Croon's Bias-Corrected Estimator Small- to Moderate-Scale Multilevel Structural Equation Models

• In Event: Methodological Challenges in Multilevel Modeling

Abstract

Maximum likelihood (ML) estimation of multilevel structural equation model (MLSEM) parameters is a preferred approach to operationalize theories involving latent variables in multilevel settings. Although this estimator has many desirable properties, a major limitation is that it often fails to converge and can incur significant bias when implemented in small-to-moderate multilevel sample studies (<100 schools). In this study, we develop an alternative two-stage estimator based on Croon’s framework for MLSEMs and probe the degree to which the estimator can retain these advantages with small to moderate multilevel samples. The estimator emerges as an alternative or complementary estimator ML because it often outperforms ML in convergence, bias, error variance, and power. The estimator is implemented in the BLINDED R package.

Authors

• Benjamin Kelcey, University of Cincinnati

• Kyle T. Cox, University of North Carolina - Charlotte

• Nianbo Dong, University of North Carolina at Chapel Hill