clock menu more-arrow no yes mobile

Filed under:

A study suggests it's easy to catch students cheating. So why don't colleges try?

About 10 percent of students in an introductory class at a top university cheated on their midterm exams.
About 10 percent of students in an introductory class at a top university cheated on their midterm exams.
Libby Nelson is Vox's policy editor, leading coverage of how government action and inaction shape American life. Libby has more than a decade of policy journalism experience, including at Inside Higher Ed and Politico. She joined Vox in 2014.

When a college professor suspected some of his students were cheating on a midterm, he did what anyone would do: He called economists and cheating experts to come up with an algorithm to catch them.

The economists, Steven Levitt and Ming-Jen Lin, found that 10 percent of the class had probably cheated on the final exam. In a recent paper published by the National Bureau of Economic Research, they argue that students have many incentives to cheat, and that colleges need to do something about it.

But there's probably a reason they don't. Although the college tried to investigate the cheaters with a judiciary hearing, it ended up canceling it under pressure from the accused students' parents.

When the professor — who teaches an "introductory natural sciences course" at an unnamed "top university" — suspected cheating, he warned his class that he was going to call in an expert if the guilty parties didn't confess.

Levitt, a professor of economics at the University of Chicago and one of the authors of Freakonomics, had previously researched teachers who cheat by changing their students' answers on high-stakes standardized tests. Levitt and Lin, a professor of economics at National Taiwan University, studied pairs of students who sat next to each other during exams.

Of 242 students, they found those who sat together were more likely to answer the same questions incorrectly — twice as often as would be expected by chance. Using their algorithm, they determined 10 percent of students cheated on the midterm, blatantly enough that statistics could catch them.

They forwarded the name of 12 students they suspected were cheating in pairs to the dean, and four confessed. (It's not clear if some of the students might have been copying off their neighbors without both students in a pair knowing about it.) All 12 were disqualified for scholarships because the university held their grades, and none of them complained.

Protests from parents mean the university never pursued disciplinary action — "a powerful explanation as to why so little effort is invested in detecting cheaters," the authors wrote. That's despite the fact that cheating scandals pop up with disturbing regularity, including at Harvard, where nearly half of students in one class were suspected of cheating.

And students persisted in cheating even after being warned that their professor was onto them: "'[They are] extremely good at catching cheating, if you have read Freakonomics,'" the professor wrote in an email to the class about his decision to bring in the economists, according to a footnote in the paper. It went on to note dryly, "Apparently, none of the cheaters had read Freakonomics."