摘要：News of dishonest acts has become increasingly frequent, either in politics or daily life, as well as in cases of scandals involving corruption schemes in companies. To explain the dishonesty process, the Maneuver Margin Theory, developed by Ariely (2012), states that the individual performs internal "maneuvers", balancing the benefits that they will receive from the act and the self-concept that they have of themselves. To verify whether individuals are more dishonest when they are about to make decisions in groups rather than individually, we did an experiment with 250 accounting students, where they should observe the die-rolling game and then report the number seen, for example, if the respondent reported that the dice presented the number 3 (three), they would earn 3 (three) points, although the number seen had been 1 (one). The participant’s reward was linked to the number reported and not the number displayed, thus enabling them to be dishonest. Additionally, participants should be divided into two groups, one where they should enter the same number to receive. Otherwise, they would not win anything, and, in the other, where the combination of the same number was not necessary for remuneration. The experiment was programmed and conducted through software specially created for this research and the McNemar statistical test was performed to compare the two samples and to verify if individuals are more dishonest when making group decisions than individual ones. The results of the research corroborated the initial hypothesis that individuals are more dishonest when making decisions in a group (36% of participants) comparing to decisions taken individually (23%). These findings corroborate previous research (e.g., Mazar and Aggarwal, 2011; Gneezy, 2005) and begin work on group dishonesty in Brazil, showing that individuals behave dishonestly when making group decisions rather than individually. The results presented in this paper reinforce the idea that, both in groups and individually, internal considerations are important when making decisions about dishonesty. This can help to understand the motivations that lead people to be dishonest as well as to seek mechanisms that are capable of inhibiting this behavior. For future researches, it is recommended that this same experiment be expanded not only in other courses but also in universities from different regions in order to provide an overview of the dishonest behavior of Brazilian students (future decision-makers in companies), especially when they interact in a group. Another suggestion is also to increase the number of people in each group so that the experiment might resemble the decision-making in real-life companies because, in real life, the number of people engaging in dishonest acts can vary greatly.
