Online Program Home
My Program

Abstract Details

Activity Number: 358 - Contributed Poster Presentations: Section on Statistics in Epidemiology
Type: Contributed
Date/Time: Tuesday, July 30, 2019 : 10:30 AM to 12:20 PM
Sponsor: Section on Statistics in Epidemiology
Abstract #304909
Title: Statistical Issues for Latent Class Analysis
Author(s): Tzu-Cheg Kao*
Companies: Uniformed Services University of the Health Sciences
Keywords: Latent class analysis; Collinearity; categorization; physical activity; sleep; diets

How collinearity or categorizations of items impact latent class analysis (LCA) was not known. The data based on the 2011 Survey of Health-Related Behaviors among US military active duty personnel were used to illustrate some of the statistical issues. For LCA examples: Ten items will be considered, six food items (fruit, vegetable, whole grains, sweets, sugar drinks, and fried foods: dichotomized as less or more), moderate activity, vigorous activity, strength training (less, moderate, more) and sleep hours were used. Given the set of 10 items, we determined the appropriate number of classes to extract, by using Lo-Mendell-Rubin adjusted likelihood ratio test, along with and other fit statistics (Log-likelihood, Akaike information criteria; Bayes information criteria, entropy). M?plus version 8 will be used for various situations of the ten items to obtain appropriate latent classes, and results will be discussed.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2019 program