Doug Lemov's field notes

The following is an excerpt from Teach Like a Champion 2.0 about the technique Reject Self-Report.  I discuss it further in this post.

Volumes of social science literature have established that self-report is highly unreliable. Questions that ask for binary (that is, yes-no) answers are particularly suspect. Throw in group dynamics in the classroom, and for the great majority of self-report questions, what we get back isn’t really even an answer so much as a formality. Whether people understand or not, they almost always say they do, particularly in groups. The result is that we get little if any data at a moment when our instincts are telling us we should be assessing.

Let’s go back to the question, “So, that’s cellular structure and the differences between plant and animal cells. Got it, guys?” Even if people thought the question was in earnest, most would be reticent to stop a group of twenty-five people and say, “Uh, no. Actually, I don’t really understand what you mean by the rigid structure of plant cells.” Even if they could, in that moment, (1) identify that there was something they didn’t know, and (2) describe it quickly so that you could understand it, most people would be unlikely to do so in front of a group that size, out of embarrassment or fear that they would co-opt the better interests of the group. They’d assume they were the only one who didn’t get it and that it wasn’t fair to speak up.

Even more compelling than implicit social pressure is a deeper problem: often students don’t know that they missed something because, well, they missed it. If they miss your description of the rigid structure of plant cells, they may not know they missed it. Similarly, students often think they understand when they don’t, or they don’t know how to reflect much on the question of whether they really understand something. When you ask, “Do you understand the differences between the structures of plant and animal cells?” what your students may be assenting to is, for all practical purposes, “Can you think of something you know about the difference between plant and animal cells?” The vague nature of most self-report questions exacerbates this. If our questions aren’t directive, they have no reason to draw students’ attention to the many things they should know; this encourages them not only to give us false positives but, possibly, to believe them themselves.

Finally, the way we often ask these questions—with a wait time of a fraction of a second; a willingness to accept silent assent without testing; a look of relief, even, when we get silent assent because we really just want the green light to move on—intimates very clearly that we’re not expecting a response. Students know not to speak up. If they do, our response can also send the message that they weren’t really supposed to answer. A teacher might offer a perfunctory recap in response to a student who says, “No, actually, I don’t get it.” “Well, David, do you remember their shape from when we looked at them under the microscope? They had a rectangular pattern. Did they pretty much all look the same? Got it now?” This isn’t a message of “Thank you for that useful data; let’s go back to that,” so much as a message of “Here’s what you missed, David.” I’ll discuss the cultural implications of the messaging around what it means to be in error to a greater extent in Chapter Two, but for now all David knows is that he was supposed to give his assent and didn’t. He’ll do better next time.

Reject Self-Report