This Structural Equation Modeling using SAS training course introduces the experienced statistical analyst to structural equation modeling (SEM) and the new PATH language in the CALIS procedure in SAS/STAT software. The course also introduces the PATHDIAGRAM statement in the CALIS procedure, which automatically draws path diagrams based on fitted models.
Structural equation modeling is a statistical technique that combines elements of traditional multivariate models, such as regression analysis, factor analysis, and simultaneous equation modeling. These models are often represented as matrices, equations, and/or path diagrams and can explicitly account for uncertainty in observed variables and for estimation bias due to measurement error. Competing models can be compared to one another, providing information about the complex drivers of the outcome variables of interest. Many applications of SEM can be found in the social, economic, and behavioral sciences, where measurement error and uncertain causal conditions are commonly encountered.
By attending Structural Equation Modeling using SAS workshop, delegates will learn to:
- Explain a regression model in terms of a structural equation model
- Compare results from the reg and calis procedures
- Produce a path diagram of your model results
- Customize a path diagram
- Specify models and evaluate model fit in the calis procedure using the path input style
- Specify mediation models and test for complete and partial mediation
- Perform complex path analysis
- Perform confirmatory factor analysis
- Specify general latent variable models
- Perform robust estimation for data with outliers
- Perform full-information maximum likelihood estimation for incomplete data
- Perform honest assessment to validate models.
- Strong background in regression modeling
- Familiar with factor analysis
- Familiar with Anova and Regression.
The Structural Equation Modeling using SAS class is ideal for:
- Most appropriately, social, behavioral, economic, and health researchers interested in fitting complex path models and latent variable models