Meta-Analysis of Complex Primary Studies

Session Type: Symposium

Abstract

The use of multivariate analysis has increased over the years in various fields, such as the social sciences, medicine, and education. Given the increased use of multivariate analyses, researchers have proposed a number of effect-size indices for use in meta-analyses of regressions and other complex studies. All proposed indices require thorough evaluation of their behavior under diverse conditions. This symposium, addresses some of the current developments in meta-analysis techniques for estimating and analyzing effects from regressions and multilevel models, and modeling heterogeneity among those effects. We also detail the state of affairs in the reporting of complex studies such as regressions and multilevel models; especially as it pertains to our ability to cumulate knowledge in meta-analyses.

Sub Unit

• SIG-Educational Statisticians

Chair

• Ariel M. Aloe, University of Iowa

Papers

• Assessing Meta-Analysis Results From Regression Models of Different Sizes - Christopher Glen Thompson, Florida State University; Ariel M. Aloe, University of Iowa

• Synthesizing Partial Effect Sizes: One- and Two-Step Approaches - Ariel M. Aloe, University of Iowa

• Combining Dependent Effects From Studies Reporting Regression Analyses - Tracey D. Gunter, Florida State University

• Dependence of Slopes From a Single Sample - Betsy J. Becker, Florida State University; Ariel M. Aloe, University of Iowa; Ingram Olkin, Stanford University

• Summary of Complex Regression Models - Ahmet Serhat Gozutok, Florida State University; Abdullah A. Alghamdi, Florida State University; Betsy J. Becker, Florida State University

Discussant

• Susan Natasha Beretvas, The University of Texas - Austin