Using Multilevel Regression Mixture Models to Identify Level-1 Heterogeneity in Level-2 Effects

M. Lee Van Horn, Yuling Feng, Minjung Kim, Andrea Lamont, Daniel J Feaster, Thomas Jaki

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

This article proposes a novel exploratory approach for assessing how the effects of Level-2 predictors differ across Level-1 units. Multilevel regression mixture models are used to identify latent classes at Level 1 that differ in the effect of 1 or more Level-2 predictors. Monte Carlo simulations are used to demonstrate the approach with different sample sizes and to demonstrate the consequences of constraining 1 of the random effects to 0. An application of the method to evaluate heterogeneity in the effects of classroom practices on students is used to show the types of research questions that can be answered with this method and the issues faced when estimating multilevel regression mixtures.

Original languageEnglish (US)
Pages (from-to)259-269
Number of pages11
JournalStructural Equation Modeling
Volume23
Issue number2
DOIs
StatePublished - Mar 3 2016

Fingerprint

Mixture Model
Regression Model
regression
Predictors
Latent Class
Students
Random Effects
Demonstrate
Sample Size
Monte Carlo Simulation
Regression
Unit
Evaluate
classroom
simulation
Mixture model
student
Monte Carlo simulation

Keywords

  • heterogeneity in contextual effects
  • multilevel regression mixtures
  • regression mixture modeling

ASJC Scopus subject areas

  • Modeling and Simulation
  • Decision Sciences(all)
  • Economics, Econometrics and Finance(all)
  • Sociology and Political Science

Cite this

Using Multilevel Regression Mixture Models to Identify Level-1 Heterogeneity in Level-2 Effects. / Van Horn, M. Lee; Feng, Yuling; Kim, Minjung; Lamont, Andrea; Feaster, Daniel J; Jaki, Thomas.

In: Structural Equation Modeling, Vol. 23, No. 2, 03.03.2016, p. 259-269.

Research output: Contribution to journalArticle

Van Horn, M. Lee ; Feng, Yuling ; Kim, Minjung ; Lamont, Andrea ; Feaster, Daniel J ; Jaki, Thomas. / Using Multilevel Regression Mixture Models to Identify Level-1 Heterogeneity in Level-2 Effects. In: Structural Equation Modeling. 2016 ; Vol. 23, No. 2. pp. 259-269.
@article{b58317e03ba344e78d421d681dab111d,
title = "Using Multilevel Regression Mixture Models to Identify Level-1 Heterogeneity in Level-2 Effects",
abstract = "This article proposes a novel exploratory approach for assessing how the effects of Level-2 predictors differ across Level-1 units. Multilevel regression mixture models are used to identify latent classes at Level 1 that differ in the effect of 1 or more Level-2 predictors. Monte Carlo simulations are used to demonstrate the approach with different sample sizes and to demonstrate the consequences of constraining 1 of the random effects to 0. An application of the method to evaluate heterogeneity in the effects of classroom practices on students is used to show the types of research questions that can be answered with this method and the issues faced when estimating multilevel regression mixtures.",
keywords = "heterogeneity in contextual effects, multilevel regression mixtures, regression mixture modeling",
author = "{Van Horn}, {M. Lee} and Yuling Feng and Minjung Kim and Andrea Lamont and Feaster, {Daniel J} and Thomas Jaki",
year = "2016",
month = "3",
day = "3",
doi = "10.1080/10705511.2015.1035437",
language = "English (US)",
volume = "23",
pages = "259--269",
journal = "Structural Equation Modeling",
issn = "1070-5511",
publisher = "Psychology Press Ltd",
number = "2",

}

TY - JOUR

T1 - Using Multilevel Regression Mixture Models to Identify Level-1 Heterogeneity in Level-2 Effects

AU - Van Horn, M. Lee

AU - Feng, Yuling

AU - Kim, Minjung

AU - Lamont, Andrea

AU - Feaster, Daniel J

AU - Jaki, Thomas

PY - 2016/3/3

Y1 - 2016/3/3

N2 - This article proposes a novel exploratory approach for assessing how the effects of Level-2 predictors differ across Level-1 units. Multilevel regression mixture models are used to identify latent classes at Level 1 that differ in the effect of 1 or more Level-2 predictors. Monte Carlo simulations are used to demonstrate the approach with different sample sizes and to demonstrate the consequences of constraining 1 of the random effects to 0. An application of the method to evaluate heterogeneity in the effects of classroom practices on students is used to show the types of research questions that can be answered with this method and the issues faced when estimating multilevel regression mixtures.

AB - This article proposes a novel exploratory approach for assessing how the effects of Level-2 predictors differ across Level-1 units. Multilevel regression mixture models are used to identify latent classes at Level 1 that differ in the effect of 1 or more Level-2 predictors. Monte Carlo simulations are used to demonstrate the approach with different sample sizes and to demonstrate the consequences of constraining 1 of the random effects to 0. An application of the method to evaluate heterogeneity in the effects of classroom practices on students is used to show the types of research questions that can be answered with this method and the issues faced when estimating multilevel regression mixtures.

KW - heterogeneity in contextual effects

KW - multilevel regression mixtures

KW - regression mixture modeling

UR - http://www.scopus.com/inward/record.url?scp=84957432599&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84957432599&partnerID=8YFLogxK

U2 - 10.1080/10705511.2015.1035437

DO - 10.1080/10705511.2015.1035437

M3 - Article

AN - SCOPUS:84957432599

VL - 23

SP - 259

EP - 269

JO - Structural Equation Modeling

JF - Structural Equation Modeling

SN - 1070-5511

IS - 2

ER -