Latent Class Analysis and Finite Mixture Modeling

Finite mixture models, which are a type of latent variable model, express the overall distribution of one or more variables as a mixture of a finite number of component distributions. In direct applications, one assumes that the overall population heterogeneity with respect to a set of manifest variables results from the existence of two or more distinct homogeneous subgroups, or latent classes, of individuals. This chapter presents the prevailing “best practices” for direct applications of basic finite mixture modeling, specifically latent class analysis (LCA) and latent profile analysis (LPA), in terms of model assumptions, specification, estimation, evaluation, selection, and interpretation. In addition, a brief introduction to structural equation mixture modeling in the form of latent class regression is provided as well as a partial overview of the many more advanced mixture models currently in use. The chapter closes with a cautionary note about the limitations and common misuses of latent class models and a look toward promising future developments in mixture modeling.