Description

Latent class modeling refers to a group of techniques for identifying unobservable, or latent, subgroups within a population. Researchers have developed and expanded methods like latent class analysis (LCA) and latent transition analysis (LTA) over the last two decades. Our current research focuses on expanding methods to include latent class variables in larger models of complex developmental processes. Latent class analysis (LCA) identifies unobservable subgroups within a population. We work to expand LCA models to allow scientists to better understand the impact of exposure to patterns of multiple risks, as well as the antecedents and consequences of complex behaviors, so that interventions can be tailored to target the subgroups that will benefit most. Latent transition analysis (LTA) is a related method that allows scientists to estimate movement between subgroups over time.

LCA example

LCA Introductory Example: Profiles of Teen Sex and Drug Use

In this example, LCA identifies five subgroups of teenagers based on their substance use and sexual behaviors. The latent variable “youth risk behavior” is measured by the observed variables “sex,” “drinking,” “smoking,” and “other drugs.” This analysis allows us to identify complex behavior patterns and variables that predict high-risk behavior patterns, as well as identify subgroups of youth who are at-risk for negative health consequences. With this information, scientists can develop interventions that target individuals with the greatest need.

Read more here.

Resources

There are a variety of resources available to help you learn more about LCA. See our Resources page for the following:

Static Models

LCA: LCA with a covariate (1-step approach)

Description This code fits a 4-class, latent-class model for marijuana use and attitudes using a model-based approach (1-step approach). It includes a covariate for grades in the model. Software Downloads Latent Gold Mplus SAS Stata Exercise Exercise 5 This exercise asks you to use a model-based approach (1-step approach) to add a covariate for grades to a 4-class model for marijuana use and attitudes that uses 7 binary indicators of the latent class variable. You have to carefully consider what latent class to use as the reference class in the multinomial logistic regression. You may wish to standardize the grades...

LCA: LCA with a covariate and a grouping variable (1-step approach)

Description This code fits a 4-class, latent-class model for marijuana use and attitudes using a model-based approach (1-step approach). It includes a covariate for grades and a grouping variable for year in the model. Software Downloads Latent Gold SAS Exercise Exercise 5 This exercise asks you to use a model-based approach (1-step approach) to add a covariate for grades and a grouping variable for year to a 4-class model for marijuana use and attitudes that uses 7 binary indicators of the latent class variable. You have to carefully consider what latent class to use as the reference class in the...

LCA: LCA with a grouping variable and without measurement variance

Description This code fits a 4-class, latent-class model for marijuana use and attitudes using 7 binary indicators of the latent class variable. It includes a grouping variable for year, and observations came from 3 different years. Measurement invariance across groups is not imposed resulting in an unrestricted latent class model with multiple groups. Software Downloads Latent Gold Mplus SAS Stata Exercise Exercise 4 This exercise asks you to add a grouping variable for year to a 4-class model for marijuana use and attitudes that uses 7 binary indicators of the latent class variable. It asks you to fit a model...

LPA: Baseline LPA with all continuous indicators and a covariate

Description This code fits a baseline, latent-profile model for the “Big 5” personality traits using 5 continuous indicators of the latent class variable and biological sex as a covariate. Software Downloads Latent Gold Mplus Exercise Exercise 6 This exercise asks you to select and interpret a latent profile model for the “Big 5” personality traits using 5 continuous indicators of the latent class variable as well as add a covariate for biological sex. Then, it asks you to interpret all parameters in the model. Note that, by default in most software packages, the variances of the indicators are restricted to...

LPA: Baseline LPA with all continuous indicators and a grouping variable with measurement invariance

Description This code fits a baseline, latent-profile model for the “Big 5” personality traits using 5 continuous indicators of the latent class variable and biological sex as the grouping variable. It also imposes measurement invariance across the groups. Software Downloads Latent Gold Mplus Exercise Exercise 6 This exercise asks you to select and interpret a latent profile model for the “Big 5” personality traits using 5 continuous indicators of the latent class variable as well as add a grouping variable for biological sex. Then, it asks you to interpret all parameters in the model. Please be sure to impose measurement...

LPA: Baseline LPA with continuous and categorical indicators (mixed indicator model)

Description This code fits a mixed indicator latent-profile model (using both continuous and categorical indicators) to identify family subgroups that conform to risk factors associated with adolescent antisocial behavior. This code corresponds to the research paper titled “Constellations of Family Risk for Long-Term Adolescent Antisocial Behavior” published in Journal of Family Psychology in 2020. The paper can be found here: https://pmc.ncbi.nlm.nih.gov/articles/PMC7375013/ Software Downloads Mplus Model Features Model Category Your Content Goes Here Model Type Your Content Goes Here Indicator Type Your Content Goes Here Available Software Your Content Goes Here Measurement Invariance Your Content Goes Here Approach to Outcomes Your...

Multilevel LPA: Baseline two-level LPA with classes at level 1 and level 2

Description This code fits a 2-level latent-profile model using a “non-parametric approach” to identify mother-father-adolescent relationship structures and dynamics on a daily basis. This code corresponds to the research paper titled “Triadic Family Structures and Their Day-to-Day Dynamics From an Adolescent Perspective: A Multilevel Latent Profile Analysis” published in Fam Process in 2022. Note that there is an important difference between the code available here and the exact code used to fit the model in the paper: in the code available here the measurement model is freely estimated, whereas in the paper the measurement model was fixed. The paper can...

Let’s stay in touch.

We are in this together. Receive an email whenever a new model or resource is added to the Knowledge Base.