When a professor at an anonymous "top American university" recently suspected cheating in a class, no student would admit to it, so he called in a big gun: Freakonomics author and University of Chicago economist Steven Levitt. Levitt and Ming-Jen Lin, of National Taiwan University, devised an algorithm to try to catch the miscreants. Using only the test answers from the class's 242 students, along with seating arrangements used during those tests, the duo report in the National Bureau of Economic Research that in fact more than 10% of the students were cheating "in a manner blatant enough to be detected by our approaches."
They looked at rates of shared incorrect answers, reports Business Insider, and singled out students sitting next to one another who had roughly twice as many shared incorrect answers as the rate that's expected due to sheer chance. They then devised a "clever trick" for the final exam, reports CBS News: allowing students to seat themselves and then rearranging them randomly. The dean was given the names of the 12 most suspicious students, four of whom confessed before the dean closed the investigation "due to pressure from parents. While this precluded any further admissions of guilt, the professor withheld grades of the presumptive guilty pairs until the first day of the next semester, which resulted in scholarship disqualification," they write. (Check out when people are most likely to cheat.)