As you may remember, last week I linked to a Schoolhouse Rock post that talked about how some teachers help their students cheat on standardized tests. I said we should at the very least rotate the teachers for the tests (so they couldn’t boost their own kids’ results), and possibly even bring all the students into a gymnasium, where all the teachers could supervise both the kids and each other. A reader wrote to Schoolhouse Rock with a similar suggestion.

It turns out the problem is widespread, and surveillance fixes it:

After my post on cheating went up on Monday, a few readers wrote to remind me of a classic work of cheating economics, “Rotten Apples: An Investigation of the Prevalence and Predictors of Teacher Cheating” . . . According to the original paper’s abstract, the authors:
estimate that serious cases of teacher or administrator cheating on standardized tests occur in a minimum of 4-5 percent of elementary school classrooms annually. . . . . . . Arne Duncan, the CEO of the Chicago public schools, contacted Levitt and Jacob after their paper was published and asked for their help in catching the cheaters. Duncan and the economists chose 120 classrooms—some classrooms with suspicious results, some regular classrooms as a control—and retested them, but with officials from Duncan’s office overseeing the test, rather than the teachers themselves. When the test results came back, the scores from the classrooms with suspicious results had dropped sharply. A dozen teachers were fired, and as Dubner and Levitt write in their book, “The final outcome of the Chicago study is further testament to the power of incentives: the following year, cheating by teachers fell more than 30 percent.”