From Factor analysis to SEM!

Path Analysis: Regression and lm as a subset of SEM

Introduces

• The idea of causality, testing (and, critically, mxCompare()-ing) ideas. And use real tasks, such as handling a ordinal outcomes as thresholded normals.
• Tutorial: Specific SEM models
• Curb your enthusiasm: By this week you should understand what boxes, circles, triangles, diamonds, and straight and curved arrows are.

Regression and lm as a subset of SEM

This week we will leverage the modelling you already know by building the equivalent of lm(y ~ x + z) in umx package.

This builds on two ideas you already understand: Factor analysis and linear models/ANOVA.</li>

Regression

The relationship of a single dependent variable (Y) to one or more independent variables (X₁, X₂ etc.) Y = 𝛽₀ + (𝛽₁ × X₁) + (𝛽₂ × X₂) + (𝛽₃ × X₃)

Regression and lm as a subset of SEM