The (Honest) Truth About Dishonesty
Ariely's research on cheating shows that most people will bend rules slightly but not dramatically — a finding that complicates simple models of self-interest. His fudge factor theory is at odds with rational-choice accounts of honesty.
Summary
Ariely reports experiments on everyday dishonesty, arguing that most cheating is done by many people cheating a small amount rather than by a few people cheating a lot.
The core experimental paradigm is the matrix task, where subjects solve simple problems for money and self-report their score under conditions that permit inflation. The consistent result is that most subjects inflate slightly and almost none inflate to the maximum. Ariely explains this with what he calls the fudge factor: people want to benefit from dishonesty while maintaining a self-image as honest, so they cheat up to the point where the self-image would be threatened.
The reported moderators include distance from money, since subjects cheat more when the reward is tokens exchanged for cash later than when it is cash; observing an in-group member cheat, which raises cheating, while observing an out-group member cheat lowers it; depletion, where subjects cheat more after exerting self-control; creativity, which predicts more cheating because creative people generate better justifications; and conflicts of interest, which shift judgement without the person being aware.
The interventions reported include signing an honesty declaration before rather than after completing a form, and recalling moral standards before the opportunity to cheat.
Readers must know that this literature has been seriously damaged. The insurance form field study reported in the book, on signing at the top of a form, failed to replicate and the original data was subsequently shown to have been fabricated, in a paper co-authored by Ariely among others. Ariely has denied responsibility for the fabrication and an institutional investigation did not establish who was responsible. Other findings in the book, including some priming effects, have replicated poorly. The central matrix task result on small widespread cheating has held up better. Treat individual claims here as unverified unless separately confirmed.
Key takeaways 7
- Most dishonesty consists of many people cheating slightly rather than a few people cheating extensively.
- People cheat up to the point where their self-image as an honest person would be threatened, and no further.
- Distance between the act and money increases cheating, so tokens and non-cash rewards are taken more freely than cash.
- Watching a member of one's own group cheat increases cheating, while watching an outsider cheat decreases it.
- Conflicts of interest change professional judgement without the professional being aware of the shift.
- Recalling a moral standard immediately before an opportunity to cheat reduces cheating.
- The signature-at-the-top finding reported in this book was based on fabricated data and should not be relied on.
Who should read it
Readers interested in behavioural findings on dishonesty who will treat the specific claims as provisional given the documented data fabrication in this literature.
About Dan Ariely
Professor of psychology and behavioral economics at Duke University who runs experiments on how people misjudge value.
Predictably Irrational reports controlled experiments showing that valuation errors are systematic and repeatable: relative comparison, the pull of free, the effect of price on expected quality. The findings bear directly on pricing, bundling, incentive design and any forecast that assumes rational buyers.
Where it sits in the reference 3
Similar books 6
Chosen by shared topics, the entries that hold them, and the people who recommended both.
