Structural Equation Modeling
Part 2. Marketing Research
Published Online: 15 DEC 2010
Copyright © 2011 John Wiley & Sons, Ltd. All rights reserved.
Wiley International Encyclopedia of Marketing
How to Cite
Baumgartner, H. 2010. Structural Equation Modeling. Wiley International Encyclopedia of Marketing. 2.
- Published Online: 15 DEC 2010
Structural equation modeling (SEM) with latent variables is a technique for investigating relationships between latent (unobserved) variables or constructs that are measured by multiple manifest (observed) variables or indicators. We provide an overview of SEM organized around the steps involved in using SEM for theory testing: model specification (graphical model specification, model identification, measurement and latent variable model specification issues, and sample size considerations); preliminary data analysis; model estimation; overall model evaluation (based on the chi-square goodness-of-fit test and alternative fit indices); model modification; and model interpretation. More advanced uses of SEM and recent developments are also briefly described. These include models for multiple populations and models with mean structures, latent variable mixture models, models incorporating nonlinear relationships, multilevel SEM, and latent curve models.
- structural equation modeling;
- latent variable models;
- covariance structure analysis;
- moment structures;
- confirmatory factor analysis;
- path analysis