Multilevel structural equation models for assessing moderation within and across levels of analysis

Kristopher J. Preacher, Zhen Zhang, Michael J. Zyphur

Research output: Contribution to journalArticle

62 Citations (Scopus)

Abstract

Social scientists are increasingly interested in multilevel hypotheses, data, and statistical models as well as moderation or interactions among predictors. The result is a focus on hypotheses and tests of multilevel moderation within and across levels of analysis. Unfortunately, existing approaches to multilevel moderation have a variety of shortcomings, including conflated effects across levels of analysis and bias due to using observed cluster averages instead of latent variables (i.e., "random intercepts") to represent higher-level constructs. To overcome these problems and elucidate the nature of multilevel moderation effects, we introduce a multilevel structural equation modeling (MSEM) logic that clarifies the nature of the problems with existing practices and remedies them with latent variable interactions. This remedy uses random coefficients and/or latent moderated structural equations (LMS) for unbiased tests of multilevel moderation. We describe our approach and provide an example using the publicly available High School and Beyond data with Mplus syntax in Appendix. Our MSEM method eliminates problems of conflated multilevel effects and reduces bias in parameter estimates while offering a coherent framework for conceptualizing and testing multilevel moderation effects.

Original languageEnglish (US)
Pages (from-to)189-205
Number of pages17
JournalPsychological Methods
Volume21
Issue number2
DOIs
StatePublished - 2016

Fingerprint

Structural Models
Statistical Models

Keywords

  • Interactions
  • Latent variables
  • Moderation
  • Multilevel modeling
  • Random slopes

ASJC Scopus subject areas

  • Psychology (miscellaneous)

Cite this

Multilevel structural equation models for assessing moderation within and across levels of analysis. / Preacher, Kristopher J.; Zhang, Zhen; Zyphur, Michael J.

In: Psychological Methods, Vol. 21, No. 2, 2016, p. 189-205.

Research output: Contribution to journalArticle

Preacher, Kristopher J. ; Zhang, Zhen ; Zyphur, Michael J. / Multilevel structural equation models for assessing moderation within and across levels of analysis. In: Psychological Methods. 2016 ; Vol. 21, No. 2. pp. 189-205.
@article{637475eac51448f69b77b2cfa1742a25,
title = "Multilevel structural equation models for assessing moderation within and across levels of analysis",
abstract = "Social scientists are increasingly interested in multilevel hypotheses, data, and statistical models as well as moderation or interactions among predictors. The result is a focus on hypotheses and tests of multilevel moderation within and across levels of analysis. Unfortunately, existing approaches to multilevel moderation have a variety of shortcomings, including conflated effects across levels of analysis and bias due to using observed cluster averages instead of latent variables (i.e., {"}random intercepts{"}) to represent higher-level constructs. To overcome these problems and elucidate the nature of multilevel moderation effects, we introduce a multilevel structural equation modeling (MSEM) logic that clarifies the nature of the problems with existing practices and remedies them with latent variable interactions. This remedy uses random coefficients and/or latent moderated structural equations (LMS) for unbiased tests of multilevel moderation. We describe our approach and provide an example using the publicly available High School and Beyond data with Mplus syntax in Appendix. Our MSEM method eliminates problems of conflated multilevel effects and reduces bias in parameter estimates while offering a coherent framework for conceptualizing and testing multilevel moderation effects.",
keywords = "Interactions, Latent variables, Moderation, Multilevel modeling, Random slopes",
author = "Preacher, {Kristopher J.} and Zhen Zhang and Zyphur, {Michael J.}",
year = "2016",
doi = "10.1037/met0000052",
language = "English (US)",
volume = "21",
pages = "189--205",
journal = "Psychological Methods",
issn = "1082-989X",
publisher = "American Psychological Association Inc.",
number = "2",

}

TY - JOUR

T1 - Multilevel structural equation models for assessing moderation within and across levels of analysis

AU - Preacher, Kristopher J.

AU - Zhang, Zhen

AU - Zyphur, Michael J.

PY - 2016

Y1 - 2016

N2 - Social scientists are increasingly interested in multilevel hypotheses, data, and statistical models as well as moderation or interactions among predictors. The result is a focus on hypotheses and tests of multilevel moderation within and across levels of analysis. Unfortunately, existing approaches to multilevel moderation have a variety of shortcomings, including conflated effects across levels of analysis and bias due to using observed cluster averages instead of latent variables (i.e., "random intercepts") to represent higher-level constructs. To overcome these problems and elucidate the nature of multilevel moderation effects, we introduce a multilevel structural equation modeling (MSEM) logic that clarifies the nature of the problems with existing practices and remedies them with latent variable interactions. This remedy uses random coefficients and/or latent moderated structural equations (LMS) for unbiased tests of multilevel moderation. We describe our approach and provide an example using the publicly available High School and Beyond data with Mplus syntax in Appendix. Our MSEM method eliminates problems of conflated multilevel effects and reduces bias in parameter estimates while offering a coherent framework for conceptualizing and testing multilevel moderation effects.

AB - Social scientists are increasingly interested in multilevel hypotheses, data, and statistical models as well as moderation or interactions among predictors. The result is a focus on hypotheses and tests of multilevel moderation within and across levels of analysis. Unfortunately, existing approaches to multilevel moderation have a variety of shortcomings, including conflated effects across levels of analysis and bias due to using observed cluster averages instead of latent variables (i.e., "random intercepts") to represent higher-level constructs. To overcome these problems and elucidate the nature of multilevel moderation effects, we introduce a multilevel structural equation modeling (MSEM) logic that clarifies the nature of the problems with existing practices and remedies them with latent variable interactions. This remedy uses random coefficients and/or latent moderated structural equations (LMS) for unbiased tests of multilevel moderation. We describe our approach and provide an example using the publicly available High School and Beyond data with Mplus syntax in Appendix. Our MSEM method eliminates problems of conflated multilevel effects and reduces bias in parameter estimates while offering a coherent framework for conceptualizing and testing multilevel moderation effects.

KW - Interactions

KW - Latent variables

KW - Moderation

KW - Multilevel modeling

KW - Random slopes

UR - http://www.scopus.com/inward/record.url?scp=84949921242&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84949921242&partnerID=8YFLogxK

U2 - 10.1037/met0000052

DO - 10.1037/met0000052

M3 - Article

C2 - 26651982

AN - SCOPUS:84949921242

VL - 21

SP - 189

EP - 205

JO - Psychological Methods

JF - Psychological Methods

SN - 1082-989X

IS - 2

ER -