A Structural After Measurement Approach for Partially Nested Structural Equation Models With Latent Interactions

• In Event: Saturday Roundtable Session 11:25 am
  In Roundtable Session: Issues and Developments in Structural Equation Modeling (Table 10)

Abstract

Conventional applications and developments of multilevel structural equation modeling (MLSEM) have focused on fully nested structures such as students within a teacher. More recent work has demonstrated the prevalence of complex structures such as partial nesting. The purpose of this study is to develop and investigate a structure after measurement estimator that utilizes a Croon’ based correction (SAM-Croon’s) for the analysis of latent interactions in partially nested structural equation models with small to moderate sample sizes. We conducted three Monte Carlo simulation studies with increasingly complex models and compared bias, efficiency, and convergence rates of several estimation approaches including maximum likelihood and Bayesian estimators. Results support SAM-Croon’s as an effective estimator for partially nested SEMs with latent interactions.

Authors

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

• Benjamin Kelcey, University of Cincinnati