A New Way for Handling Student Mobility With Longitudinal Data in Educational Research

• In Event: Analyzing Longitudinal Data in Multilevel Modeling and Latent Growth Modeling

Abstract

This study proposes a multiple membership growth curve model (MM-GCM), a new specification of the three-level growth curve model used for multiple membership data structures when wanting the intercept to represent the final status. Real data with a multiple membership structure, where some students changed schools over time, were used to demonstrate the MM-GCM. Baseline and conditional MM-GCMs are presented, and parameter estimates are compared to two approaches – one that deletes mobile students and another that only models the final school. The results suggest that researchers using longitudinal data should not delete mobile students, and be cautious when making inferences about the impact of level-3 predictors on the growth rate and the between-schools growth rate variability if ignoring student mobility.

Authors

• Congying Sun, Georgia State University

• Audrey J. Leroux, Georgia State University

• Christopher Cappelli, Georgia State University