latent class modeling: uncovering population profiles

These pages are currently under construction and will be based on content from The Methodology Center at Penn State.

Latent class modeling refers to a group of techniques for identifying unobservable, or latent, subgroups within a population. Methods like latent class analysis (LCA) and latent transition analysis (LTA) have been developed and expanded by a variety of researchers over the last two decades. LCA identifies unobservable subgroups within a population based on multiple observed indicators (e.g., multiple self-reported risk factors). Current research in the social, behavioral, and health sciences focuses on expanding methods to include latent class variables in larger models of complex developmental processes. This research allows scientists to better understand, for example, the impact of exposure to patterns of multiple risk factors, as well as the antecedents and consequences of complex behaviors, so that interventions can be tailored to target the subgroups that will benefit most. LTA is a related method that allows scientists to estimate movement between subgroups over time.

introductory examples

latent class analysis (LCA): profiles of teen sex and drug use

latent transition analysis (LTA): changes in teen sexual risk profiles

free software

programming code repository for SAS, Mplus, and Latent Gold

SAS PROC LCA & PROC LTA

SAS macros for PROC LCA

LCA Stata plug-in

resources

recommended reading list

videos and podcasts

teachers’ corner: materials for graduate-level methods courses

latent class analysis FAQ and ask a methodologist

PROC LCA & PROC LTA FAQ

methodological research

latent class analysis with an outcome

latent class moderation

multilevel latent class analysis

latent class analysis and causal inference

applied research

substance use and other addictive behaviors

HIV/AIDS

intervention science

risk and protective factors

contact info

Bethany C. Bray, Ph.D.
bcbray@uic.edu

Stephanie T. Lanza, Ph.D.
slanza@psu.edu

John J. Dziak, Ph.D.
jjd264@psu.edu



Portions of this website and the related scientific research were funded by National Institute on Drug Abuse awards P50 DA039838 and P50 DA010075 to The Methodology Center at Penn State.