Abstract
Criteria for evaluating structural equation models with latent variables are defined, critiqued, and illustrated. An overall program for model evaluation is proposed based upon an interpretation of converging and diverging evidence. Model assessment is considered to be a complex process mixing statistical criteria with philosophical, historical, and theoretical elements. Inevitably, the process entails some attempt at a reconcilation between so-called objective and subjective norms.
Get full access to this article
View all access options for this article.
