Methodological & Technical Research Topics

LCA With a Distal Outcome

The %LCA_Distal macro provides a model-based approach to estimating the association between C and Z.

Latent class membership can be used to predict a distal outcome (an outcome at a later time). One motivating example appears in Lanza & Rhoades (2013), in which the authors identified six adolescent risk profiles characterized by exposure to peer, household, and neighborhood risk. This study showed how one can examine differential treatment effects across latent classes; the authors developed an Excel calculator that provides a model-based method for predicting a binary distal outcome from the latent class variable.

 

This work then led to the Methodology Center's more comprehensive model-based approach, packaged as a SAS macro, to estimating the effect of latent class membership on a distal outcome. See Lanza, Tan, & Bray (2013). This work led to the development of the %LCA_Distal SAS macro for estimating the association between membership in a latent class and a binary, continuous, count or categorical distal outcome.

 

A related research topic is the development of improved methods for classify-analyze approaches, such as multiple pseudo-class draws.

 

References

Lanza, S. T. & Rhoades, B. L. (2013). Latent class analysis: An alternative perspective on subgroup analysis in prevention and treatment. Prevention Science, 14, 157-168. PMCID: PMC3173585

Lanza, S. T., Tan, X., & Bray, B. C. (2013). Latent class analysis with distal outcomes: A flexible model-based approach. Structural Equation Modeling: A Multidisciplinary Journal. Advance online publication. doi: 10.1080/10705511.2013.742377 PMC Journal- In process


 

Latent Class and Latent Transition Analysis

Comprehensive Book for Applied Researchers

Linda Collins and Stephanie Lanza have authored a book titled Latent Class and Latent Transition Analysis: With Applications in the Social, Behavioral, and Health Sciences. The book, published by Wiley in 2010, provides a comprehensive introduction to the use of latent class analysis and latent transition analysis in applied research.

 

SAS syntax for the applied examples are available so that researchers can conduct the same analyses in their work. All latent class analyses in the book's examples were performed using PROC LCA.

 

 


 

Software

ItemResponsePlot

We distribute free software to researchers so they can use LCA accurately and easily. The Methodology Center first released PROC LCA for SAS in 2007, and has regularly added important features to the software.

 

We develop macros to enhance PROC LCA functionality.

 

We also develop non-SAS based software for LCA.

 


 

Syntax for easy to use PROC LTA Software

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;

Latent Transition Analysis (LTA)

Latent transition analysis (LTA) is used to model the transitions between underlying subgroups over time. This means that for longitudinal data (i.e., repeated-measures data) you can determine patterns of change over time, for example modeling transitions in stages of substance use in late adolescence (Lanza, Patrick, & Maggs, 2010).

  

The Methodology Center has been a pioneer in the development of LTA, and The Methodology Center's PROC LTA is a widely-used, straightforward software program for performing LTA. LTA is a separate (though related) research area.

  

Visit the LTA webpage.

 

Reference

Lanza, S. T., Patrick, M. E., & Maggs, J. L. (2010). Latent transition analysis: Benefits of a latent variable approach to modeling transitions in substance use. Journal of Drug Issues, 40(1), 93-120. PMCID: PMC2909684

Follow Us