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, t[superscript 2][subscript res] --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 t[superscript 2][subscript res] estimation in combination with two methods, the Wald-type X[superscript 2] and "F" tests, to test the statistical significance of the moderator. Results suggest that the "F" test using a pooled estimate of t[superscript 2][subscript res] across categories is the best option in most conditions, although the "F" test using separate estimates of t[superscript 2][subscript res] is preferable if the residual heterogeneity variances are heteroscedastic.