Fig. 2
Meta-analytic estimates and heterogeneity estimates of gender differences in the five outcome variables. (a) Plotted are the meta-analytic effect sizes and their 95% and 99.5% CIs, based on random-effects meta-analyses (using the restricted maximum likelihood estimator for τ) of k = 39 countries surveyed in the ESS, for gender differences (0 “male”, 1 “female”) in the five outcome variables. The point estimates and confidence intervals are standardized using the pooled standard deviations of the outcome variables and can be interpreted in terms of Cohen’s d units (see Methods for details). (b) Plotted are the point estimates of τ, i.e., the standard deviation of true effect sizes across the included countries, and their 95% and 99.5% CIs, for gender differences in the five outcome variables. All estimates are reported in Cohen’s d units.
