Brilliant: The Science of Smart

In Defense of School Testing

While the anti-testing movement gains steam, let's not deprive our children of a valuable tool for learning

  • Share
  • Read Later
Getty Images

The school year is winding down, but one faction within the world of education is ratcheting up: the anti-testing movement. More parents are pulling their kids out of end-of-year math and literacy assessments. More teachers and administrators are speaking up against testing — like the group of school district superintendents in Georgia who are calling on the state legislature to reconsider its test-based accountability system. And a national resolution condemning testing has now attracted the endorsements of more than 300 organizations and 8,500 individuals. Standardized testing is “an inadequate and often unreliable measure of both student learning and educator effectiveness,” the resolution reads in part, and “the over-reliance on high-stakes standardized testing in state and federal accountability programs is undermining educational quality and equity in U.S. public schools.”

(MORE: What Everyone Missed on the Pineapple Test Question)

Many of these complaints have merit: tests can warp the learning process and cause students enormous anxiety. But such concerns are mainly about “high-stakes” testing, in which a student’s achievement is judged, and a school’s status and funding are determined, by just a few big-deal tests. What we need, instead, is a regime of no-stakes testing. Entirely apart from their assessment function, tests are an enormously effective way to improve understanding and recall. By using the test as a learning tool instead of an evaluation instrument, we could help students avoid the perils of testing while garnering all its benefits.

(MORE: Do Mothers Hamper Their Daughters in Math?)

What are those benefits? Decades of research on the “testing effect” have documented that calling up a fact or a concept from memory actually helps us remember it better the next time. Foreign language learners who tested themselves on vocabulary words remembered 80 percent of the words they studied, according to one experiment, while learners who used other study methods remembered only about a third of them. Likewise, students learning from a science textbook who were queried about what they read retained about 50 percent more of the information than students who studied in other ways.

(MORE: Talking to Yourself: Not So Crazy After All)

No-stakes testing — in which scores aren’t counted or even calculated — is now being successfully implemented in real-life school settings. Through a collaboration with researcher Henry Roediger and his colleagues at Washington University in St. Louis, teachers at nearby Columbia Middle School have incorporated what psychologists call “retrieval practice” into seventh- and eighth-grade science and social studies courses. At the end of each class, students are given a quiz on what they just learned. The quiz is not graded; it’s intended solely to promote retention. This simple exercise has dramatically improved students’ recall of the material.

Although we often conceive of memory as something like a storage tank, and a test as a kind of dipstick that measures how much information is inside, that’s not actually how the brain works. Every time we pull up a memory, we make it stronger and more lasting — so that testing doesn’t just measure what students know, it changes what they know. Reading over material to be learned, taking notes, even engaging in teacher-favored techniques like concept-mapping: none of these practices are as effective as testing at improving retention.

(MORE: What Actors Can Teach Us About Memory and Learning)

Opponents of high-stakes testing want to replace the current regime with — in the words of the national resolution — “a system based on multiple forms of assessment which does not require extensive standardized testing, more accurately reflects the broad range of student learning, and is used to support students and improve schools.” In this vision, students would be judged, say, on a portfolio of the work they produced over the course of the semester. That seems a little pie-in-the-sky, but more power to them. More importantly, we should also be implementing exams that serve a different purpose: tests that have no stakes at all, other than helping students learn.

MORE: Ten Ideas That Are Changing Your Life


Nice treatment of a particular method to improve retention, which should be examined as an alternative to traditional or newer methods of instruction and checking for understanding.

However, this article is mis-titled. The content falls considerable short of a defense of current trends in school testing.


Great to see this articulated here, it's a fabulous example of taking the timeless analysis you've been sharing on your blog, Annie, and applying it to something very much in the news.  I love it!  

Since reading your post about how Recall is the most effective method of learning, especially vs re-reading and underlining, I've been adding the things I learn at work to a mind map each day, then to a flash-card app on my phone that I flip through on shuffle whenever I can in order to get those lessons learned well.  Chuck Frey says that learning knew knowledge consists of a process consisting of gathering, discerning, assimilating and utilizing that knowledge.  I'm working on putting some of the things he's written together with some of the things you've written and this no-stakes testing idea fits nicely into that.   My co-workers, my company's investors and I all thank you for your ongoing public work!