The best value in formative assessment

Abstract

Ready-made benchmark tests cannot substitute for day-to-day formative assessment conducted by assessment-literate teachers. Recently a school leader asked us to provide an example of a good test item on a formative assessment and then show how that item would be different when used on a summative test. He wanted to explain to his staff the difference between formative and summative assessment. His end goal was for teachers to develop assessments to measure how well students were mastering the content standards that would appear on the state accountability test before the test was given in the spring. His question reflects the confusion many educators have about formative and summative assessment. This confusion isn't surprising: Definitions of formative assessment abound, resulting in multiple and sometimes conflicting understandings. And in part because of these varying definitions and views, practices labeled as formative assessment in schools today vary widely. One result of No Child Left Behind has been a surge in student testing—much of it voluntary, going well beyond what federal law or state assessment systems require. Many schools and districts administer tests with names like benchmark, short-cycle, and interim assessments to predict student performance on high-stakes tests and to identify students needing additional help. This increasingly popular level of testing has contributed to the widening scope of what is called formative assessment. Testing companies in the K–12 education market, seeking to support the trend toward more testing, sometimes advertise products as "formative assessments." This adds to the confusion by encouraging the idea that it's the test itself that's formative (Chappuis, 2005).

Full text

December 2007/January 2008 | Volume 65 | Number 4
Informative Assessment Pages 14-19
The Best Value in Formative
Assessment
Stephen Chappuis and Jan Chappuis
Ready-made benchmark tests cannot substitute for day-
to-day formative assessment conducted by assessment-
literate teachers.
Recently a school leader asked us to provide an example of a good test item on a formative
assessment and then show how that item would be different when used on a summative test. He
wanted to explain to his staff the difference between formative and summative assessment. His
end goal was for teachers to develop assessments to measure how well students were mastering
the content standards that would appear on the state accountability test before the test was
given in the spring.
His question reflects the confusion many educators have about formative and summative
assessment. This confusion isn't surprising: Definitions of formative assessment abound, resulting
in multiple and sometimes conflicting understandings. And in part because of these varying
definitions and views, practices labeled as formative assessment in schools today vary widely.
One result of No Child Left Behind has been a surge in student testing—much of it voluntary,
going well beyond what federal law or state assessment systems require. Many schools and
districts administer tests with names like benchmark, short-cycle, and interim assessments to
predict student performance on high-stakes tests and to identify students needing additional
help. This increasingly popular level of testing has contributed to the widening scope of what is
called formative assessment.
Testing companies in the K–12 education market, seeking to support the trend toward more
testing, sometimes advertise products as "formative assessments." This adds to the confusion by
encouraging the idea that it's the test itself that's formative (Chappuis, 2005).
In reality, this level of testing is often little more than a series of minisummative tests, not always
tightly aligned to what was taught in the classroom. There is nothing inherently formative in such
tests—they may or may not be used to make changes in teaching that will lead to greater student
learning.
The Difference Between Summative and Formative
What is formative assessment, then? First, it's not a product. That was the central
misunderstanding of the administrator who asked for an example of a good formative test item.
Even though assessments will continue to be labeled formative or summative, how the results are
used is what determines whether the assessment is formative or summative.
To begin, let's look at summative assessment. In general, its results are used to make some sort
of judgment, such as to determine what grade a student will receive on a classroom assignment,
measure program effectiveness, or determine whether a school has made adequate yearly
progress. Summative assessment, sometimes referred to as assessment of learning, typically
December 2007/January 2008

documents how much learning has occurred at a point in time; its purpose is to measure the level
of student, school, or program success.
Formative assessment, on the other hand, delivers information during the instructional process,
before the summative assessment. Both the teacher and the student use formative assessment
results to make decisions about what actions to take to promote further learning. It is an
ongoing, dynamic process that involves far more than frequent testing, and measurement of
student learning is just one of its components.
Summative Assessment Used in Formative Ways
Almost any assessment instrument can be used for summative or formative purposes, but some,
by design, are better suited to summative use and others to formative use. For example, state
assessments, although they may also have some limited formative use, are designed to provide
accountability data and to compare schools and districts. Because their primary purpose is
summative, the results may not be communicated in ways that teachers and students can easily
interpret and work with. Further, the results are often delivered months after the administration
of the tests. For these reasons, such state tests usually do not function well in a formative way:
They can't contribute much information to guide day-to-day instruction or help determine the
next learning steps of individual students.
Benchmark assessments, either purchased by the district from commercial vendors or developed
locally, are generally meant to measure progress toward state or district content standards and to
predict future performance on large-scale summative tests. A common misconception is that this
level of assessment is automatically formative. Although such assessments are sometimes
intended for formative use—that is, to guide further instruction for groups or individual students—
teachers' and administrators' lack of understanding of how to use the results can derail this
intention. The assessments will produce no formative benefits if teachers administer them, report
the results, and then continue with instruction as previously planned—as can easily happen when
teachers are expected to cover a hefty amount of content in a given time.
Teachers also select or develop their own summative assessments—those that count for a grade.
Compared with state and district tests, these classroom assessments can more readily be
adapted to formative use because their results are more immediately available and their learning
targets have been more recently taught. When teachers know what specific learning target each
question or task on their test measures, they can use the results to select and reteach portions of
the curriculum that students haven't yet mastered. Carefully designed common assessments can
be used this way as well.
Students, too, can use summative test results to make decisions about further study. If the
assessment items are explicitly matched to the intended learning targets, teachers can guide
students in examining their right and wrong answers in order to answer questions such as these:
zWhat are my strengths relative to the standards?
zWhat have I seen myself improve at?
zWhere are my areas of weakness?
zWhere didn't I perform as desired, and how might I make those answers better?
zWhat do these results mean for the next steps in my learning, and how should I prepare
for that improvement?
For students to make maximum use of these questions to guide further study, however, teachers
must plan and allow time for students to learn the knowledge and skills they missed on the
summative assessment and to retake the assessment. Lack of time for such learning is one of the
biggest hindrances to formatively using summative classroom assessments.
Assessment for Learning
When teachers assess student learning for purely formative purposes, there is no final mark on

