What Happened to Fraud in Influential Procrastination and Honesty Studies Involving Dan Ariely?

The controversy surrounding Dan Ariely's research centers on two highly influential papers: a 2012 study on honesty pledges and a 2002 study on procrastination. The initial wave of allegations emerged in August 2021, when the blog Data Colada, run by researchers Uri Simonsohn, Leif Nelson, and Joe Simmons, published a detailed analysis revealing evidence of fabricated data in a 2012 paper co-authored by Ariely, Francesca Gino, and others, titled 'Signing at the Beginning Makes Ethics Salient and Decreases Dishonest Self-Reports in Comparison to Signing at the End'. This study, which claimed that signing an honesty pledge at the top of a form reduced cheating, was widely cited and influenced policy. Data Colada's analysis highlighted statistically impossible patterns in the data, such as unrealistically uniform mileage reports from an insurance company dataset.

The 2012 paper was subsequently retracted in 2021. Ariely, who was the only co-author to have received the dataset in question, denied manipulating the data, suggesting it might have been fabricated by the insurance company that provided it. However, Excel metadata indicated Ariely as both the file's creator and the last person to modify it. Duke University, where Ariely is a professor, launched a three-year investigation into the matter. By February 2024, the investigation concluded that while the data was indeed falsified, there was no evidence that Ariely had knowingly used fake data. Ariely attributed errors to negligence and a lack of awareness, stating he hadn't personally opened data files in two decades due to a disability.

The most recent developments, as of August 2026, concern Ariely and Klaus Wertenbroch's 2002 paper, 'Procrastination, Deadlines, and Performance: Self-Control by Precommitment.' This influential study, with over 2,100 citations, found that externally imposed deadlines improved performance more than self-imposed ones. In August 2026, Kyle Hyndman and Alberto Bisin published a replication of Study 2 from this paper in Psychological Science, which failed to reproduce the original results. Concurrently, on August 31, 2026, Data Colada published a two-part analysis of the original data files for Studies 1 and 2, which Hyndman had received from Ariely's MIT email in 2006. Data Colada's analysis identified multiple red flags, including duplicated observations, implausibly large effects, and non-rounded values where rounding would be expected from human estimates, leading them to conclude the data was 'severely tampered with or fabricated'.

Following Data Colada's findings, Ariely and Wertenbroch reached out to Psychological Science to request the article's retraction. Wertenbroch stated he never had access to the data for the studies, while Ariely denied that the files received by Hyndman were the actual data and did not provide alternative data. These ongoing revelations have intensified the broader 'replication crisis' and concerns about scientific integrity within behavioral economics, prompting discussions among researchers, journal editors, and institutions about strengthening data integrity and transparency. The National Academies of Science, Engineering and Medicine hosted a workshop in April 2026 to address these issues, focusing on proactive strategies to enhance transparency and accountability.