TY - JOUR

T1 - Statistical Power to Detect the Correct Number of Classes in Latent Profile Analysis

AU - Tein, Jenn-Yun

AU - Coxe, Stefany

AU - Cham, Heining

N1 - Funding Information:
We gratefully acknowledge that this work was conducted with support from the National Institute of Mental Health (NIMH), NIMH R01 Grant MH49155; NIMH R01 Grant MH68920; and NIMH P30 Grant MH068685. This article is based on a poster presentation given at the 74th Annual and the 16th International Meeting of the Psychometric Society, Cambridge, United Kingdom, July 2009.

PY - 2013/10

Y1 - 2013/10

N2 - Little research has examined factors influencing statistical power to detect the correct number of latent classes using latent profile analysis (LPA). This simulation study examined power related to interclass distance between latent classes given true number of classes, sample size, and number of indicators. Seven model selection methods were evaluated. None had adequate power to select the correct number of classes with a small (Cohen's d =.2) or medium (d =.5) degree of separation. With a very large degree of separation (d = 1.5), the Lo-Mendell-Rubin test (LMR), adjusted LMR, bootstrap likelihood ratio test, Bayesian Information Criterion (BIC), and sample-size-adjusted BIC were good at selecting the correct number of classes. However, with a large degree of separation (d =.8), power depended on number of indicators and sample size. Akaike's Information Criterion and entropy poorly selected the correct number of classes, regardless of degree of separation, number of indicators, or sample size.

AB - Little research has examined factors influencing statistical power to detect the correct number of latent classes using latent profile analysis (LPA). This simulation study examined power related to interclass distance between latent classes given true number of classes, sample size, and number of indicators. Seven model selection methods were evaluated. None had adequate power to select the correct number of classes with a small (Cohen's d =.2) or medium (d =.5) degree of separation. With a very large degree of separation (d = 1.5), the Lo-Mendell-Rubin test (LMR), adjusted LMR, bootstrap likelihood ratio test, Bayesian Information Criterion (BIC), and sample-size-adjusted BIC were good at selecting the correct number of classes. However, with a large degree of separation (d =.8), power depended on number of indicators and sample size. Akaike's Information Criterion and entropy poorly selected the correct number of classes, regardless of degree of separation, number of indicators, or sample size.

KW - interclass distance

KW - latent profile analysis

KW - mixture models

KW - model section methods

KW - power

UR - http://www.scopus.com/inward/record.url?scp=84886897982&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84886897982&partnerID=8YFLogxK

U2 - 10.1080/10705511.2013.824781

DO - 10.1080/10705511.2013.824781

M3 - Article

AN - SCOPUS:84886897982

VL - 20

SP - 640

EP - 657

JO - Structural Equation Modeling

JF - Structural Equation Modeling

SN - 1070-5511

IS - 4

ER -