Thompson: Can Testing Be a Learning Tool?

I don't know what is the best thing about Make It Stick, by Peter Brown, Henry Roediger, and Mark McDaniel. Perhaps it is its concise explication of new cognitive science findings, or maybe it is the reinterpretation of education research.

Brown, Roediger, and McDaniel make a great case for forcing learners to practice “retrieval” skills. They argue for quizzes and other graded and ungraded assessments to help students draw from long term memory and to engage in “reflection” on the learning process. They don’t argue that testing is the only way to help calibrate students’ judgments, but they seem to believe it is one necessary tool that should be used frequently and, preferably, with some stakes attached.  

I worry that education reformers will misuse Make It Stick. Rather than reflect on the many ways that this outstanding book explains why test-driven accountability failed. I’m afraid it will be cited in support of the new reform meme that we need “tests worth teaching to.” But, if that happens, it is not the fault of this though-provoking analysis. 

I also read Brown et. al for advice regarding my own classroom instruction. This new synthesis of cognitive science is very consistent with the professional development I received during the last 25 years. Rightly or wrongly, my current teaching position requires frequent quizzing, and I have started to share the book’s recommendations with my students. I also wonder if I should have tried harder to incorporate frequent quizzing into my instruction when teaching regular high school history classes.

Brown, Roediger, and McDaniel make a strong case for frequent quizzing, with low stakes attached – or at least they do so in the abstract. They acknowledge the policy implications of their work while remaining agnostic about specific reforms. But, teaching is a political process, full of give and take and reaching accommodations that teachers, students, parents, and administrators can all live with.  I wonder whether the novelist and the cognitive scientists who compiled this narrative have witnessed the dynamics by which testing creates so much damage. Those who haven't taught in public school may not appreciate how much effort is necessary to teach students to unlearn the deplorable habits that are legacies of testing.

Twice, a decade or more ago, I made major efforts to implement reasonable and frequent assessment programs. I gave up on an online reading assessment after failing to persuade students to implement it as it was intended. Whenever I turned my back, too many teens would revert back to trying to memorize answers so they could pass the next test without reading. As one student explained, the purpose of online testing is “exercising your right click finger.” 

We abandoned the second major effort to give frequent quizzes not so much because of the program’s fault, as a much as the context. Our school was trying so many worthy, new programs that there simply wasn’t the time necessary to implement all of them. And, in the light of Make it Stick, that brings me back to the question as to whether I should have made more of an effort to prioritize the no-stakes quizzes during that promising time (before NCLB-type testing sucked the oxygen out of school improvements.)

Political economist Andy Sobel experimented his way to an attractive compromise which could point the path to better and more frequent assessments. He replaced midterm and final exams with nine quizzes per semester, and students responded well. In other words, Sobel did something that runs counter to the normative school reform process. When he put something new on his and his students’ plates, something equally effortful was taken off.    

I don’t deny my biases or that I disliked standardized testing as a child, and I have never cared for a competitive approach to learning. Still, when I began teaching in the wake of our system’s 1980s infatuation with bubble-in accountability, I was shocked by the way it undermined teaching and learning and fostered destructive habits. During the last decade, my students haveagain  told me horror story after horror story about the way that testing robbed them of a chance for a real education. I’ve only had a half century of contact with testing, but I find it hard to conceive of a testing system that doesn’t bring out the worst in people, and undermine educational values. 

Neither do I deny that my policy positions have been informed by our classes’ great success in using Socratic questioning and class discussions to help students retrieve information and to teach them how to reflect on it. I also have had great success in preparing students for college by weaning them off of being motivated by grades. So, I doubt I will ever be as comfortable with testing as Brown, Roediger, and McDaniel are. But, I would be open to a deal which might get us out of this test, sort, and punish mindset that is causing so much damage.  If stakes were stripped from standardized tests and replaced by frequent and low stakes quizzes, teachers would be able to draw upon the wisdom of Brown et. al, and we and our students  could be allowed to get back to real teaching and learning for mastery.- JT (@drjohnthompson)