Essays on Moral Decisions

Abstract

People are often faced with decisions about right and wrong. Undoubtedly, individual dispositions and material incentives can influence those decisions to a large extent. However, beyond personal characteristics and expected costs versus benefits, what else may affect moral decisions? Using a combination of scenario-based and incentivized experiments, this dissertation comprises five articles that present causal evidence on how moral intentions and behavior can systematically vary across contexts. Article 1, conducted in the registered report format, tried to replicate and extend Conway and Peetz’s (2012) influential hypothesis that recalling behaviors from the recent (distant) past should lead to compensatory (consistent) moral behavior. With one of the largest single-lab studies (N = 5,091), investigating sequential moral behavior, we robustly show that recalling moral behavior led to higher prosocial intentions than recalling either immoral or neutral behavior, irrespective of recalling from the recent or distant past. Article 2 examines how the mere size of an organization can affect dishonest behavior against it. Across eight scenario-based and incentive-aligned experiments (combined N = 5,670), we find that people are more likely to both intend to and actually cheat big businesses than small businesses for selfish gain, rendering a meta-analytic effect size corresponding to .31 of a standard deviation. Further, based on mediation analyses, we also suggest that one important explanation of this biased dishonesty is that people perceive big businesses as less moral and less vulnerable than small businesses. Article 3 investigates intergroup bias in selfish and coalitional dishonesty. In two experiments, we tempted Democrat and Republican voters to double their earnings (or the earnings of someone else) by self-reporting a correct guess of a die-roll. In Experiment 1 (N = 1,176), we found that individuals were equally likely to cheat their political ingroup and

outgroup members to double their payoffs. In Experiment 2 (N = 1,710), participants lied at a significantly higher rate to benefit an ingroup member than to benefit an outgroup member (9 percentage points). Article 4 aims to answer a simple question: Do people believe that others are similarly, more, or less dishonest than they truly are? In a research program spanning three years (2022-24) and a total of 31 different effects (combined N = 8,127), we tempted participants to cheat without any repercussions or detection risks, and asked them to estimate what percentage of other participants would lie in the same situation. Our meta-analysis revealed a significant overestimation of others' dishonesty by an average of 14 percentage points, suggesting the world is less dishonest than people think. Article 5 tests for gender bias in interpersonal dishonesty by recruiting a total of 3,166 participants from nine countries and providing them an opportunity to cheat and increase their payoffs at the cost of another male, female, or sex-unmentioned participants. Overall, females were cheated significantly less (22%) than sex-unmentioned participants. Interestingly, the effect was significantly stronger among female decision-makers, who cheated other females substantially less (53.6%) than other male participants. Theoretically, the dissertation contributes to the moral decision-making literature across topics such as sequential moral behavior, organizational perceptions, intergroup relations, beliefs, and gender biases. The findings are relevant to fellow researchers studying both basic judgment and decision-making, and those in applied settings such as organizations and marketplaces. Methodologically, most studies in the dissertation used incentivized economic experiments to study psychological phenomena, providing behavioral evidence to respective research questions. Practically, articles in this dissertation can inform managers, policymakers, and society at large on how everyday people make moral decisions.