Seminar for Advanced Data Analysis: Introduction to latent class analysis via Latent GOLD 6.0

Content

This workshop will introduce you to latent class (LC) analysis, an elegant, flexible and conceptually simple model-based clustering tool. Using examples from social and behavioral science research, Prof. Vermunt discusses LC models for response variables that are dichotomous, ordinal, nominal, continuous, or counts, or any combination of these. After presenting the basic LC model, he will introduce some more extended models, such as models with covariates and models that relax the local independence assumption.
Besides the more typical cluster analysis applications, he will also pay attention to LC analysis based on a regression model, a technique that is also known as mixture regression analysis. This method that is similar to multilevel regression analysis and especially useful in longitudinal and experimental research, where the aim is to find groups of individuals with similar change patterns or treatment effects. Such LC models for longitudinal data are often referred to as mixture growth models.
The various types of LC models will be illustrated using applications from survey research, psychological testing/diagnosis, longitudinal studies, and experimental research. During the workshop you will learn how to use of the Latent GOLD software for LC analysis.

Preparation

In preparation for each workshop day, participants are asked to watch weblectures recorded by Professor Vermunt. The link to those videos and more details can be found here. Watching the video’s takes 2-3 hours per workshop day.

Prerequisites

Knowledge of descriptive and inductive statistics (as is usually taught at master’s level in a behavioral science program). Participants should bring a laptop to be able to work with the software Latent GOLD during the workshop.

Instructor

Jeroen K. Vermunt received his PhD degree in social sciences research methods from Tilburg University in the Netherlands in 1996. He is currently a full professor in the Department of Methodology and Statistics at Tilburg University, where he has been on the faculty since 1992. In 2005, he received the Leo Goodman early career award from the methodology section of the American Sociological Association. His research interests include latent class and finite mixture models, IRT modeling, longitudinal and event history data analysis, multilevel analysis, and generalized latent variable modeling. He is the co-developer (with Jay Magidson) of the Latent GOLD software package.