Demonstration and evaluation of a method for assessing mediated moderation

Antonio A. Morgan-Lopez, David Mackinnon

Research output: Contribution to journalArticle

126 Citations (Scopus)

Abstract

Mediated moderation occurs when the interaction between two variables affects a mediator, which then affects a dependent variable. In this article, we describe the mediated moderation model and evaluate it with a statistical simulation using an adaptation of product-of-coefficients methods to assess mediation. We also demonstrate the use of this method with a substantive example from the adolescent tobacco literature. In the simulation, relative bias (RB) in point estimates and standard errors did not exceed problematic levels of ±10%, although systematic variability in RB was accounted for by parameter size, sample size, and nonzero direct effects. Power to detect mediated moderation effects appears to be severely compromised under one particular combination of conditions: when the component variables that make up the interaction terms are correlated and partial mediated moderation exists. Implications for the estimation of mediated moderation effects in experimental and nohexperimental research are discussed.

Original languageEnglish (US)
Pages (from-to)77-87
Number of pages11
JournalBehavior Research Methods
Volume38
Issue number1
StatePublished - Feb 2006

Fingerprint

Sample Size
Tobacco
Research
Moderation
Evaluation
Simulation
Interaction
Mediation
Mediator

ASJC Scopus subject areas

  • Psychology(all)
  • Psychology (miscellaneous)
  • Experimental and Cognitive Psychology

Cite this

Demonstration and evaluation of a method for assessing mediated moderation. / Morgan-Lopez, Antonio A.; Mackinnon, David.

In: Behavior Research Methods, Vol. 38, No. 1, 02.2006, p. 77-87.

Research output: Contribution to journalArticle

@article{46895b7bdbaa461faa35fffe4a579e32,
title = "Demonstration and evaluation of a method for assessing mediated moderation",
abstract = "Mediated moderation occurs when the interaction between two variables affects a mediator, which then affects a dependent variable. In this article, we describe the mediated moderation model and evaluate it with a statistical simulation using an adaptation of product-of-coefficients methods to assess mediation. We also demonstrate the use of this method with a substantive example from the adolescent tobacco literature. In the simulation, relative bias (RB) in point estimates and standard errors did not exceed problematic levels of ±10{\%}, although systematic variability in RB was accounted for by parameter size, sample size, and nonzero direct effects. Power to detect mediated moderation effects appears to be severely compromised under one particular combination of conditions: when the component variables that make up the interaction terms are correlated and partial mediated moderation exists. Implications for the estimation of mediated moderation effects in experimental and nohexperimental research are discussed.",
author = "Morgan-Lopez, {Antonio A.} and David Mackinnon",
year = "2006",
month = "2",
language = "English (US)",
volume = "38",
pages = "77--87",
journal = "Behavior Research Methods, Instruments, and Computers",
issn = "1554-351X",
publisher = "Springer New York",
number = "1",

}

TY - JOUR

T1 - Demonstration and evaluation of a method for assessing mediated moderation

AU - Morgan-Lopez, Antonio A.

AU - Mackinnon, David

PY - 2006/2

Y1 - 2006/2

N2 - Mediated moderation occurs when the interaction between two variables affects a mediator, which then affects a dependent variable. In this article, we describe the mediated moderation model and evaluate it with a statistical simulation using an adaptation of product-of-coefficients methods to assess mediation. We also demonstrate the use of this method with a substantive example from the adolescent tobacco literature. In the simulation, relative bias (RB) in point estimates and standard errors did not exceed problematic levels of ±10%, although systematic variability in RB was accounted for by parameter size, sample size, and nonzero direct effects. Power to detect mediated moderation effects appears to be severely compromised under one particular combination of conditions: when the component variables that make up the interaction terms are correlated and partial mediated moderation exists. Implications for the estimation of mediated moderation effects in experimental and nohexperimental research are discussed.

AB - Mediated moderation occurs when the interaction between two variables affects a mediator, which then affects a dependent variable. In this article, we describe the mediated moderation model and evaluate it with a statistical simulation using an adaptation of product-of-coefficients methods to assess mediation. We also demonstrate the use of this method with a substantive example from the adolescent tobacco literature. In the simulation, relative bias (RB) in point estimates and standard errors did not exceed problematic levels of ±10%, although systematic variability in RB was accounted for by parameter size, sample size, and nonzero direct effects. Power to detect mediated moderation effects appears to be severely compromised under one particular combination of conditions: when the component variables that make up the interaction terms are correlated and partial mediated moderation exists. Implications for the estimation of mediated moderation effects in experimental and nohexperimental research are discussed.

UR - http://www.scopus.com/inward/record.url?scp=33746699697&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33746699697&partnerID=8YFLogxK

M3 - Article

C2 - 16817516

AN - SCOPUS:33746699697

VL - 38

SP - 77

EP - 87

JO - Behavior Research Methods, Instruments, and Computers

JF - Behavior Research Methods, Instruments, and Computers

SN - 1554-351X

IS - 1

ER -