Post-Conference Workshop

Finite Mixture Modeling: Statistical Methods for Correlated Data by Jeffrey R. Harring

The post-conference workshop will be dealing with model-based clustering methods, commonly referred to as finite mixture modeling. These methods have been applied to a wide variety of cross-sectional and longitudinal data to account for heterogeneity in population characteristics. This workshop is intended as both a theoretical and practical introduction to finite mixture modeling as it pertains to statistical methods used to model correlated data. It is assumed that workshop attendees are well-grounded in basic statistical methods and have had some exposure to more advanced quantitative methods such as structural equation modeling and multilevel modeling.

Although finite mixture modeling has proven advantageous across many disciplines, the examples used in the workshop draw primarily from social science research, including the fields of education and psychology. Models will be presented in several formats — path diagrams, equations, and software syntax. Data, annotated Mplus script files along with annotated output for all of the examples will be provided. It is not required to bring Mplus on a laptop to attend the workshop — participants will still benefit from a comprehensive set of slides, input files and data so that they could practice running workshop examples on their own. Participants who have Mplus base package + combination add-on will be able to run the examples in real time during the workshop.

The post-conference workshop will be broken into approximately 4-90 minute sessions: 2 in the morning and 2 in the afternoon with lunch in between.