Introduction to Finite Mixture Modeling, Day 3

The Center for Integrated Latent Variable Research (CILVR)

presents

     INTRODUCTION TO FINITE MIXTURE MODELING

                    APRIL 28-30, 2021

taught by

Jeffrey R. Harring, University of Maryland
 

Short Description

Model-based clustering methods, commonly referred to as finite mixture modeling, have been applied to a wide variety of cross-sectional and longitudinal data to account for heterogeneity in population characteristics. Finite mixture models represent a type of latent variable model that expresses 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 variables is due to the existence of two or more distinct homogeneous subgroups, or latent classes, of individuals. These approaches are often termed “person-centered” analyses in contrast to the “variable-centered” analyses of conventional factor analytic models. This three-day course is intended as both a theoretical and practical introduction to finite mixture modeling as it pertains to statistical methods regularly used in educational, behavioral, and social science research.

We will begin by introducing mixture modeling principles in familiar contexts such as univariate and multivariate distributions and quickly move to more complex modeling environments such as regression, factor analysis, and latent growth modeling. Along the way, we will cover aspects of mixture modeling including model construction and specification, graphical representations of models, estimation, class enumeration, evaluating hypotheses, assessing data-model fit, and model comparisons. In addition, content covered will draw, as time allows, from such topics as: latent class analysis, hidden Markov models, latent transition analysis, multilevel mixture modeling. Although this material is necessarily more complex, it will be presented in an approachable hands-on manner for the applied researcher.

Examples used in presentations draw primarily from social science research, including the fields of education and psychology, and will be accompanied by annotated input and output using the Mplus software package. Throughout the course, participants will be able to do practice exercises using Mplus.

Course Fees