Dan Ariely's fabricated honesty study — the professor who faked data on dishonesty

Dan Ariely was one of the most famous behavioral economists in the world. A professor at Duke University, author of the New York Times bestseller Predictably Irrational, and the inspiration for the NBC series The Irrational. His research on dishonesty was widely cited by policymakers and companies. His 2012 paper in PNAS, co-authored with Francesca Gino and others, claimed that having people sign an honesty pledge at the beginning of a form reduced dishonest behavior. The finding was influential and used in real-world policy.

On August 17, 2021, the blog Data Colada — run by researchers Uri Simonsohn, Leif Nelson, and Joe Simmons — published evidence of fraud in the 2012 paper. The raw data showed impossible patterns consistent with tampering: identical sequences of numbers across conditions that should have been random. The data appeared to have been manually altered to produce the desired results. The Economist reported that the study was 'based on fraudulent data.' The paper was retracted.

Duke University launched a three-year investigation. In 2024, Business Insider reported that Duke concluded the data had been falsified but found no evidence that Ariely knowingly used fake data. Duke declined to comment. Ariely told the press he 'undoubtedly made a mistake' but remained a professor at Duke. The scandal did not cost him his job, but it cost him his reputation — and exposed how easily fabricated data can survive peer review when the finding is what everyone wants to believe.