StepMix – Modeling heterogeneous and complex data using generalized mixture models: presentation of the package and its applications in the social sciences

The growing popularity of generalized mixture models is partly explained by the fact that many concepts in the social and behavioral sciences cannot be observed directly, such as the existence of subgroups within a population.
Based on patterns of responses to sets of indicators, mixture models (e.g., latent class/profile analysis) offer an intuitive solution to many questions involving the existence of subgroups within a population.