The effect of minimal-induced group identity on collaborative behavior: A meta-study in experimental economics

Although Balliet et al. (2014) and Lane et al. (2016) analyze the effect of minimal-induced group identity on discrimination they do not investigates the drivers of differences in co- operation within groups. Applying meta-analytic techniques, we analyze cooperation for a subset (49) of the original meta-study (77) by Lane et al. (2016) for which we can code the rate of cooperation as a continuous variable for in- and out-groups separately. Our included games consist of the dictator, trust, ultimatum, and public good games. With the help of an OLS regression and an arcsine transformation, we study in detail which factors affect cooperative behavior with in- and out-groups separately. Additionally, we run regressions on the corresponding standard deviations to analyze the interpersonal variance within groups. The interpersonal variance only differs significantly for the returner in the trust game. In contrast to other studies, we find that students discriminate less and that the possibility of statistical discrimination increases discrimination. The latter two are level and difference effects because they influence cooperative behavior within groups and drive discrimination between groups. Additionally, we find a persistent negative time trend. We discuss the implications of our findings and suggest directions for future research.