poLCA is an R package for estimating latent class and latent class regression models for polytomous (categorical) outcome variables. It uses the expectation-maximization (EM) algorithm and Newton-Raphson methods to find maximum likelihood estimates of model parameters. The basic latent class model is a finite mixture model where component distributions are multi-way cross-classification tables with all variables mutually independent. The latent class regression model extends this by allowing covariates to predict latent class membership. poLCA can handle both dichotomous and polytomous outcomes, and it is the only package that estimates latent class regression models with covariates. It also provides standard error estimates, model selection criteria (AIC, BIC), and goodness-of-fit statistics. The package includes sample data sets and commands for generating simulated data. Users can specify the number of latent classes, covariates, and other model parameters. The package also allows for reordering latent classes and avoiding local maxima by running the model multiple times. The output includes estimated class-conditional response probabilities, mixing proportions, posterior class membership probabilities, and various model fit statistics. poLCA is useful for analyzing multivariate categorical data, such as in political science, social science, and survey research.poLCA is an R package for estimating latent class and latent class regression models for polytomous (categorical) outcome variables. It uses the expectation-maximization (EM) algorithm and Newton-Raphson methods to find maximum likelihood estimates of model parameters. The basic latent class model is a finite mixture model where component distributions are multi-way cross-classification tables with all variables mutually independent. The latent class regression model extends this by allowing covariates to predict latent class membership. poLCA can handle both dichotomous and polytomous outcomes, and it is the only package that estimates latent class regression models with covariates. It also provides standard error estimates, model selection criteria (AIC, BIC), and goodness-of-fit statistics. The package includes sample data sets and commands for generating simulated data. Users can specify the number of latent classes, covariates, and other model parameters. The package also allows for reordering latent classes and avoiding local maxima by running the model multiple times. The output includes estimated class-conditional response probabilities, mixing proportions, posterior class membership probabilities, and various model fit statistics. poLCA is useful for analyzing multivariate categorical data, such as in political science, social science, and survey research.