The Good, The Bad, and The Ugly: What We Know Today About Latent Class Analysis with Distal Outcomes
Spring 2015 Emerging Scholars Series
Latent class analysis (LCA) is an increasingly popular statistical tool among social and behavioral scientists interested in explaining population heterogeneity by identifying underlying subgroups of individuals. The subgroups (classes) are comprised of individuals who are similar in their responses to a set of observed variables, with latent class membership inferred from responses to those variables. In many studies, researchers are interested in understanding which characteristics predict latent class membership. For example, do adolescents’ friendship goals (i.e., a risk factor) predict substance use patterns (i.e., a latent class variable)? The mathematical model for predicting class membership from a covariate is well understood. However, determining the association between a latent class predictor and distal outcome (LCA with distal outcomes) presents a more difficult methodological problem. Solving this problem is currently a “hot topic” in the methodological literature.
This keynote presentation provides a scientific summary of the three state-of-the-art approaches to LCA with distal outcomes. Each approach has been shown to work well under certain conditions in recently published simulation studies. However, there has not yet been a comprehensive overview of the approaches and their assumptions or an integration of “take-home messages” across simulation studies. In order to lay the foundation for recent advances in LCA with distal outcomes, this presentation will (1) describe the three approaches to it, (2) clarify the assumptions of each approach, (3) integrate “take-home messages” of published simulation studies, and (4) summarize available software options.