Latent Class Analysis
To see a sample of the course materials, click here.
Latent class analysis (LCA) is an intuitive and rigorous tool for uncovering hidden subgroups in a population. It can be viewed as a special kind of structural equation modeling in which the latent variables are categorical rather than continuous. This seminar will give you the theoretical background and applied skills to address interesting research questions using LCA. You will also be introduced to latent transition analysis (LTA), a longitudinal extension of LCA.
Other topics include model identification, model selection, model interpretation, multiple-groups LCA, measurement invariance across groups, and LCA with covariates and distal outcomes. The seminar will combine lectures, software demonstrations, computer exercises, and discussion. There will be opportunities to discuss how LCA and LTA can be applied in your own research.
Starting December 10, we are offering this seminar as a 3-day synchronous*, remote workshop for the first time. Each day will consist of a 4-hour live lecture held via the free video-conferencing software Zoom. You are encouraged to join the lecture live, but will have the opportunity to view the recorded session later in the day if you are unable to attend at the scheduled time.
Each lecture session will conclude with a hands-on exercise reviewing the content covered, to be completed on your own. An additional session will be held Thursday and Friday afternoons as an “office hour”, where you can review the exercise results with the instructor and ask any questions.
*We understand that scheduling is difficult during this unpredictable time. If you prefer, you may take all or part of the course asynchronously. The video recordings will be made available within 24 hours of each session and will be accessible for one week after the seminar, meaning that you will get all of the class content and discussions even if you cannot participate synchronously.
This remote seminar is held via Zoom, a free video conferencing application. Instructions for joining a session via Zoom are available here. Before the seminar begins, you will receive an email with the meeting code and password you must use to join.
All examples and exercises will use SAS and the free add-on procedures PROC LCA and PROC LTA developed by Dr. Lanza and her colleagues. Coverage will include both basic and advanced features of PROC LCA, but only basic features of PROC LTA. Previous experience with SAS is highly desirable. If you prefer Mplus, Stata, or Latent Gold, you can get equivalent program code for these packages by request.
To complete the exercises, you will need your own computer with a recent version of SAS and the free add-on procedures, PROC LCA and PROC LTA, installed. Both PROCs can be downloaded at https://www.methodology.psu.edu/downloads/proclcalta/.
Note: PROC LCA and PROC LTA do not function with the free University Edition of SAS.
WHO SHOULD Register?
If you plan to analyze cross-sectional or longitudinal data and believe that there are meaningful subgroups of individuals characterized by the intersection of multiple characteristics, this seminar is for you. These subgroups might be defined by patterns of problem behavior, risk exposure, product preference, political alignment, and many other hard-to-measure constructs.
You should have a good working knowledge of the principles and practice of multiple regression; familiarity with logistic regression is helpful.
Day 1: Introduction to latent class analysis (LCA)
- Conceptual introduction to LCA
- An example: latent classes of adolescent drinking behavior
- Parameters estimated in LCA
- Technical considerations: model identification, model selection
- Software options
Exercises on fitting a baseline LCA, model identification, and model selection
Day 2: Adding features to “baseline” models
- Including a grouping variable
- Review of logistic regression
- Predicting latent class membership
- Predicting a distal outcome
- Conceptual introduction to latent profile analysis (LPA)
Exercises on adding a grouping variable, adding predictors, and fitting a baseline LPA
Day 3: Introduction to latent transition analysis (LTA)
- Conceptual introduction to LTA
- An example: latent classes of dating and sexual risk behavior
- Technical details
- Including a grouping variable
- Predicting transitions
- Connections between LCA, LTA, and growth mixture modeling
Exercise on fitting a baseline LTA, adding a grouping variable, and adding predictors
“This has been an excellent introduction to latent class analysis. Drs. Lanza and Bray are extremely knowledgeable and effective teachers. They appropriately covered basic concepts and linked them to more complex statistical approaches. I also benefited from the hands-on learning by practicing analyses with the statistical software. I look forward to applying these methods to my research.”
Kaylee Crockett, University of Alabama at Birmingham
“This course provided me with ideas about my future research. The instructors created a very positive learning environment.”
Hannah Dong, B.C. Centre for Excellence in HIV/AIDS
“The instructors are extremely knowledgeable and also great teachers. Whether you’ve already tried using LCA or are new to it, I highly recommend taking this course.”
Talha Ali, Yale University
“This class will give you a great intro to use these techniques.”
Nick Huntington, Brandeis University
“Having only read a few of the articles on SAS LCA prior to coming to this workshop and with no prior experience with LCA, I found the speed, breadth, and depth to be a perfect dive into this type of analysis. The instructors Stephanie and Bethany were extremely engaging and were patient and informative in answering students’ questions. I learned a ton from listening to their troubleshooting of students during exercises and breaks and from their answers to class questions. What a fantastic course! Well worth the time and money. I look forward to jumping into this new analytic technique!”
Krissy Moehling, University of Pittsburgh
“This course was a great overview of LCA with a focus on the foundations. The instructors were so knowledgeable and helpful and provided a lot of specialized attention.”
Annette Ponnock, Yale University
“As a graduate student, I was very intimidated about taking this workshop. But it really is tailored to the needs of the participants, and I never felt as if I were asking a stupid question. I had read the Collins and Lanza book, but this was supplemental and incredibly useful.”
Lauren Aaron, University of New Orleans
“I was initially concerned that this workshop would be totally over my head, but the instructors made the material very digestible. They were very friendly and engaging and I appreciated their openness to questions. They also pointed us to several helpful resources.”
Samantha Stevens, Pennsylvania State University
“The teachers are passionate about the topic, and they explain the concepts in easy-to-understand ways. It’s an absolute nerd-fest in the best possible way. If you’re considering LCA, LPA, or LTA, this is the class to take!”
Will Beckham, Johns Hopkins University
“The course provides a solid background in LCA while also providing specific code in multiple packages for running analyses.”
Michael Strambler, Yale University
“Excellent instructors and instruction!”
Leah Varga, George Washington University & D.C. Department of Health