When incentives beat nudges but not bounded rationality: Partial effects of incentives on academic cheating

Abstract

The paper explores the effect of nudges on dishonest academic behavior in a 3x4 factorial treatment design. Subjects had to throw a physical die 50 times, report the outcome and were given partial credit for their participation. 435 students were assigned to the following conditions: a) a nudge presenting an authoritarian instruction; 2) a nudge referring to the college code of honor; 3) a nudge that presented reasons for not cheating; and 4) a condition presenting no instruction at all. Additionally, within each condition, the partial credit incentive obtained from participating in the study varied. Some did not have it (n=166), others had a partial credit that represented 15% of the final grade (n=145), and, lastly, some of them had a partial credit that represented 2% or less of the final grade (n=124). Our results showed a significant difference between the group that had an academic incentive and the group that did not: t(433) = -2.35, p = 0.0190. However, we did not find any significant difference between any of the nudge conditions. Among students who could obtain academic credit, the incentive’s magnitude did not have an effect on the results of the task: t(267)=-0.90, p=0.3651. This indicates that the presence of incentives increases dishonest behavior, but their magnitude does not. These results suggest that when given an academic incentive to cheat, students will be dishonest, and they shed light on Colombian culture and on the effects of the education grading system. Finally, the study showed that the nudges that are reported in the literature are not as effective as they are said to be. In fact, more research should be dedicated to the effectiveness of nudges in different cultures and it should be done to look for effective nudges on academic settings.