MM4DEBR (Cohort 2)
MM4DBER Training Materials
Mixture Modeling for Discipline-Based Education Researchers (MM4DBER) is an NSF-funded training grant to support STEM Education scholars in integrating mixture modeling into their research.
“Parking Lot” document with questions and answers HERE
Office hour appointment link (Adam): HERE
Day 1 Training (August 23, 2024): Introduction to the Latent Class Analysis (LCA) Model
Learning Outcomes:
- Recognize the LCA statistical model
- Interpret LCA model parameters
- Evaluate conditional item probability plots
- Have a basic understanding of the Mplus code and output for an LCA model
Synchronous Activity:
- Slide deck here
Asynchronous Activity:
- Activity 1: Run your first LCA model and do the scavenger hunt looking for important parts of the Mplus output. All materials are found here
Answer key: Questions & Answers , Annotated Mplus Output
- Activity 2: Review this introductory paper on LCA and answer the questions found in this guided reading exercise
Day 2 Training (August 30, 2024):
Learning Outcomes:
- Understand the principles of item selection in LCA
- Know the principles of evaluating mixture models
- Understand the statistical tools available to evaluate models
Synchronous Activity:
- Slide deck here
Asynchronous Activity:
Activity 1:
- Watch this code-along video and follow along in Rstudio: Video
- Tutorial handout: Link here
- Download the Github repository here: Intro_to_LCA
Day 3 Training (September 6, 2024):
Learning Outcomes:
- Understand the principles of evaluating mixture models
- Been exposed to the complex enumeration process
- Understand the multi-step process for auxiliary variables
Synchronous Activity:
- Slide deck here
Asynchronous Activity:
Day 4 Training (September 13, 2024):
Learning Outcomes:
- Understand the utility of auxiliary variables in mixture modeling
- Learn about the multi-step approach to auxiliary variables
- Understand how to incorporate covariate and distal outcomes into mixture models
Synchronous Activity:
- Slide deck Here
- Mplus output file example: c4_covariate_only.out
Asynchronous Activity:
Activity 1: Automated ML 3-Step Method
- Watch this code-along video and follow along in Rstudio: Video
- Tutorial handout: Link here
- Download the Github repository here: 3-Step
Activity 2: Interpretation of auxiliary variables
- Watch this code-along video and follow along in Rstudio: Video
- Tutorial handout: Link here
- Download the Github repository here: Interpretation of auxiliary variables
Additional resources: How to use the here() function
Bonus slides & 3-step article:
Day 5 Training (September 20, 2024):
Learning Outcomes:
- Understand what LPA is and how it differs from LCA
- Have exposure to extensions of the LCA and LPA models
- Know more about the year long MM4DBER training program
Synchronous Activity:
Additional Resources (Latent Profile Analysis; LPA):
NOTE: For the LPA diabetes example (from slides) the final model chosen is discussed on page 595 (Masyn, 2013).
- Reading: Masyn, 2013 (book chapter; p.36-52)
- Handout: LPA in R code-along
- Github repository: Intro_to_LPA
- Video link: Cohort 1 - LPA Training Recording (2023)
Helpful Links:
-
Please visit our website to learn more.
-
Visit our GitHub account to access all the MM4DBER materials.
-
Follow us on Twitter!