the paper and no summative grade in the grade book. Rather, assessment serves as practice for
students, just like a meaningful homework assignment does. This is formative assessment at its
most valuable. Called assessment for learning, it supports learning in two ways:
zTeachers can adapt instruction on the basis of evidence, making changes and
improvements that will yield immediate benefits to student learning.
zStudents can use evidence of their current progress to actively manage and adjust their
own learning. (Stiggins, Arter, Chappuis, & Chappuis, 2006)
Assessment for learning can take many different forms in the classroom. It consists of anything
teachers do to help students answer three questions (Atkin, Black, & Coffey, 2001):
Where am I going?
zGive students a list of the learning targets they are responsible for mastering, written in
student-friendly language.
zShow students anonymous strong and weak examples of the kind of product or
performance they are expected to create and have them use a scoring guide to determine
which one is better and why.
Where am I now?
zAdminister a nongraded quiz part-way through the learning, to help both teacher and
students understand who needs to work on what.
zHighlight phrases on a scoring guide reflecting specific strengths and areas for
improvement and staple it to student work.
zHave students identify their own strengths and areas for improvement using a scoring
guide.
zHave students keep a list of learning targets for the course and periodically check off the
ones they have mastered.
How can I close the gap?
zGive students feedback and have them use it to set goals.
zHave students graph or describe their progress on specific learning targets.
zAsk students to comment on their progress: What changes have they noticed? What is
easy that used to be hard? What insights into themselves as learners have they
discovered?
When students use feedback from the teacher to learn how to self-assess and set goals, they
increase ownership of their own success. In this type of assessment environment, teachers and
students collaborate in an ongoing process using assessment information to improve rather than
j
udge learning. It all hinges on the assessment's ability to provide timely, understandable, and
descriptive feedback to teachers and students.
Feedback: The Key Difference
Feedback in an assessment for learning context occurs while there is still time to take action. It
functions as a global positioning system, offering descriptive information about the work, product,
or performance relative to the intended learning goals. It avoids marks or comments that judge
the level of achievement or imply that the learning journey is over.
Effective descriptive feedback focuses on the intended learning, identifies specific strengths,
points to areas needing improvement, suggests a route of action students can take to close the
gap between where they are now and where they need to be, takes into account the amount of
corrective feedback the learner can act on at one time, and models the kind of thinking students
will engage in when they self-assess. These are a few examples of descriptive feedback:

zYou have interpreted the bars on this graph correctly, but you need to make sure the
marks on the x and y axes are placed at equal intervals.
zWhat you have written is a hypothesis because it is a proposed explanation. You can
improve it by writing it as an "if … then … " statement.
zThe good stories we have been reading have a beginning, a middle, and an end. I see that
your story has a beginning and a middle, just like those good stories do. Can you draw and
write an ending?
zYou have described the similarities between _____ and _____ clearly in this paper, and
you have identified key differences. Work on illustrating those differences with concrete
examples from the text.
In contrast, the feedback from a summative assessment—whether given in the classroom or in a
larger context—tells teachers and students who made it to the learning destination and who
didn't. The assessment's coded, evaluative feedback—B+, 84%, Meets Standards, Great Job,
Proficient, and so on—does not identify individual student strengths and areas needing
improvement. It does not offer specific information for course correction.
Advantages of Formative Classroom Assessment
Although all formative assessment practices have the potential to increase student learning,
assessment for learning in the classroom offers a number of distinct benefits:
zThe timeliness of results enables teachers to adjust instruction quickly, while learning is in
progress.
zThe students who are assessed are the ones who benefit from the adjustments.
zThe students can use the results to adjust and improve their own learning.
When we try to teacher-proof the assessment process by providing a steady diet of ready-made
external tests, we lose these advantages. Such tests cannot substitute for the day-to-day level of
formative assessment that only assessment-literate teachers are able to conduct. The greatest
value in formative assessment lies in teachers and students making use of results to improve
real-time teaching and learning at every turn.
References
Atkin, J. M., Black, P., & Coffey, J. (2001). Classroom assessment and the national
science standards. Washington, DC: National Academies Press.
Chappuis, S. (2005). Is formative assessment losing its meaning? Education Week, 24
(44), 38.
Stiggins, R., Arter, J., Chappuis, J., & Chappuis, S. (2006). Classroom assessment for
student learning: Doing it right—using it well. Portland, OR: Educational Testing
Service.
Stephen Chappuis (schappuis@ets.org) and Jan Chappuis (jchappuis@ets.org) work with the ETS Assessment
Training Institute in Portland, Oregon (www.ets.org/ati).
Copyright © 2007 by Association for Supervision and Curriculum Development
© Copyright ASCD. All rights reserved.