Measure for measure: Effects of money exposure, reward size and loss aversion on cheating

Abstract

People sometimes behave dishonestly to collect undeserved monetary rewards. Prior research has shown that people put more effort into avoiding monetary losses than into making gains, and accordingly they cheat more to avoid losses than to acquire the equivalent amount (loss aversion). However, there has been a lack of research about how reward size and money exposure affect levels of cheating. Using a real effort task, we implemented a between-subjects experimental design to test the effects of framing (loss vs gain), reward size (small vs large) and money exposure (money vs no money) on individual real performance and cheating levels. The results revealed no significant differences in real performance. However, for cheating levels, all two-way interaction effects turned out to be significant (i.e., frame by size – frame by exposure – size by exposure). To disentangle the effects of the loss frame on cheating levels, a double moderated model was tested with reward size and money exposure as moderators. The model was significant with conditional effects revealing that the loss frame generally causes increased cheating level unless (i) participants were informed about a possible large reward they had not been exposed to, and (ii) participants were informed about, and exposed to, a small reward. Our results offer a partial replication of the finding that the level of cheating is higher within the loss frame than in the gain framing, which suggests that the relationship between framing and cheating behaviour can be moderated by other variables such as reward size and exposure to a reward. They also pose new questions for future research about complex joint effects on cheating behaviour, such as the combined influence of framing and default choices.