EN
In latent class analysis it is assumed that each observation comes from one of a number of classes (groups) and models each with its own probability distribution. When longitudinal data are to be analyzed, the research questions concern some form of change over time. Latent transition analysis (LTA) also known as latent Markov model, is a variation of the latent class model that is designed to model not only the prevalence of latent class membership, but the incidence of transitions over time in latent class membership. We used latent class analysis for grouping and detecting inhomogeneities of Polish attitude to saving money. We analyzed data collected as part of the Social Diagnosis, based on panel research using depmixS4 package of R.