If there is a field that puts everything, we think we know about human behavior under question… it is behavioral economics. Founded in the latter half of the twentieth century, behavioral economics goes a step further than traditional economics and boldly questions human rationality. Instead of portraying the average person as Homo economicus–or an otherwise rational, self-interested, and reasonable individual–behavioral economics grounds its theories in empirical research to shed light on human irrationality. Studies have shown that people are not only irrational, but they often act against their long-term self-interest due to the covert influences of psychological factors like stress, arousal, temptation, anxiety, and rationalization.
At the forefront of this view is an American / Israeli star researcher, author, professor, and, predictably, behavioral economist, Dan Ariely. His main preoccupations span from studying spending habits and decoding dishonesty to exploring the impact of incentives and compensation on performance and behavior. While books like Dollars and Sense and The Honest Truth About Dishonesty have garnered much acclaim–both for the author and the broader field of behavioral economics–a paper that Ariely co-authored with James Heyman made an irrational dent in his reputation.
Titled “Effort for Payment: A Tale of Two Markets,” this paper, published in 2004, suggested that the relationship between effort and payment depends on the context. In monetary contexts–also known as monetary markets–compensation is key; this is why the average employer may expect to motivate his subordinates by giving them a raise. Yet, that is not the case for social markets, where social norms are paramount. As Ariely later pointed out in his book Predictably Irrational, no man could pull out his wallet in the middle of Thanksgiving dinner and offer his mother-in-law monetary compensation for preparing a feast; this would earn him a scoff, if not a squabble, and not just from the mother-in-question. Monetary compensation is not welcome in social markets, where reciprocity, kindness, and long-term relationships come before short-term financial gain.
But back to the “Effort for Payment: A Tale of Two Markets” paper. In it, Ariely and Heyman summarized multiple experiments where participants were offered various compensation forms, including money and non-monetary rewards such as candy. The experiments demonstrated that monetary incentives led to lower levels of effort compared with social incentives because the moment that financial incentives entered the relationships, they crowded out the social motivation. These findings contributed to a deeper understanding of how different types of motivations can affect human behavior, providing valuable insights for both economic theory and practical applications in areas like human resources, marketing, and organizational behavior.
While the paper’s interpretation of data and conclusions initially garnered great interest, about seventeen years after it was originally published, the study came under some scrutiny for possibly containing a statistical issue. In 2021, Professor Gilad Feldman from Hing Kong University and his colleagues reported to the journal that while they could replicate the results of Ariely’s 2004 paper, they uncovered discrepancies between the statistical values originally reported and their significance level. Given that, so much time has passed and that the original data was no longer available (data standards require that data be kept for five years, and this was more than fifteen years), the editors of the journal Patricia Bauer and Dan Ariely issued an Expression of Concern. In their statement, they acknowledged the issue, noting that the original data was no longer available, which made it impossible to fully address the ambiguities.
Thankfully, this was another case that showed how science is self-correcting. Not only did Professor Gilad Feldman claim that he replicated the results of the original Ariely paper, thereby supporting the original findings, but Professor Gregory Francis from Purdue University identified the reason for what seemed like a critical error in the paper’s analysis: the paper had reported F-values in place of t-values. As Professor Gregory Francis noted, for one degree of freedom tests, one can report either the F-values or the t-values. In this case, Ariely reported the t-value but mistakenly reported it as an F-value. Professor Gregory Francis further showed that the concussions of the study remain the same.
In a blog post from 2024, Ariely reflected on this, attributing the discrepancies to possible issues with the software or human oversight. Right after he described the mistake as “a typo… a serious typo, but a typo nevertheless,” he suggested that they should have indicated the utilization of t-statistic; had they done that, “the table would have been just fine.” Ariely continued, admitting: “My guess is that the statistical software pointed out that this was either an F or a t, and we used the F instead of t.” And for that, he lamented, he is deeply sorry.
So, in the end, the concerns about this specific paper were mainly for nothing. The results were replicated, they were statistically significant, and so was their interpretation, But what about mistakes? What about researchers’ commitment to accurate reporting?
In general, behavioral economics, human beings, and science are not exempt from human error, bias, and oversight. The statistical error–the innocent typo–in Ariely’s paper poses a reminder that even the most meticulous researchers are vulnerable to mistakes. When these mistakes happen, it is important to point them out, acknowledge them, and explore how to reduce the chance that these mistakes will happen again. It is also clear that the scientific community needs to develop more robust practices around data preservation, replication, and ethical responsibility.
It is also clear that cases such as Ariely’s need to be brought under the spotlight, not as reasons to shed a researcher’s reputation to pieces, but as opportunities to highlight ethical behavior amid controversy and foster a culture where transparency and correction are celebrated instead of stigmatized. In any field, including behavioral economics, mistakes will inevitably occur. Human behavior is too complex, and sometimes, so is data. Yet, what ultimately matters is how the people who make these mistakes acknowledge them and how those mistakes are addressed.
In the end, the future of science rests not just on scientific precision but on how well we confront and learn from mistakes. Only our attitude holds the power to ensure that ethical standards are upheld, public trust is preserved, and progress is achieved.
