Evaluation of Level-Specific Goodness-of-fit Indices in Testing Factorial Invariance in Multilevel Data

• In Event: 65.060 - Multilevel Modeling: Nonnested Multilevel Data, Measurement Invariance, and Bayesian Approaches

Abstract

The pursuit of evaluating various goodness of fit indices for testing measurement invariance has been a focus over the past decade (e.g., Chen, 2007; Cheung & Rensvold, 2002; Meade, Johnson, & Braddy, 2008). Nevertheless, the pursuit is continuously aspired due to new contexts in educational, psychological, and behavioral research. For instance, multilevel data are commonly obtained in educational settings. The performance of goodness-of-fit indices for testing factorial invariance in multilevel data has yet to be investigated. The present study expands the previous work (Chen, 2007) to multilevel settings and evaluates the sensitivity of goodness-of-fit indexes to lack of factorial invariance in multilevel data. Ryu’s (2014a) and Ryu and West’s (2009)’s studies can be referred to guide the direction. By using a Monte Carlo simulation experiment, the performance of Δgoodness-of-fit indexes and chi-square difference test in examining factorial invariance in multilevel data is investigated, in which the level-specific model fit evaluation is also compared to the “standard” approach.

Authors

• Leina Zhu, The University of Houston

• Oiman Kwok, Texas A&M University - College Station