Discovering SEM in R
What this Project Is
What this Project Isn’t
1
Introduction to confirmatory factor analysis
1.1
The principal component factor analysis approach
1.2
Alpha reliability for our nine-item scale
1.3
Generating a factor score rather than a mean of summative score
1.4
What can CFA add?
1.5
Fitting a CFA model
1.6
Interpreting and presenting CFA results
1.7
Assessing goodness of fit
1.7.1
Final model and estimating scale reliability
1.8
A two-factor model
1.8.1
Evaluating the depression dimension
1.8.2
Estimating a two-factor model
2
Using structural equation modeling for path models
2.1
A substantive example of a path model
2.2
Estimating a model with correlated residuals
2.2.1
Strengthening our path model and adding covariates
2.3
Auxiliary Variables
2.4
Testing equality of coefficients
2.5
A cross-lagged panel design
2.6
Moderation
2.7
Nonrecursive models
3
Structural equation modeling
3.0.1
The classic example of a structural equation model
3.0.2
Fitting a full structural equation model
3.0.3
Modifying our model
3.0.4
Indirect effects
3.1
Equality constraints
3.2
Programming constraints
3.3
Structural model with formative indicators
3.3.1
Identification and estimation of a composite latent variable
3.3.2
Multiple indicators, multiple causes model
4
Latent growth curves
4.1
Discovering growth curves
4.2
A simple growth curve model
4.3
An example of a linear latent growth curve
4.3.1
Fitting a latent growth curve model
4.3.2
Adding correlated adjacent error terms
4.3.3
Adding a quadratic latent slope growth factor
4.4
How can we add time-invariant covariates to our model?
5
Group Comparisons
Published with bookdown
Discovering Structural Equation Modeling Using
Stata
R, the Tidyverse, and Lavaan
Chapter 5
Group Comparisons
In progress.