An Examination of Heterogeneity Estimators for Meta-Analysis With Dependent Effect Size Estimates

• In Event: Methodological Innovations for Conducting Research Syntheses

Abstract

Meta-analyses in educational research often encounter the challenge of dealing with statistically dependent effect size estimates. Robust variance estimation (RVE) methods offer a viable approach to incorporate multiple dependent effect sizes into a single meta-regression model, even in the presence of unknown dependency structures. Various working models have been employed for meta-analysis of dependent effect sizes within the RVE framework. We investigate heterogeneity estimators derived from different working models and provide guidance on their interpretation. Additionally, we present an empirical example to illustrate the disparities in heterogeneity estimates obtained from distinct working models. We conclude by discussing the implications of these findings, acknowledging limitations, and suggesting potential avenues for future research.

Authors

• Jingru Zhang, University of Wisconsin - Madison

• James E. Pustejovsky, University of Wisconsin - Madison