Not-So-Strong Evidence for Gender Differences in Risk Taking

In their article "Strong Evidence for Gender Differences in Risk Taking", Gary Charness and Uri Gneezy (2012) review a number of experimental studies regarding investments in risky assets, and claim that these yield strong evidence that females are more risk averse than males. This study replicates and extends their article, demonstrating that its methods are highly problematic. While the methods used would be appropriate for categorical, individual-level differences, the data reviewed are not consistent with such a model. Instead, modest differences (at most) exist only at aggregate levels, such as group means. The evidence in favor of gender difference is thus considerably less strong than claimed. This analysis has important implications both for the design of behavioral and economic research and for policies related to discrimination and stereotyping.