- Behavioral science has faced challenges in replicating past studies, undermining faith in existing theories.
- A large-scale replication study was conducted on mental accounting, an important behavioral economics concept.
- The research involved samples from 21 countries, totaling 5,589 participants.
- Findings strongly support the theory of mental accounting across time, cultures, and demographic groups.
Although replication problems1 have affected some fields within behavioral science (mostly social psychology) more than others, I was excited to come across research that appears to be a very successful replication of a core idea from behavioral economics: mental accounting.
According to the theory of mental accounting, people think of money in relative rather than absolute terms. They derive pleasure not just from an object’s value, but also from the quality of the deal—its transaction utility. For example, a consumer may be more likely to buy a soft drink with a $3.50 price tag at a hotel shop than a grocery store. Similarly, consumers are more likely to travel to another store location if they are told a $15 dollar product is available there at a $5 discount than a $150 dollar product at a $5 discount, even though there is $5 remaining in their pocket in either scenario.
Mental accounting can be seen in a variety of other behaviors, showing that people treat money depending on context and factors such as the money’s origin and intended use, rather than thinking of it in terms of the “bottom line,” as in formal accounting. For example, people are more likely to spend $20 on a theater ticket if they have just lost a $20 bill than if they have to replace a lost ticket worth $20. Why? In losing the ticket, the loss is allocated to a "movie" mental account. Buying a replacement ticket adds up to a total cost of $40 in the movie account. Losing a $20 bill, on the other hand, is not recorded against the movie account, and the cost of the movie is perceived as $20.
A new study that involved dozens of researchers looked at the replicability of seven studies similar to the examples above, covering the most important effects that have been associated with mental accounting. The research was pre-registered, which means that the hypotheses tested were publicly announced before the data were collected. This enhances transparency and reduces the problems associated with reproducibility.
One of the problems addressed by the study was the global generalizability of research. Historically, a lot of social science has involved white undergraduate students from the Western industrialized world (primarily the U.S.). This study involved samples from 21 countries, each containing at least 250 participants, adding up to a total of 5,589 participants. The samples were culturally diverse, involving participants from Canada to Brazil and Denmark to Vietnam. In their data analysis, the researchers estimated seven parameters (i.e., different types of mental accounting effects) for each country, for a total of 147 parameters. The findings revealed that in 90.5 percent of those parameters, the results from the new studies were statistically significant and showed the same trend as the results reported in the original studies (i.e., more biased choices). In simpler terms, the new research supported what was previously found.
The researchers also explored whether other variables affected people’s tendency to apply mental accounting. More specifically, they looked at age, education, financial literacy, gender, and income. Findings were mixed. For example, out of the seven studies, higher age reduced mental accounting in three of the studies, while men were more biased than women in one study and women more than men in another.
The relationship between financial literacy and mental accounting is quite interesting. Intuitively, we might expect less biased decisions about money by people who know more about finances. However, financial literacy can also make people appreciate the need to keep their expenses in check, and mental accounting can help them reach that goal (despite its shortcomings). Results in this research show a small positive association between financial literacy and mental accounting—not enough to help researchers draw a convincing conclusion.
In summary, the authors conclude that their new research provides very good evidence that mental accounting is a universal behavior, aligning with the original ideas of the theory. It spans across cultures and time. The replication has practical implications for finance and education, helping individuals make better decisions and improve financial well-being.
Behavioral science has been experiencing its share of challenges in recent years, particularly in the form of what’s known as a replication crisis—the difficulty of repeating past studies and finding the same results. Numerous interrelated factors are thought to contribute to problems in recreating studies, including underpowered research (essentially small sample sizes that lead to a lack of robustness), a lack of unifying theoretical frameworks, a lack of globally generalizable samples, publication bias (a tendency to report research that supports rather than refutes hypotheses), as well as research conduct problems (from p-hacking to data falsification).
Priolo, G., Stablum, F., Vacondio, M., D'Ambrogio, S., Caserotti, M., Conte, B., … Rubaltelli, E. (2023, July 18). The robustness of mental accounting: a global perspective. https://doi.org/10.31219/osf.io/apc26