How to Decompose the Between and Within Effects in Contextual Models

• In Event: Methodological Challenges in Multilevel Modeling

Abstract

In contextual studies, group compositions are often extracted from individual data in the sample, in order to estimate the group compositional effects controlling for interindividual differences. As the same variable is used at both group level and individual level, an appropriate decomposition of between and within effects is a key to providing a clearer picture of these organizational and individual processes (Zhang, Zyphur, & Preacher, 2009). A new approach with within-group finite population correction (fpc) was proposed in the current study, and compared with the manifest and latent aggregation approaches. When the within-group sampling ratio was moderate, the between effect estimates from the new approach were generally of less biases and higher observed coverage rates.

Authors

• Siwen Guo, Renmin University of China

• Richard T. Houang, Michigan State University

• William H. Schmidt, Michigan State University