### Overview

PROC LCA and PROC LTA are SAS procedures for latent class analysis (LCA) and latent transition analysis (LTA) developed by the Methodology Center. These straightforward procedures make it possible to pre-process data, fit a variety of latent class and latent transition models, and post-process the results without leaving the SAS environment. Compatible with SAS version 9.1 or higher. PROC LCA is intended for individual installations and is not tested for server installations of SAS. Features include

- simple model specification,
- multiple-groups LCA and LTA,
- LCA and LTA with covariates (prediction of latent class membership and transitions),
- baseline-category multinomial logit model or binary logit model for prediction,
- posterior probabilities saved to SAS data file,
- parameter estimates saved to SAS data file,
- optional Bayesian stabilizing prior to handle sparseness issues in estimation, and
- accounts for sampling weights and clusters.
- Ability to assess identification of models with covariates via multiple random starts.
- Indication of which latent class is the best match for each individual.
- Option to generate 20 random draws for each individual’s class membership based on posterior probabilities.

Read about latent class analysis or latent transition analysis.

### Recommended Citations

PROC LCA & PROC LTA (Version 1.3.0) [Software]. (2013). University Park: The Methodology Center, Penn State. Retrieved from http://methodology.psu.edu

Lanza, S. T., Dziak, J. J., Huang, L., Wagner, A., & Collins, L. M. (2013). *PROC LCA & PROC LTA users' guide *(Version 1.3.0)*. *University Park: The Methodology Center, Penn State. Retrieved from http://methodology.psu.edu

### PROC LCA for Latent Class Analysis

In its simplest form, PROC LCA allows the user to fit a latent class model by specifying a SAS data set, the number of latent classes, the items measuring the latent variable, and the number of response categories for each item. Multiple-groups LCA can be run using the GROUPS statement; users can examine measurement invariance across groups by adding the MEASUREMENT statement. Additional parameter restrictions can be provided in a SAS data file.

Continuous and categorical covariates can be included in the COVARIATES statement in order to examine the relation between each covariate and the probability of latent class membership. Prediction can be modeled using a baseline-category multinomial logit model or a binary logit model.

A Bayesian stabilizing prior can be invoked when sparseness is an issue for parameter estimation. Random starting values can be generated by the program, or the user can provide starting values in a SAS data file.

An empirical demonstration of PROC LCA appeared in *Structural Equation Modeling*:

Lanza, S. T., Collins, L. M., Lemmon, D. R., & Schafer, J. L. (2007). PROC LCA: A SAS procedure for latent class analysis. *Structural Equation Modeling, 14*(4), 671-694. PMCID: PMC2785099 View article

### PROC LTA for Latent Transition Analysis

LTA is a longitudinal extension of LCA used to estimate latent status membership probabilities at Time 1, along with probabilities of transitions in latent status membership over time. The transition probabilities characterize development over time based on longitudinal data.

In its simplest form, PROC LTA allows the user to fit a latent transition model by specifying a SAS data set, the number of latent statuses, the number of times, the items measuring the latent variable, and the number of response categories for each item. Multiple-groups LTA can be run using the GROUPS statement, and users can examine measurement invariance across groups or times by adding the MEASUREMENT statement. Additional parameter restrictions can be provided in a SAS data file.

Covariates can be included in the model to examine the relation between each covariate and the probability of latent status membership and/or transition probabilities. Prediction can be modeled using a baseline-category multinomial logit model or a binary logit model, adding to the set of questions that can be addressed in LTA. A Bayesian stabilizing prior can be invoked when sparseness is an issue for parameter estimation.

An empirical demonstration of PROC LTA appeared in *Developmental Psychology*:

Lanza, S. T., & Collins, L. M. (2008). A new SAS procedure for latent transition analysis: Transitions in dating and sexual behavior. *Developmental Psychology, 42*(2) 446-456. PMCID: PMC2846549 View arcticle

EXAMPLE OF THE SAS PROCEDURE FOR LATENT TRANSITION ANALYSIS

PROC LTA DATA=SEX OUTPOST=SEX_POST;

TITLE1 'Time 1 substance use predicting Time 1 risky sex, by gender';

TITLE2 'Measurement invariance across times and groups';

TITLE3 'Posterior probabilities saved to SAS data file';

NSTATUS 5;

NTIMES 3;

ITEMS date_98 sex_98 partners_98 exposed_98 date_99 sex_99 partners_99 exposed_99 date_00 sex_00 partners_00 exposed_00;

CATEGORIES 3 2 3 2;

GROUPS gender;

GROUPNAMES male female;

MEASUREMENT TIMES GROUPS;

COVARIATES1 cig_98 drunk_98 marij_98;

REFERENCE1 1;

SEED 409621;

RUN;