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

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 estimatio...

Full description

Saved in:
Bibliographic Details
Published in:The Journal of experimental education 2020-02, Vol.88 (2), p.288-310
Main Authors: Rubio-Aparicio, María, López-López, José Antonio, Viechtbauer, Wolfgang, Marín-Martínez, Fulgencio, Botella, Juan, Sánchez-Meca, Julio
Format: Article
Language:English
Subjects:
Classification
Computation
Differences
Effect Size
Meta Analysis
mixed-effects model
Monte Carlo simulation
residual between-studies variance
Statistical Analysis
Statistical Significance
subgroup analyses
Systematic review
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary: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.
ISSN:0022-0973
1940-0683
DOI:10.1080/00220973.2018.1561404