Tuesday, January 17, 2017

Designing Questions to Probe Learning


If you ask me today what is the best book to read about teaching, I will say Embedded Formative Assessment by Dylan Wiliam. After the many good books that I’ve read the past five years, I don’t know why I didn’t come across this gem earlier. No doubt I will encounter better books in the future, but this one really gave me that kick-in-the-pants encouragement to make larger inroads into improving my teaching, which would hopefully improve student learning. Why “hopefully”? That’s because Wiliam is very realistic about the centrality of assessment in the learning process. This is not Assessment of the generate-more-paperwork administrative mandate pervading our institutions. It is the assessment we should all be doing as teachers embedded in our lesson plans.

In chapter two, where the author makes the case for formative assessment, there is a section titled “Assessment: The Bridge Between Teaching and Learning”. I will quote several sentences. “Assessment occupies such a central position in good teaching because we cannot predict what students will learn, no matter how we design our teaching… Students do not [necessarily] learn what we teach. If they did, we would not need to keep gradebooks. We could, instead, simply record what we have taught. But anyone who has spent any time in a classroom knows that what students learn as a result of our instruction is unpredictable. We teach what we think are good lessons, but after we collect our students’ [work], we wonder how they could have misinterpreted what we said so completely.”

Before I elaborate on the author’s suggestions, let me first describe how I presently assess student learning. Outside of occasional project work and writing assignments, I generally grade four types of content-based assessments: worksheets, quizzes, homework (or problem sets) and exams. One might call these summative assessments, although they partly function as formative assessments – at least in the standard lingo. (It’s not the grading that makes something formative or summative in my opinion.) I don’t always collect worksheets or homework, i.e., sometimes the work is ungraded, so to speak. While a significant portion of my class time is spent systematically going through worked examples (one of the best ways to learn many chemical concepts), students also have opportunities to “work problems” and get feedback from me in class. In addition I get feedback from them as I observe what difficulties they run into.

In each of the four “graded” items mentioned above, I get information on different scales. Worksheets tell me immediately which students understood the lesson material of the day, who didn’t understand it, and who helped whom. (My classes are small enough that I can circulate through all groups at least once, often more.) Quizzes tell me if students individually understood one key thing from the previous class period. (My quizzes are taken in the first five minutes of class.) Problem Sets tell me if students learned a chunk of material usually over the course of 3-4 class meetings, and in particular whether they were able to apply the concepts to different but related problems (not identical to ones worked in class). Sometimes solving the problems requires integrating multiple concepts. Exams assess individual student learning over a larger chunk, usually about a month; the final exam is cumulative Students can work together on problem sets and worksheets. Quizzes and exams are individual assessments.

What feedback do the students get? On worksheets and problem sets, the students get their graded work back along with a detailed answer key. If there is a particularly egregious common error, I discuss it briefly the next class meeting. For quizzes, I provide the answer immediately in class before I’ve seen the student responses. I grade a stack of 3x5 index cards after class. Rarely do I look through the responses in class. For graded exams, the students can see what they got right and wrong accompanied by a detailed answer key, but I don’t discuss it in class. In rare instances, when I observe a widely common error, I would mention this in class. I attribute the rarity of such an occurence to students having to work individually on exams. When they work in groups with other students, the same error tends to propagate widely – often traced to one of the more capable and confident students who first made the error.

While I think I am mostly asking the right questions on exams from a summative assessment perspective, I’m not sure I’m doing well in the earlier stages. My questions focus on what I think the students should know, but don’t necessarily reveal how they think or where a misconception may lie. Of course when I discover a common misconception when grading, I make it a point to relay this to the students. But my in-class questions might not being do the best or even the right job from a formative assessment point of view. Wiliam writes: “Questions that provide a window into students’ thinking are not easy to generate, but they are crucially important if we are to improve the quality of students’ learning.”

I think that’s the really hard part: constructing excellent questions at the formative stage. These should be questions that expose previous learning and probe for potential misconceptions. Given the brevity of his book (a good thing overall), Wiliam provides just a handful of examples of standard questions, and then builds on these with much better constructed questions. While his examples are mainly in middle school math (his training is as a math teacher), one gets the gist of the type of question that really gets at the nub of learning. I need to ask much better questions but haven’t taken the time to do so. Lest, I use this as an excuse, Wiliam addresses the matter head-on.

“One common objection […] is that teachers do not have time to develop such questions, not least because they are too busy grading, but this just shows how ineffective many of our standard classroom routines are. Every teacher has had the experience of writing the same thing on [multiple student] notebooks because the students were allowed to leave the classroom before the teacher discovered that the students had failed to understand some crucial point. So the important issue is this: does the teacher only discover this once he looks at the students’ notebooks? Viewed from tis perspective, grading can be seen as the punishment given to teachers for failing to find out that they did not achieve the intended learning when the students were in front of them.”

I consider myself appropriately schooled. What I really need to do is build up a stock of excellent questions for formative assessment, particularly at the introductory level. I have enough teaching experience to know where the tricky parts are in my courses. I should also work with others who are teaching the same sections to pool and share knowledge; we have many of these at the introductory levels. One thing I am going to do this semester is to devote some time to coming up with good in-class probing questions. And perhaps, I can do less grading which seems like a more-than-reasonable trade-off.

Although a short book, Embedded Formative Assessment is chock full of great practical examples. The author does a fantastic job getting straight to the point, providing the necessary theory and scaffolding, chooses good representative examples to make his point, and keeps moving things forward. All these are marks of an excellent teacher. His experience shows. On the other hand, as I went through the book I was partly self-berating myself for not reading this book sooner, and thinking about how I’d potentially let down hundreds of students in the course of my teaching while floundering around repeating the same mistakes. (On average I teach 100-150 students per year.) I could have built up a set of really good formative assessment questions by now. As it is, I have plenty of great summative assessment questions (that I do reuse) but a limited set of great formative assessment questions. The author anticipates my despair! Here’s what he says in the epilogue.

“The problem with being provided with so many techniques is that […] too much choice can be paralyzing and dangerous. When teachers try to change more than two or three things about their teaching at the same time, the typical result is that their teaching deteriorates and they go back to doing what they were doing before. My advice is that each teacher chooses one or two of the techniques, […] tries them out […] If they appear to be effective […] practice them until they become second nature. If they are [not] … try another technique or [modify a previous one].”

I’ve decided to pick two things to work on this semester. The first is to come up with some great formative assessment questions in class. For a start, I’ll set myself the goal of one insightful one each week that I will build into my classes. The second is one I had mentioned in a blog post last summer during an assessment exercise. I came up with a great suggestion: having students critique written answers from students in a previous year. Of course, these would be carefully curated for maximum learning to take place. (Wiliam recommends this in his book!) I got so busy this past Fall that I did not do this at all. Shame on me! This semester, I will remedy things. I dug up final exams from the equivalent class last year, and I plan to make appropriate selections when the corresponding topic is covered in class. I should also commit on blogging about my trials and errors – for accountability!

I end this post with a final quote from the author. He exhorts teachers to “accept the need to improve practice, not because [we] are not good enough, but because [we] can be even better, and focus on the things that make the biggest difference to [our] students”. What great advice! His book receives my highest recommendation.

No comments:

Post a Comment