Back to search

University of North Texas

Indicator Measurement Noninvariance in Latent Profile Analysis

Abstract

dc:description

Latent profile analysis is a person-centered approach to modeling unobserved heterogeneity with homogenous subgroups based on similar, defined characteristics. Observed variables are selected to create meaningful subgroups based on similar patterns in responses to those variables to guide decisions on treatments or interventions needed. Measurement invariance is an important assumption for observed variables used in a latent profile analysis to ensure the constructs measure the same across groups. Few studies have been designed to examine the impact of increasing degrees of measurement noninvariance if as few as one indicator variable has group differences. This Monte Carlo simulation study examined the impact of increasing degrees of measurement noninvariance on the accuracy of model selection in latent profile analysis. A fully crossed factorial design was used, varying sample size, number of indicator variables, class separation, and degree of measurement noninvariance using Cohen’s d. Results showed that as measurement noninvariance increased, AIC selected inaccurate models earlier than BIC and sBIC, particularly under conditions of smaller sample sizes. These findings highlight the sensitivity of model selection criteria to measure noninvariance and underscore the importance of evaluating measurement assumptions in latent profile analysis.

Degree

thesis:*
Grantor dc:publisher
University of North Texas
Year dc:date
2026

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Jackson, Sarah 1978-
Contributors dc:contributor
  • Henson, Robin K.
  • Allen, Carrie D.
  • Uanhoro, James O.
  • Hull, Darrell

Subjects

dc:subject × 3

Rights

dc:rights
Statement dc:rights
  • Public
  • Jackson, Sarah 1978-
  • Copyright
  • Copyright is held by the author, unless otherwise noted. All rights Reserved.
Language dc:language
English

Identifiers

dc:identifier.*
Identifier
local-cont-no: submission_5060
ark: ark:/67531/metadc2652549

Chain of custody

source
Harvested from
University of North Texas
Base URL
digital.library.unt.edu/oai/
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
citation

Jackson, Sarah 1978-. Indicator Measurement Noninvariance in Latent Profile Analysis. University of North Texas, 2026. https://digital.library.unt.edu/ark:/67531/metadc2652549/