Multilevel modeling of individual and group level mediated effects

Jennifer L. Krull, David Mackinnon

Research output: Contribution to journalArticle

738 Citations (Scopus)

Abstract

This article combines procedures for single-level mediational analysis with multilevel modeling techniques in order to appropriately test mediational effects in clustered data. A simulation study compared the performance of these multilevel mediational models with that of single-level mediational models in clustered data with individual- or group-level initial independent variables, individual- or group-level mediators, and individual level outcomes. The standard errors of mediated effects from the multilevel solution were generally accurate, while those from the single-level procedure were downwardly biased, often by 20% or more. The multilevel advantage was greatest in those situations involving group-level variables, larger group sizes, and higher intraclass correlations in mediator and outcome variables. Multilevel mediational modeling methods were also applied to data from a preventive intervention designed to reduce intentions to use steroids among players on high school football teams. This example illustrates differences between single-level and multilevel mediational modeling in real-world clustered data and shows how the multilevel technique may lead to more accurate results.

Original languageEnglish (US)
Pages (from-to)249-277
Number of pages29
JournalMultivariate Behavioral Research
Volume36
Issue number2
StatePublished - 2001

Fingerprint

Multilevel Modeling
Clustered Data
Mediator
Multilevel Analysis
Football
Group
Intraclass Correlation
Multilevel Models
Multilevel Methods
Steroids
group size
Standard error
Modeling Method
Biased
Simulation Study
simulation
school
performance

ASJC Scopus subject areas

  • Mathematics (miscellaneous)
  • Statistics and Probability
  • Psychology(all)
  • Experimental and Cognitive Psychology
  • Social Sciences (miscellaneous)

Cite this

Multilevel modeling of individual and group level mediated effects. / Krull, Jennifer L.; Mackinnon, David.

In: Multivariate Behavioral Research, Vol. 36, No. 2, 2001, p. 249-277.

Research output: Contribution to journalArticle

@article{9a78f8d080e2445796b125bcb7bfeafa,
title = "Multilevel modeling of individual and group level mediated effects",
abstract = "This article combines procedures for single-level mediational analysis with multilevel modeling techniques in order to appropriately test mediational effects in clustered data. A simulation study compared the performance of these multilevel mediational models with that of single-level mediational models in clustered data with individual- or group-level initial independent variables, individual- or group-level mediators, and individual level outcomes. The standard errors of mediated effects from the multilevel solution were generally accurate, while those from the single-level procedure were downwardly biased, often by 20{\%} or more. The multilevel advantage was greatest in those situations involving group-level variables, larger group sizes, and higher intraclass correlations in mediator and outcome variables. Multilevel mediational modeling methods were also applied to data from a preventive intervention designed to reduce intentions to use steroids among players on high school football teams. This example illustrates differences between single-level and multilevel mediational modeling in real-world clustered data and shows how the multilevel technique may lead to more accurate results.",
author = "Krull, {Jennifer L.} and David Mackinnon",
year = "2001",
language = "English (US)",
volume = "36",
pages = "249--277",
journal = "Multivariate Behavioral Research",
issn = "0027-3171",
publisher = "Psychology Press Ltd",
number = "2",

}

TY - JOUR

T1 - Multilevel modeling of individual and group level mediated effects

AU - Krull, Jennifer L.

AU - Mackinnon, David

PY - 2001

Y1 - 2001

N2 - This article combines procedures for single-level mediational analysis with multilevel modeling techniques in order to appropriately test mediational effects in clustered data. A simulation study compared the performance of these multilevel mediational models with that of single-level mediational models in clustered data with individual- or group-level initial independent variables, individual- or group-level mediators, and individual level outcomes. The standard errors of mediated effects from the multilevel solution were generally accurate, while those from the single-level procedure were downwardly biased, often by 20% or more. The multilevel advantage was greatest in those situations involving group-level variables, larger group sizes, and higher intraclass correlations in mediator and outcome variables. Multilevel mediational modeling methods were also applied to data from a preventive intervention designed to reduce intentions to use steroids among players on high school football teams. This example illustrates differences between single-level and multilevel mediational modeling in real-world clustered data and shows how the multilevel technique may lead to more accurate results.

AB - This article combines procedures for single-level mediational analysis with multilevel modeling techniques in order to appropriately test mediational effects in clustered data. A simulation study compared the performance of these multilevel mediational models with that of single-level mediational models in clustered data with individual- or group-level initial independent variables, individual- or group-level mediators, and individual level outcomes. The standard errors of mediated effects from the multilevel solution were generally accurate, while those from the single-level procedure were downwardly biased, often by 20% or more. The multilevel advantage was greatest in those situations involving group-level variables, larger group sizes, and higher intraclass correlations in mediator and outcome variables. Multilevel mediational modeling methods were also applied to data from a preventive intervention designed to reduce intentions to use steroids among players on high school football teams. This example illustrates differences between single-level and multilevel mediational modeling in real-world clustered data and shows how the multilevel technique may lead to more accurate results.

UR - http://www.scopus.com/inward/record.url?scp=0035535604&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0035535604&partnerID=8YFLogxK

M3 - Article

VL - 36

SP - 249

EP - 277

JO - Multivariate Behavioral Research

JF - Multivariate Behavioral Research

SN - 0027-3171

IS - 2

ER -