Structural Equation Modeling (2013)

10 Commandments of SEM (Thompson, 2000)

1. No small samples.
2. Analyze covariance, not correlation matrices.
3. Simpler models are better.
4. Verify distributional assumptions (multivariate normality)
5. Consider theoretical and practical significance, not just statistical significance.
6. Report multiple fit statistics.
7. Use two-step modeling for structural regression models. (Test the measurement model then the structural model.)
8. Consider theoretically plausible alternative models.
9. Respecify rationally. (Test the respecified model on new or split-halves data.)
10. Acknowledge equivalent models.

Checklist for SEM (Tabachnick & Fidell, 2007: Table 14.24)

• Issues
  • Sample size and missing data
  • Normality of sampling distributions
  • Outliers
  • Linearity
  • Adequacy of covariances
  • Identification
  • Path diagram---hypothesized model
  • Estimation method
• Major analyses
  • Assessment of fit
    • Residuals
    • Model chi square
    • Fit indices
  • Significance of specific parameters
  • Variance in a variable accounted for by a factor
• Additional analyses
  • Lagrange Multiplier test (not endorsed in my SEM course!)
    • Tests of specific parameters
    • Addition of parameters to improve fit
  • Wald test for dropping parameters
  • Correlation between hypothesized and final model or cross-validate model
  • Diagram---final model

Course References

Kline, R. B. (2010). Principles and practice of structural equation modeling (3rd ed.). New York: Guilford Press.

Tabachnick, B. G., & Fidell, L. S. (2007). Using multivariate statistics (5th ed.). Boston: Pearson.

Thompson, B. (2000). Ten commandments of structural equation modeling. In L. Grimm & P. Yarnell (Eds.), Reading and understanding more multivariate statistics (pp. 261-284). Washington, DC: American Psychological Association.