The Impacts of Ignoring Multiple-Membership Data Structures in Estimating a Piecewise Growth Model

• In Event: 79.046 - Studies Addressing Issues Related to the Modeling of Multiple Membership Data

Abstract

The three-level piecewise growth model (3L-PGM) can be used to break up nonlinear growth into multiple distinct components, providing the opportunity to examine potential sources of variation in individual and contextual growth within different segments of the model. The multiple membership piecewise growth model (MM-PGM) extends the 3L-PGM to handle multiple membership data structures frequently found in educational datasets when there are mobile students who change schools across a study’s timeframe. In this study, we used a real data analysis and a simulation study to compare and evaluate estimation of the MM-PGM and 3L-PGM. MM-PGM estimates were less biased (especially in the cluster-level coefficient estimates), although we found substantial bias in level-3 variance component estimates across conditions for both models.

Authors

• Audrey J. Leroux, Georgia State University

• Christopher Cappelli, Georgia Institute of Technology

• David R J Fikis, Georgia State University