Getting Better All the Time? Evaluating Multilevel Longitudinal Measurement Invariance in School Climate Research

• In Event: Marriott Roundtable Session 41
  In Roundtable Session: Modeling and Detecting Measurement Invariance

Abstract

School climate is critical in school improvement and principal evaluation policies. Typically, school climate indicators are constructed by aggregating responses to survey questionnaires. When school climate is measured over time, indicators can be used to assess improvement and progress toward school goals. However, such uses presuppose the surveys measure the same things in the same metric at each time point (i.e., measurement invariance). There is little methodological guidance on how to evaluate the measurement invariance of school climate indicators. This study outlines a method for evaluating factorial invariance in longitudinal designs applicable to school climate research. The method is illustrated using data from a multidistrict school climate survey. Potential benefits of the method for research and policy are discussed.

Authors

• Jonathan Schweig, The RAND Corporation

• Kun Yuan, Association of American Medical Colleges