Prevalence of Multiple Membership in a Statewide Longitudinal Data System

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

Abstract

This study examined the prevalence of multiple membership in a statewide longitudinal data system and applied multiple membership modeling to compare results to the traditional HLM approach. Twenty-two percent of high school students in Maryland public schools attended more than one school, and 17% of middle school students attended more than one school. Student mobility rates were largely consistent across cohorts, and students who were eligible for FARMS had higher mobility rates. Comparison of a multiple membership modeling approach and a traditional HLM approach indicated that failure to account for students’ membership in multiple schools may introduce bias into research analyses examining long-term college and career outcomes. Implications of multiple membership on multilevel model estimates will be discussed.

Authors

• Angela K. Henneberger, University of Maryland

• Yi Feng, University of Maryland - College Park

• Tessa Johnson, University of Maryland - College Park

• Yating Zheng, University of Maryland - College Park

• Bess A. Rose, University of Maryland

• Laura M. Stapleton, University of Maryland