A comparison of methods to test mediation and other intervening variable effects

David P. MacKinnon; Chondra M. Lockwood; Jeanne M. Hoffman; Stephen G. West; Virgil Sheets

doi:10.1037/1082-989X.7.1.83

A comparison of methods to test mediation and other intervening variable effects

David P. MacKinnon, Chondra M. Lockwood, Jeanne M. Hoffman, Stephen G. West, Virgil Sheets

Psychology

Research output: Contribution to journal › Article › peer-review

7135 Scopus citations

Abstract

A Monte Carlo study compared 14 methods to test the statistical significance of the intervening variable effect. An intervening variable (mediator) transmits the effect of an independent variable to a dependent variable. The commonly used R. M. Baron and D. A. Kenny (1986) approach has low statistical power. Two methods based on the distribution of the product and 2 difference-in-coefficients methods have the most accurate Type I error rates and greatest statistical power except in 1 important case in which Type I error rates are too high. The best balance of Type I error and statistical power across all cases is the test of the joint significance of the two effects comprising the intervening variable effect.

Original language	English (US)
Pages (from-to)	83-104
Number of pages	22
Journal	Psychological Methods
Volume	7
Issue number	1
DOIs	https://doi.org/10.1037/1082-989X.7.1.83
State	Published - Mar 2002

ASJC Scopus subject areas

Psychology (miscellaneous)

Access to Document

10.1037/1082-989X.7.1.83

Cite this

@article{8a38353fdaae4c52957a362a4b963a44,

title = "A comparison of methods to test mediation and other intervening variable effects",

abstract = "A Monte Carlo study compared 14 methods to test the statistical significance of the intervening variable effect. An intervening variable (mediator) transmits the effect of an independent variable to a dependent variable. The commonly used R. M. Baron and D. A. Kenny (1986) approach has low statistical power. Two methods based on the distribution of the product and 2 difference-in-coefficients methods have the most accurate Type I error rates and greatest statistical power except in 1 important case in which Type I error rates are too high. The best balance of Type I error and statistical power across all cases is the test of the joint significance of the two effects comprising the intervening variable effect.",

author = "MacKinnon, {David P.} and Lockwood, {Chondra M.} and Hoffman, {Jeanne M.} and West, {Stephen G.} and Virgil Sheets",

year = "2002",

month = mar,

doi = "10.1037/1082-989X.7.1.83",

language = "English (US)",

volume = "7",

pages = "83--104",

journal = "Psychological Methods",

issn = "1082-989X",

publisher = "American Psychological Association Inc.",

number = "1",

}

TY - JOUR

T1 - A comparison of methods to test mediation and other intervening variable effects

AU - MacKinnon, David P.

AU - Lockwood, Chondra M.

AU - Hoffman, Jeanne M.

AU - West, Stephen G.

AU - Sheets, Virgil

PY - 2002/3

Y1 - 2002/3

N2 - A Monte Carlo study compared 14 methods to test the statistical significance of the intervening variable effect. An intervening variable (mediator) transmits the effect of an independent variable to a dependent variable. The commonly used R. M. Baron and D. A. Kenny (1986) approach has low statistical power. Two methods based on the distribution of the product and 2 difference-in-coefficients methods have the most accurate Type I error rates and greatest statistical power except in 1 important case in which Type I error rates are too high. The best balance of Type I error and statistical power across all cases is the test of the joint significance of the two effects comprising the intervening variable effect.

AB - A Monte Carlo study compared 14 methods to test the statistical significance of the intervening variable effect. An intervening variable (mediator) transmits the effect of an independent variable to a dependent variable. The commonly used R. M. Baron and D. A. Kenny (1986) approach has low statistical power. Two methods based on the distribution of the product and 2 difference-in-coefficients methods have the most accurate Type I error rates and greatest statistical power except in 1 important case in which Type I error rates are too high. The best balance of Type I error and statistical power across all cases is the test of the joint significance of the two effects comprising the intervening variable effect.

UR - http://www.scopus.com/inward/record.url?scp=85035930668&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85035930668&partnerID=8YFLogxK

U2 - 10.1037/1082-989X.7.1.83

DO - 10.1037/1082-989X.7.1.83

M3 - Article

C2 - 11928892

AN - SCOPUS:85035930668

SN - 1082-989X

VL - 7

SP - 83

EP - 104

JO - Psychological Methods

JF - Psychological Methods

IS - 1

ER -

A comparison of methods to test mediation and other intervening variable effects

Abstract

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this