SI Online course: Introduction to Latent Class Modeling
Next course: October 16 to November 6
Latent class (LC) modeling is a technique for analyzing case level data with the goal of finding and introducing to the model “latent classes,” or segments that characterize similar groups of cases (e.g. customer segments, medical diagnoses, types of survey respondents, etc.) based on categorical or continuous variables or a combination of these. In this course we introduce LC as a probability model and describe various applications using the demo version of Latent GOLD®. Initial lessons focus on model fitting strategies and the interpretation of output. Subsequent lessons consider several advanced topics including identification issues, LC models for longitudinal data, data sparseness, and use of bivariate residuals. Scroll down for more information about this course.
For Commercial attendees the cost is $495 and for Academic participants it is $295. Additional attendees from the same organization receive a 50% discount. To receive this discount, all attendees should register at the same time. This discount is automatically applied to your order.
To register, please select the attendee type.
Course Overview: Latent class (LC) modeling is a technique for analyzing case level data with the goal of finding and introducing to the model “latent classes,” or segments that characterize similar groups of cases (e.g. customer segments, medical diagnoses, types of survey respondents, etc.). Interest in LC models is increasing rapidly. This is due to the growing evidence that they provide better solutions than the more traditional approaches to cluster, factor and regression analysis when the population is not homogeneous. In particular, LC modeling has now become the gold standard for cluster analysis.
In this course we introduce LC as a probability model and describe various applications using the demo version of Latent GOLD®. Initial lessons focus on model fitting strategies and the interpretation of output. Subsequent lessons consider several advanced topics including identification issues, problems caused by data sparseness, use of bivariate residuals and repeated observations. The course also introduces the emerging topics of latent class trees (LCTs) and the proper use of covariates to deal with the potential bias of Differential Item Functioning (DIF).
Course Structure: The course takes place online. Course participants will be given a username and password for access to a private bulletin board that serves as a forum for discussion and interaction with the instructor. The course is divided into three weekly sessions. Attendees typically spend about 5-15 hours on each session. At the beginning of each week, participants receive the relevant material, in addition to answers to exercises from the previous session. All course materials are posted to a dedicated course homepage, which can be accessed via the same username and password. During each session, participants review the course materials and work through exercises using the Latent GOLD® program. The instructor will provide answers to the exercises and to posted questions, but participants may also engage in discussions with other course participants.
Session 1: Introduction to Latent Class Cluster Models
- Basic ideas of latent class analysis
- Interpreting Latent GOLD® output
- Examples from survey analysis
- Modeling Latent Class Trees (NEW TOPIC)
- Including covariates in LC models and dealing with DIF (NEW TOPIC)
- Extension to continuous variables and other scale types
- Boundary and local solution issues; Bayes constants
- Including direct effects to relax the assumption of local independence
- Example with Diabetes data (obtaining scoring equations)
Session 2: Simple LC Regression Models
- Guidelines for estimating LC regression models
- LC Regression Models with Predictors
- LC Regression Models with repeated measurements
- LC Growth Models and other examples (obtaining scoring equations)
Session 3: Introduction to some advanced topics
- Incorporating/ accommodating cases with known class membership
- Latent Markov modeling for longitudinal data analysis
- Multilevel models
- Continuous factors (CFactor) and individual-level parameters
- Controlling for the ‘level’ effect: specifying a random intercept model as an alternative to ‘centering’
- Controlling for the ‘scale’ effect: using scale classes and scale CFactors
- Using previously estimated models to score new cases
Who should sign up for this course: Marketing researchers, biomedical researchers, survey analysts, and anyone who wants to learn the latest tools to analyze data in more depth than allowed by conventional methods, and to identify latent (hidden) segments that underlie survey data, customer, patient or prospect databases, diagnostic, test or other cross-sectional or longitudinal data.
Prerequisite: Participants should have taken at least two courses in statistics, and be familiar with the use of linear regression.
Course Material: No text required — copies of published articles and other material will be made available. All participants will have free access to the demo version of Latent GOLD®, which allows unrestricted analyses of all course datasets. Participants need not license a copy of the Latent GOLD® program. All participants will have free access to the demo version of Latent GOLD®, which allows unrestricted analyses of all course datasets.
Instructor: Dr. Jay Magidson, founder and president of Statistical Innovations Inc.. Dr. Magidson’s clients have included A.C. Nielsen Co., Household Finance Corp., Blue Cross Blue Shield Association, and Pfizer. He taught statistics at Tufts and Boston University and is widely published on the theory and applications of multivariate statistical methods. Dr. Magidson designed SPSS CHAID, SI-CHAID®, GOLDMineR® and CORExpress®, is the co-developer (with Jeroen K. Vermunt) of the Latent GOLD® and Latent GOLD® Choice programs.