cultures constrain and afford assessment practices. Curriculum Journal, 16(2), 249-261.

Black, P., & Wiliam, D. (2005b). Developing a theory of formative assessment. In J. Gardner

(Ed.), Assessment and learning. London: Sage.

Bloom, B. S. (1969). Some theoretical issues relating to educational evaluation. In R. W. Tyler

(Ed.), Educational evaluation: New roles, new means. The 63rd yearbook of the National

33
 Society for the Study of Education, part 2. (Vol. 69, pp. 26-50). Chicago: University of

Chicago Press.

Borko, H. (1997). New forms of classroom assessment: Implications for staff development.

Theory into practice, 36(4), 231-238.

Borko, H. (2004). Professional development and teacher learning: Mapping the terrain.

Educational Researcher, 33(8), 3-15.

Borko, H., Mayfield, V., Marion, S. F., Flexer, R. J., & Cumbo, K. (1997). Teachers' developing

ideas and practices about mathematics performance assessment: Successes, stumbling

blocks, and implications for professional development (Vol. 423). Boulder: University of

Colorado, Boulder Center for Research on Evaluations, Standards and Student Testing

(CRESST).

Brousseau, G. (1984). The crucial role of the didactical contract in the analysis and construction

of situations in teaching and learning mathematics. In H. G. Steiner (Ed.), Theory of

mathematics education, pp. 110-119. Occasionnel Paper 54, Bielefeld, Germany,

University of Bielefeld, Institut für Didaktik der Mathematik.

Brousseau, G. (1997). Theory of didactical situations in mathematics (N. Balacheff, M. Cooper,

R. Sutherland, & V. Warfield, Trans.). Dordrecht, Netherlands: Kluwer.

Bryk, A. S., & Raudenbush, S. W. (1988). Toward more appropriate conceptualization of

research on school effects: A three-level hierarchical linear model. American Journal of

Education, 97, 65-108.

34
Clarke, S. (2003). Enriching feedback in the primary classroom. London: Hodder & Stoughton.

Clements, D. H. (2002). Linking research and curriculum development. In L. D. English (Ed.)

Handbook of international research in mathematics education (pp. 599-630). Mahwah,

NJ: Lawrence Erlbaum Associates.

Cobb, P., McClain, K., Lamberg, T. d. S., & Dean, C. (2003). Situating teachers' 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: University of Pennsylvania Consortium for Policy

Research in Education.

Crooks, T. J. (1988). The impact of classroom evaluation practices on students. Review of

Educational Research, 58(4), 438-481.

Darling-Hammond, L., Holtzman, D. J., Gatlin, S. J., & Vasquez Heilig, J. (2005). Does teacher

preparation matter? Evidence about teacher certification, teach for America, and

teacher effectiveness. Stanford, CA: Stanford University School of Education.

Dickens, W. T., & Flynn, J. R. (2001). Heritability estimates versus large environmental effects:

The IQ paradox resolved. Psychological Review, 108, 346-369.

Elmore, R. F. (2002). Bridging the gap between standards and achievement: Report on the

imperative for professional development in education. Washington, DC: Albert Shanker

Institute.

English, L. D., Jones, G., Lesh, R., Tirosh, D., & Bussi, M. B. (2002). Future issues and

directions in international mathematics education research. In L. D. English (Ed.),

35
 Handbook of international research in mathematics education (pp. 787-812). Mahwah,

NJ: Lawrence Erlbaum Associates.

Fernandez, C., & Yoshida, M. (2004). Lesson study: A Japanese approach to improving

mathematics teaching and learning. Mahwah, NJ: Lawrence Erlbaum Associates.

Fullan, M. (1991). The new meaning of educational change. London: Cassell.

Garet, M. S., Birman, B. F., Porter, A. C., Desimone, L., & Herman, R. (1999). Designing

effective professional development: Lessons from the Eisenhower program. Washington,

DC: U. S. Department of Education.

Marshall, B. (2004). Goals or horizons—the conundrum of progression in English: Or a possible

way of understanding formative assessment in English. Curriculum Journal, 15(2), 101-

113.

Gritz, R. M., & MaCurdy, T. (1992). Participation in low-wage labor markets by young men

(Discussion Paper). Washington, DC: Bureau of Labor Statistics.

Hanushek, E. A. (2004). Some simple analytics of school quality (NBER working paper No.

W10229). Washington, DC: National Bureau of Economic Research.

Hess, F. M., Rotherham, A. J., & Walsh, K. (Eds.). (2004). A qualified teacher in every

classroom? Appraising old answers and new ideas. Cambridge, MA: Harvard Education

Press.

Jepsen, C., & Rivkin, S. G. (2002). What is the tradeoff between smaller classes and teacher

quality? (NBER working paper No. 9205). Cambridge, MA: National Bureau of

Economic Research.

36
Kazemi, E., & Franke, M. L. (2003). Using student work to support professional development in

elementary mathematics: A CTP working paper. Seattle, WA: Center for the Study of

Teaching and Policy.

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(2), 254-284.

Leahy, S., Lyon, C., Thompson, M., & Wiliam, D. (2005). Classroom assessment: Minute-by-

minute and day-by-day. Educational Leadership, 63(3), 18-24.

Levin, H. M. (1972). The costs of inadequate education. In Select Committee on Equal

Educational Opportunity - United States Senate (Ed.), Toward equal educational

opportunity (pp. 171-186). Washington, DC: U.S. Government Printing Office.

