Mplus Lab Sessions

Education 231E, Spring 2004

T.A.: Karen Nylund

Email: knylund@ucla.edu

This page will be updated throughout the quarter, following the content of the course.

Week 1: Introduction to Mplus using Reading Data

• Summary of Mplus Language
  
• Type=Basic Missing output
• Graphics: Summary of histograms and scatter plots.

Week 2: EFA and CFA

• Example Description
• Example Correlation Matrix
• Example Covariance Matrix
• Mplus EFA Output
• Mplus EFA within CFA (orthogonal) Output
• Mplus EFA within CFA (oblique) Output
• Mplus Simple CFA Output

Holzinger-Swineford Data:

• Holzinger-Swineford Correlation Matrix
• EFA with all 19 variables
• EFA with 15 variables
• EFA with 13 variables

Week 3: Measurement Invariance and Population Heterogeneity

Topics covered: Forms of measurement invariance and population heterogeneity, CFA with covariates vs. multiple group analysis, H&S example for CFA with covariates.

• H & S Data set
• Mplus Simple CFA output
• Mplus CFA with indirect effects output
• Mplus CFA with indirect and direct effects output
• Mplus CFA, 7th grade only, output
• Mplus CFA, 8th grade only, output
• Mplus Multiple Group, Complete non-invariance output
• Mplus Multiple Group, Factor loading invariance output
• Mplus Multiple Group, Factor loading and intercept invariance output
• Mplus Multiple Group, Factor loading and intercept partial invariance output
• Mplus Multiple Group, Residual variance invariance output
• Mplus Multiple Group, Factor variance invariance output
• Mplus Multiple Group, Factor mean invariance output

Week 4: Latent Variable Growth Modeling

Topics covered: Basic latent variable linear growth models, interpretation, nonlinear growth models, piecewise growth models, growth models with time-invariant covariates.

• Data for Example
• Mplus basic analysis output
• Mplus latent variable linear growth model output
• Mplus LV nonlinear growth model output
• Mplus LV alternate nonlinear growth model output
• Mplus LV nonlinear GM with freed residual covariance between adjacent time points output
• Mplus LV nonlinear GM with time-invariant covariate output
• Mplus LV nonlinear piecewise GM output
• Mplus LV nonlinear piecewise GM with time-invariant covariate output
• Mplus LV nonlinear piecewise reduced GM with time-invariant covariate output

Week 5: Latent Variable Growth Modeling Continued; Multiple Group Analysis

• Data for Example
• Mplus basic analysis output for females
• Mplus latent variable linear growth model output for females
• Mplus LV GM output for females, grade 7-10 Correlated Residuals using Pwith
• Mplus LV nonlinear GM output for females; fixed time scores
• Mplus basic analysis output for males
• Mplus latent variable linear growth model output for males
• Mplus LV GM output for males, grade 7-10 Correlated Residuals using Pwith
• Mplus LV nonlinear GM output for males; fixed time scores
• Mplus LV multiple group GM output for males and females; complete non-invariance across groups
• Mplus LV multiple group GM output for males and females; invariance of factor means and variance
• Mplus LV multiple group GM output for males and females; invariance of intercept mean and variance
• Mplus LV multiple group GM output for males and females; Non-linear growth, invariance of intercept mean and variance

Week 6: Parallel Process Growth Modeling using LSAY

• LSAY Data
• Mplus latent variable linear growth model for math
• Mplus latent variable linear growth model for math with freed time scores
• Mplus latent variable linear growth model for math with fixed time scores
• Mplus latent variable linear growth model for ?Enjoy Math?
• Mplus latent variable linear growth model for ?Enjoy Math? with freed time scores
• Mplus latent variable linear growth model for ?Enjoy Math? with different freed time scores
• Mplus Parallel Process Growth model: Math7-10 and Enj7-10
• Mplus Parallel Process Non-Linear Growth models
• Mplus Parallel Process Non-Linear Growth models with regression among random effects
• Mplus Parallel Process Non-Linear Growth models with regression among random effects and correlation among residual variances
• Mplus Parallel Process Non-Linear Growth models with more regression among random effects and correlation among residual variances

Week 7: LCA and CFA with Categorical Indicators 

• Rindskopf and Rindskopf Data
• Mplus 1-class LCA Model output
• Mplus 2-class LCA Model output
• Mplus 3-class LCA Model output
• Mplus Restricted 3-class LCA Model output
• Mplus 1-factor CFA model output
  
  

Week 8: LCA Continued and LCA with Covariates

• ASB Data Set
• Mplus ASB type=Basic with 17 items
• Mplus 4-class LCA Output
• Mplus Alternative Specification for the 4-class Model Output
• Mplus 5-class LCA Output
• Mplus 4-class LCA Output with Covariates

Week 9: Growth Mixture Modeling with Simulated Data from Mplus Version 3 User?s Guide Example 8.1

• Simulated Data Set
• Mplus Type=Basic for Simulated Data
• Mplus 1-Class GMM Solution
• Mplus 2-Class GMM Solution
• Mplus 2-Class GMM Solution with class specific statement
• Mplus 3-Class GMM Solution

Week 10: More Growth Mixture Modeling (GMM) with Simulated Data from Mplus Version 3 User?s Guide Example 8.1

• Eight Steps for GMM Analysis in Mplus Version 3
• Step A1: GMM without covariates
• Step A2: GMM without covariates, intercept variance non-invariant
• Step A2: GMM without covariates, slope variance non-invariant
• Step B1: GMM with covariate, invariant covariate effect on slope and intercept
• Step B2: GMM with covariate, non-invariant effect of slope and intercept

Week 11: Multilevel Analysis -- Multilevel Factor Analysis with Simulated Data from Mplus Version 3 User?s Guide Example 9.8

• Six Steps for Two-Level SEM Analysis in Mplus Version 3
• Example 9.8 Simulated Data Set
• Step 0: Mplus Type=Basic run: estimate within and between matrices, estimate intra class correlations (saving sigmaw)
• Step 0: Mplus Type=Basic run: estimate within and between matrices, estimate intra class correlations (saving sigmab)
• Step 1A: SEM ignoring clustering
• Step 1B: SEM with type=complex
• Step 1C: EFA ignoring clustering
• Step 2: CFA using the pooled-within matrix
• Step 4: EFA on the between correlation matrix
• Step 5: Two-Level SEM

Click here to access the previous discussion section page (Ed231E, Spring 2003).

Back to course webpage

Updated 06-10-2004