Comparisons of Reliability Estimates for Correcting Measurement Errors of the Exogenous Composites in Interaction Models

• In Event: 34.094 - Fairmont Royal York Roundtable Session Six
  In Roundtable Session: 34.094-4 - Finer Details in Modeling Latent Interactions, Moderated Mediation, and Multicategory Indicators

Abstract

Social and behavioral research often examines interaction effects. Many studies involved testing interaction effects have relied on creating observed composite scores based on multiple observed items or variables, which lead to biased estimates. Previous simulation has shown that the latent moderated structural equation (LMS) and reliability-adjusted product indicator (RAPI) methods have been shown to yield accurate interaction estimates by adjusting for the exogenous composites’ measurement errors. Conditions with congeneric equivalent variables have not been considered. Additionally, what reliability coefficients should be utilized is unknown. In the present study, three reliability estimates (Cronbach’s alpha, omega total, and greatest lower bound) were applied to adjust for the exogenous composites’ measurement errors under the RAPI and the LMS methods while estimating interaction effects.

Authors

• Yu-Yu Hsiao, University of New Mexico

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

• Hok Chio (Mark) Lai, University of Southern California