Latent Class Analysis (LCA) is a statistical method for identifying unmeasured class membership among subjects using categorical and/or continuous observed variables. For example, you may wish to categorize people based on their drinking behaviors (observations) into different types of drinkers (latent classes). This could lead to finding categories such as abstainers, social drinkers, and alcohol abusers. You could try to create models to predict why one falls into particular class memberships (why do people become alcohol abusers), and you can also seek to explore the consequences of such class memberships (does being an alcohol abuser/not abuser predict other variables). You can even combine latent class analysis with other techniques. For example, you might use survival analysis to model time to first use of alcohol and find that latent class analysis identifies a class of long term abstainers and whose survival is modeled separately from non-abstainers. Or if you were using latent growth curve modeling of alchohol use over time, you could apply latent class analysis to the trajectories of alcohol use to identify classes such as abstainers, early drinkers who taper off, and chronic alcohol abusers. LCA can be used in many disciplines such as Health Sciences, Psychology, Education, and the Social Sciences. Examples will be shown using Mplus version 3.
Statistical Methods and Data Analytics
