In this post Dylan Wiliam, lead educator on Assessment for Learning in STEM Teaching, discusses how the right questions at the right time make a big difference in the classroom.
When I was teaching full-time, the decision I made most often every day was this: “Do I need to go over this one more time, or can I move on to the next thing?” I made that decision dozens of times every day, and like most teachers all over the world, I made that decision by asking a question. I would make up a question right there and then, which I would then ask the class.
Typically four or five students would raise their hands, and I would select one of those students for a response. If the answer was correct, I would say, “Good” and move on. In other words, I was using the response of a single student—a single, confident, generally high-achieving student at that—to make a decision about the learning needs of thirty or thirty-five students.
The response of a single student is not enough
I am fairly sure that if someone had asked me then whether the response of that single student was a good guide to the level of understanding of the whole class, I would have said it was not. I don’t think that I was that naïve even when I started teaching. But nobody did ask me that question, and so for years, I persisted with this rather strange idea that if one or two students had understood the idea, that the others did too, or, at the very least, that they were ready to move on having heard one of their classmates supply the correct response.
Part of this, no doubt, was wishful thinking. I had a syllabus to cover, and students who didn’t understand what they were meant to understand were going to hold up the class’s progress—a real problem in science in particular where most countries have stuffed way too much content in their curricula. Part of the problem, too, was that I knew that hearing from every single student in the class would take too long.
Teachers often say, “I’ll just get every student to explain their answer” but no teacher in the history of the world has ever asked every single member of a class to explain an answer, because by the time you have heard from ten students, everyone, including the teacher, is losing the will to live.
Well-designed questions can make a big difference
Of course, my real failure was a failure of imagination; a failure to imagine being able to come up with questions that are so well designed that if students answer correctly, then that is a pretty good indicator that they have reached the desired understanding. No question can possibly be so well designed that a correct response always indicates a good understanding of the material at hand, and an incorrect response always indicates a deficient or inadequate understanding.
But by paying careful attention to the questions we ask our students—and in particular, by planning them in advance—we can minimise the likelihood that students will get the right answer for the wrong reason, or that they will get a wrong answer for the right reason.
Now some people advocate using these questions with electronic voting systems, or ‘classroom clickers’ as they are sometimes called. Those selling this equipment point out that with such systems, one can record every student’s response. This seems like a bad idea to me for two reasons. First, if you want to create a classroom where students feel OK about making mistakes, the last thing you should do is record every single one of them in an Excel spreadsheet until the end of time.
Second, recording every single student’s response gives the teacher more information than is usable, and is, in any case, unnecessary. What a teacher needs is evidence to support the decision she or he needs to take right now. Not so much “data-driven decision making”, but more like “decision-driven data collection”; collecting the minimum amount of information that you need to make the decision you need to make in a smarter way.
Of course, designing such questions is hard. It’s practically impossible for teachers to think up good questions “on the fly” and still pretty difficult when you have time to plan these questions as part of lesson planning. The good news, however, is that your questions will be much better if you plan them with others. And that is why designing high-quality questions, and getting feedback on the questions from others, is a key part of the Assessment for Learning course.
What we can do in a short online course will be rather modest. Still, we hope that by participating in this course, you will be able to get some ideas about how you can work with your colleagues to develop your practice of formative assessment in general, and in particular to get better at writing questions that give you useful and actionable information on your students’ learning.
The work will never be finished—no matter how good you are as a teacher, you can always get better—but as my colleagues and I have worked on the course, we have become convinced that this is an excellent place to start.
Develop your ability to come up with useful questions to ask students. Join Assessment for Learning in STEM Teaching today.