Testing Categorical Moderators in Mixed-Effects Meta-analysis in the Presence of Heteroscedasticity

Maria Rubio-Aparicio, Jose Antonio Lopez-Lopez, Wolfgang Viechtbauer, Fulgencio Marin-Martinez, Juan Botella, Julio Sanchez-Meca*

*Corresponding author for this work

Research output: Contribution to journalArticleAcademicpeer-review

20 Citations (Web of Science)
171 Downloads (Pure)

Abstract

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

Original languageEnglish
Pages (from-to)288-310
Number of pages23
JournalJournal of Experimental Education
Volume88
Issue number2
Early online date7 Aug 2019
DOIs
Publication statusPublished - 3 Feb 2020

Keywords

  • Meta-analysis
  • mixed-effects model
  • subgroup analyses
  • residual between-studies variance
  • EFFECTS META-REGRESSION
  • VARIANCE ESTIMATORS
  • HETEROGENEITY
  • ANOVA

Cite this