OB Research Seminar - Matthieu Légeret, University of Amsterdam

Despite evidence pointing to differences between hypothetical and real moral decisions, experimental research in social sciences often relies on hypothetical measures. To assess the extent of these differences, we conducted three experiments on donation decisions. We first ran two studies on Amazon Mechanical Turk (MTurk). In Study 1, we examine how levels of wealth affect the difference between hypothetical and real donation decisions (n = 1132). In Study 2, we manipulate the perceived ownership of the donation by framing donation decisions as either giving to or taking from a charity (n = 1023). To substantiate our research, we replicate Study 2 in an online experiment with a student sample (n = 516). Our results show that hypothetical donations are systematically higher than real ones, although the difference between hypothetical and real donation decisions is not affected by levels of wealth. Yet, we find that framing has an asymmetrical impact on the difference between hypothetical and real donation decisions for MTurk workers: whereas workers overstate how much they would give to a charity, hypothetical and real decisions to take from the charity did not differ. However, participants from the student sample reported overestimated donation decisions in both framings. We conclude by discussing the implications of our results for experimental methods, in terms of design as well as choice of sample.