Mixed-effects models can be used to examine the association between a categorical moderator and the magnitude of the effect size. Two approaches are available to estimate the residual between-studies variance, s 2res —namely, separate estimation within each category of the moderator versus pooled estimation across all categories. We examine, by means of a Monte Carlo simulation study, both approaches for s 2res estimation in combination with two methods, the Wald-type X 2 and F tests, to test the statistical significance of the moderator. Results suggest that the F test using a pooled estimate of s 2res across categories is the best option in most conditions, although the F test using separate estimates of s 2res is preferable if the residual heterogeneity variances are heteroscedastic.
- mixed-effects model
- residual between-studies variance
- subgroup analyses