A Bayesian Approach for Estimating Mediation Effects With Missing Data

Craig K. Enders, Amanda J. Fairchild, David Mackinnon

Research output: Contribution to journalArticle

18 Citations (Scopus)

Abstract

Methodologists have developed mediation analysis techniques for a broad range of substantive applications, yet methods for estimating mediating mechanisms with missing data have been understudied. This study outlined a general Bayesian missing data handling approach that can accommodate mediation analyses with any number of manifest variables. Computer simulation studies showed that the Bayesian approach produced frequentist coverage rates and power estimates that were comparable to those of maximum likelihood with the bias-corrected bootstrap. We share an SAS macro that implements Bayesian estimation and use 2 data analysis examples to demonstrate its use.

Original languageEnglish (US)
Pages (from-to)340-369
Number of pages30
JournalMultivariate Behavioral Research
Volume48
Issue number3
DOIs
StatePublished - May 2013

Fingerprint

Bayes Theorem
Mediation
Missing Data
Bayesian Approach
Computer Simulation
Data Handling
Bayesian Estimation
Bootstrap
Maximum Likelihood
Data analysis
Coverage
Simulation Study
Estimate
Range of data
Demonstrate

ASJC Scopus subject areas

  • Experimental and Cognitive Psychology
  • Statistics and Probability
  • Arts and Humanities (miscellaneous)

Cite this

A Bayesian Approach for Estimating Mediation Effects With Missing Data. / Enders, Craig K.; Fairchild, Amanda J.; Mackinnon, David.

In: Multivariate Behavioral Research, Vol. 48, No. 3, 05.2013, p. 340-369.

Research output: Contribution to journalArticle

Enders, Craig K. ; Fairchild, Amanda J. ; Mackinnon, David. / A Bayesian Approach for Estimating Mediation Effects With Missing Data. In: Multivariate Behavioral Research. 2013 ; Vol. 48, No. 3. pp. 340-369.
@article{8dcdcaed228b4368ac0b7456d77c3620,
title = "A Bayesian Approach for Estimating Mediation Effects With Missing Data",
abstract = "Methodologists have developed mediation analysis techniques for a broad range of substantive applications, yet methods for estimating mediating mechanisms with missing data have been understudied. This study outlined a general Bayesian missing data handling approach that can accommodate mediation analyses with any number of manifest variables. Computer simulation studies showed that the Bayesian approach produced frequentist coverage rates and power estimates that were comparable to those of maximum likelihood with the bias-corrected bootstrap. We share an SAS macro that implements Bayesian estimation and use 2 data analysis examples to demonstrate its use.",
author = "Enders, {Craig K.} and Fairchild, {Amanda J.} and David Mackinnon",
year = "2013",
month = "5",
doi = "10.1080/00273171.2013.784862",
language = "English (US)",
volume = "48",
pages = "340--369",
journal = "Multivariate Behavioral Research",
issn = "0027-3171",
publisher = "Psychology Press Ltd",
number = "3",

}

TY - JOUR

T1 - A Bayesian Approach for Estimating Mediation Effects With Missing Data

AU - Enders, Craig K.

AU - Fairchild, Amanda J.

AU - Mackinnon, David

PY - 2013/5

Y1 - 2013/5

N2 - Methodologists have developed mediation analysis techniques for a broad range of substantive applications, yet methods for estimating mediating mechanisms with missing data have been understudied. This study outlined a general Bayesian missing data handling approach that can accommodate mediation analyses with any number of manifest variables. Computer simulation studies showed that the Bayesian approach produced frequentist coverage rates and power estimates that were comparable to those of maximum likelihood with the bias-corrected bootstrap. We share an SAS macro that implements Bayesian estimation and use 2 data analysis examples to demonstrate its use.

AB - Methodologists have developed mediation analysis techniques for a broad range of substantive applications, yet methods for estimating mediating mechanisms with missing data have been understudied. This study outlined a general Bayesian missing data handling approach that can accommodate mediation analyses with any number of manifest variables. Computer simulation studies showed that the Bayesian approach produced frequentist coverage rates and power estimates that were comparable to those of maximum likelihood with the bias-corrected bootstrap. We share an SAS macro that implements Bayesian estimation and use 2 data analysis examples to demonstrate its use.

UR - http://www.scopus.com/inward/record.url?scp=84878983668&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84878983668&partnerID=8YFLogxK

U2 - 10.1080/00273171.2013.784862

DO - 10.1080/00273171.2013.784862

M3 - Article

AN - SCOPUS:84878983668

VL - 48

SP - 340

EP - 369

JO - Multivariate Behavioral Research

JF - Multivariate Behavioral Research

SN - 0027-3171

IS - 3

ER -