Pledging to Behave Honestly

Abstract

There are many situations in which people are asked to pledge or to commit to behaving honestly. While there have been studies demonstrating the positive effects of pledges, some of these findings have been challenged, and there have been other studies showing that pledges could be ineffective or even counter-productive. Moreover, several important questions, such as how pledges interact with sanctions, whether their effects last, or how should they be worded, have not been examined systematically. To address these issues, we developed an innovative online paradigm – the “Online Matrix Task” – that facilitates the examination of these and related questions. We examined the effects of pledges in two studies: the first involving two phases of data collection, and the second with a larger and representative sample. We found that pledges can reduce dishonesty significantly in sequential decisions, and not just one-shot decisions. They also reduce dishonesty considerably compared to fines, repeatedly across different times, and consistently across different extents of cheating, different individuals and several different wordings. Our results establish the effects of pledges on reducing cheating, and suggest several directions for future research in order to better understand the underlying mechanisms of how and when pledges can make people behave more honestly.