Lewis, C. C. (2002). Lesson study: A handbook of teacher-led instructional change.

Philadelphia: Research for Better Schools.

Linn, R. L. (1995). Assessment-based reform: Challenges to educational measurement (Report

No. PIC-ANG1). Princeton, NJ: William H. Angoff Memorial Lecture Series,

Educational Testing Service.

McLaughlin, M., & Talbert, J. (1993). Contexts that matter for teaching and learning: Strategic

opportunities for meeting the nation's educational goals. Palo Alto, CA: Stanford

University Center for Research on the Context of Secondary School Teaching.

37
Mercer, N., Dawes, L., Wegerif, R., & Sams, C. (2004). Reasoning as a scientist: Ways of

helping children to use language to learn science. British Educational Research Journal,

30(3), 359-377.

National Assessment of Educational Progress. (2006). The nation's report card: Mathematics

2005 (NCES 2006-453). Washington, DC: Institute of Education Sciences.

National Staff Development Council. (2001). NSDC standards for staff development. Oxford,

OH: National Staff Development Council.

Natriello, G. (1987). The impact of evaluation processes on students. Educational Psychologist,

22(2), 155-175.

Newton, P. (2003). The defensibility of national curriculum assessment in England. Research

Papers in Education, 18(2), 101-127.

Nyquist, J. B. (2003) The benefits of reconstruing feedback as a larger system of formative

assessment: A meta-analysis. Unpublished Master of Science thesis, Vanderbilt

University, Nashville, TN.

Perrenoud, P. (1991). Towards a pragmatic approach to formative evaluation. In P. Weston (Ed.)

Assessment of pupil achievement (pp. 79-101). Amsterdam, Netherlands: Swets &

Zeitlinger.

Perrenoud, P. (1998). From formative evaluation to a controlled regulation of learning. Towards

a wider conceptual field. Assessment in Education: Principles, Policy, and Practice, 5(1),

85-102.

38
Putnam, R. T., & Borko, H. (2000). What do new views of knowledge and thinking have to say

about research on teacher learning? Educational Researcher, 29(1), 4-15.

Ramaprasad, A. (1983). On the definition of feedback. Behavioural Science, 28(1), 4-13.

Reeves, J., McCall, J., & MacGilchrist, B. (2001). Change leadership: Planning,

conceptualization and perception. In J. MacBeath, & P. Mortimore (Eds.), Improving

school effectiveness (pp. 122-137). Buckingham, UK: Open University Press.

Sandoval, W., Deneroff, V., & Franke, M. L. (2002). Teaching, as learning, as inquiry: Moving

beyond activity in the analysis of teaching practice. Paper presented at the 2002 Annual

Meeting of the American Educational Research Association, New Orleans, LA.

Schweinhart, L. J., Montie, J., Xiang, Z., Barnett, W. S., Belfield, C. R., & Nores, M. (2005).

Lifetime effects: The High/Scope Perry Preschool study through age 40. Ypsilanti, MI:

High/Scope Educational Research Foundation.

Scriven, M. (1967). The methodology of evaluation (Vol. 1). Washington, DC: American

Educational Research Association.

Stiggins, R. J., & Bridgeford, N. J. (1985). The ecology of classroom assessment. Journal of

Educational Measurement, 22(4), 271 - 286.

Supovitz, J. A. (2001). Translating teaching practice into improved student achievement. In S. H.

Fuhrman (Ed.), From the capitol to the classroom: Standards-based reform in the States

(Vol. 2, pp. 81-98). Chicago: University of Chicago Press.

39
Torrance, H. (1993). Formative assessment: Some theoretical problems and empirical questions.

Cambridge Journal of Education, 23(3), 333-343.

Tyler, J. H., Murnane, R. J., & Willett, J. B. (2000). Estimating the labor market signaling value

of the GED (Research Brief). Boston: National Center for the Study of Adult Learning &

Literacy.

Wiliam, D. & Black, P. J. (1996). Meanings and consequences: A basis for distinguishing

formative and summative functions of assessment? British Educational Research

Journal, 22(5), 537-548.

Wiliam, D. (1999, May) A template for computer-aided diagnostic analyses of test outcome

data. Paper presented at 25th annual conference of the International Association for

Educational Assessment, Bled, Slovenia.

Wiliam, D. (2003a). National curriculum assessment: How to make it better. Research Papers in

Education, 18(2), 129-136.

Wiliam, D. (2003b). The impact of educational research on mathematics education. In A. Bishop,

M. A. Clements, C. Keitel, J. Kilpatrick, & F. K. S. Leung (Eds.), Second international

handbook of mathematics education (pp. 469-488). Dordrecht, Netherlands: Kluwer

Academic Publishers.

Wiliam, D., Lee, C., Harrison, C., & Black, P. J. (2004). Teachers developing assessment for

learning: Impact on student achievement. Assessment in Education: Principles, Policy,

and Practice, 11(1), 49-65.

Wilson, S. M. & Berne, J. (1999). Teacher learning and the acquisition of professional

knowledge: An examination of research on contemporary professional development. In

40
A. Iran-Nejad, & P. D. Pearson (Eds.), Review of research in education (pp. 173-209).

Washington, DC: American Educational Research Association.

41