Opinion  /  Rick Hess Straight Up Blog
Relying on Formative Assessments to Improve Instruction? Think Again

This week Dylan Wiliam, eclectic Wales native and emeritus professor at University College London, takes over the blog. Dylan began his career as a math teacher in London (having followed his "jazz-folk" band to the capital city) before eventually stumbling into academe. His books include Creating the Schools Our Children Need and Leadership for Teacher Learning. Across a varied career, he has taught in urban public schools, directed a large-scale testing program, spent three years as senior research director at the Educational Testing Service (ETS), and served a number of roles in university administration, including dean of a school of education at King's College London. Dylan, whose advice usually costs a pretty penny, will spend the week offering pro bono thoughts and tips to educators struggling to get school going this fall.

— Rick

All over the U.S., schools are using formative assessments to find out what their students know, and, before I go any further, let me make it clear that I think this is a good thing. Any well-run organization should be monitoring its progress toward its goals, and if I were an administrator, I would not accept anyone's assurance that things were "on track" without some evidence. As the old cliché has it, "In God we trust; others must provide data."

And while it is not essential that these assessments are developed by teachers, it is important that grade teams, and in secondary schools, subject teams, agree that the assessments being used capture what is important that students learn. If teachers within a school use different assessments for the same courses, then the scope for powerful conversations is reduced. If my Algebra 1 classes average 70 percent, and those of a colleague average 85 percent, I could argue that she set an easy test or is a lenient scorer. If we agreed on the assessment, and administered it in the same way, I've run out of excuses. We can then identify individual students who aren't making the progress they need to make and figure out what to do about it.

But if that is all we are doing under the banner of "formative assessment," then we will have pretty similar problems next year. We will, to be sure, identify students who are falling behind, but we'll probably have just as many as we did this year, because there is nothing in the process I have just described that is designed to improve instruction. By thinking about formative assessment more broadly, we can make sure that this year's students get a good deal but also make sure that next year's students get a better deal.

If, by thinking about formative assessment as a process rather than as a set of tests and tasks, we can help each teacher get better by just 1 percent, then every student taught by that teacher from then on will get the benefit of that 1 percent improvement. And if that teacher gets better by another 1 percent the year after, every student from then on will get the benefit of a 2 percent improvement, and so on; compound interest—the eighth wonder of the world—at work.

So how do we do this? The first thing is to recognize that there is no such thing as "a formative assessment" because any assessment can be used formatively. If I give students a practice AP Calculus test, rather than score it myself, the next day I might give students back their unscored papers and ask them, in groups of four, to produce the best composite response they can and then lead a whole-class discussion of the groups' responses. A test that was designed originally for an entirely summative purpose—conferring college credit—is being used to improve learning.

If I give a 2nd grader a test of 20 randomly selected number facts from 1x1 to 10x10, and he gets 15 correct, I can conclude that he knows approximately 75 percent of his number facts—a summative conclusion. However, if I notice that he seems to be having difficulty with the seven times table, that gives me something to work on with him—a formative conclusion. The same assessment, and even the same assessment information, can function both summatively and formatively. Formative and summative are therefore best thought of as descriptions of the uses that we make of assessment information, rather than of the assessments themselves.

Second, we need to be clear what our formative-assessment practices form. We can use common formative assessments to monitor student progress, as discussed above, but we can also use them to assess the alignment between standards and instruction. If we have a spreadsheet of students' scores standard by standard, we can look "horizontally" to see which students are struggling. But we can also look vertically to see which standards are being less well taught and thus improve the curriculum for next year's students.

However, the greatest impact on student achievement comes when formative assessment is used not just month by month but day to day and even minute to minute. Teachers have, of course, always used a range of informal techniques to assess student comprehension in real time. However, typically, such "checks for understanding" involve responses from just one or two self-selected, confident, articulate, and usually high-achieving students. Teachers can get better evidence by getting responses from students who have not raised their hands—what Doug Lemov calls "cold calling"—but it's even better to systematically get information from all members of the class, with an "all-student response" system.

It is possible to use sophisticated technology here, such as classroom "clickers," but in my view, if we want to create classrooms where students feel OK about making mistakes, the last thing we should do is record every single one of them in a spreadsheet. That is why simple technology—such as mini-whiteboards (in my view the most important development in educational technology since the slate) and, even simpler, as I mentioned in Monday's blog, finger-voting—is often the best technology. The teacher can immediately scan the responses and make a rapid decision about whether to go on, to reteach, to explore interesting student responses, or get students to discuss their responses with their neighbors. By getting responses from all students, rather than "the usual suspects," teaching is more engaging. Just as importantly, by having more extensive evidence, teachers can make their instruction more responsive to the needs of the whole group, rather than just those who are happy to share their ideas with the class.

We now know, from a randomized control trial involving 140 high schools in England, that helping teachers develop their use of minute-to-minute and day-by-day classroom formative assessment is one of the most cost-effective ways of improving student achievement. Groups of eight to 12 teachers met monthly, in cross-grade, teacher-led teams and made commitments about changes they were going to make in their classroom formative-assessment practice. They also held each other accountable for making those changes. Their students made approximately 10 percent to 20 percent more progress in 9th and 10th grade.

Regular common formative assessments every six to 10 weeks do have a role to play in making sure that students do not "slip through the net." But considerably greater improvements are possible if we can harness the power of formative assessment not just to monitor student progress but also to improve day-to-day instruction. When teachers have better evidence, they make better instructional decisions, and the result is better learning.

— Dylan