Robustness of Structural Means Model to Non-Normality: A Simulation Study

• In Event: 30.074 - Issues in Parameter Estimation and Model Comparison

Abstract

Maximum likelihood (ML) is one of the most commonly used estimation methods in structural equation modeling (SEM) because ML has desirable asymptotic properties (i.e., large sample properties), for example, unbiasedness, if the assumption of multivariate normality of measured variables is satisfied.

One special and powerful model in SEM is the structural means model (SMM) which provides researchers with capacity of testing hypotheses about the means of the latent variables by adding a mean structure to SEM's basic covariance structure.

The purpose of this study is to investigate the impact of the extent to which multivariate normality assumption is violated on the robustness of parameter estimates obtained by ML in SMMs.

Author

• Meng Qiu, Avar Consulting, Inc.