Abstract

Bayesian methods have the potential for increasing power in mediation analysis (Koopman, Howe, Hollenbeck, & Sin, 2015; Yuan & MacKinnon, 2009). This article compares the power of Bayesian credibility intervals for the mediated effect to the power of normal theory, distribution of the product, percentile, and bias-corrected bootstrap confidence intervals at N ≤ 200. Bayesian methods with diffuse priors have power comparable to the distribution of the product and bootstrap methods, and Bayesian methods with informative priors had the most power. Varying degrees of precision of prior distributions were also examined. Increased precision led to greater power only when N ≥ 100 and the effects were small, N < 60 and the effects were large, and N < 200 and the effects were medium. An empirical example from psychology illustrated a Bayesian analysis of the single mediator model from prior selection to interpreting results.

Original languageEnglish (US)
Pages (from-to)1-18
Number of pages18
JournalStructural Equation Modeling
DOIs
StateAccepted/In press - Apr 25 2017

Fingerprint

Mediation
Small Sample
mediation
Bayesian Methods
Bootstrap Confidence Intervals
distribution theory
Distribution Theory
Mediator
Percentile
Bootstrap Method
Credibility
Bayesian Analysis
Prior distribution
great power
credibility
psychology
confidence
Interval
Small sample
trend

Keywords

  • Bayesian statistics
  • power
  • single mediator model
  • small sample sizes

ASJC Scopus subject areas

  • Decision Sciences(all)
  • Modeling and Simulation
  • Sociology and Political Science
  • Economics, Econometrics and Finance(all)

Cite this

Power in Bayesian Mediation Analysis for Small Sample Research. / Miočević, Milica; Mackinnon, David; Levy, Roy.

In: Structural Equation Modeling, 25.04.2017, p. 1-18.

Research output: Contribution to journalArticle

@article{71a08cdb1d7a41578ccfc54abb437a22,
title = "Power in Bayesian Mediation Analysis for Small Sample Research",
abstract = "Bayesian methods have the potential for increasing power in mediation analysis (Koopman, Howe, Hollenbeck, & Sin, 2015; Yuan & MacKinnon, 2009). This article compares the power of Bayesian credibility intervals for the mediated effect to the power of normal theory, distribution of the product, percentile, and bias-corrected bootstrap confidence intervals at N ≤ 200. Bayesian methods with diffuse priors have power comparable to the distribution of the product and bootstrap methods, and Bayesian methods with informative priors had the most power. Varying degrees of precision of prior distributions were also examined. Increased precision led to greater power only when N ≥ 100 and the effects were small, N < 60 and the effects were large, and N < 200 and the effects were medium. An empirical example from psychology illustrated a Bayesian analysis of the single mediator model from prior selection to interpreting results.",
keywords = "Bayesian statistics, power, single mediator model, small sample sizes",
author = "Milica Miočević and David Mackinnon and Roy Levy",
year = "2017",
month = "4",
day = "25",
doi = "10.1080/10705511.2017.1312407",
language = "English (US)",
pages = "1--18",
journal = "Structural Equation Modeling",
issn = "1070-5511",
publisher = "Psychology Press Ltd",

}

TY - JOUR

T1 - Power in Bayesian Mediation Analysis for Small Sample Research

AU - Miočević, Milica

AU - Mackinnon, David

AU - Levy, Roy

PY - 2017/4/25

Y1 - 2017/4/25

N2 - Bayesian methods have the potential for increasing power in mediation analysis (Koopman, Howe, Hollenbeck, & Sin, 2015; Yuan & MacKinnon, 2009). This article compares the power of Bayesian credibility intervals for the mediated effect to the power of normal theory, distribution of the product, percentile, and bias-corrected bootstrap confidence intervals at N ≤ 200. Bayesian methods with diffuse priors have power comparable to the distribution of the product and bootstrap methods, and Bayesian methods with informative priors had the most power. Varying degrees of precision of prior distributions were also examined. Increased precision led to greater power only when N ≥ 100 and the effects were small, N < 60 and the effects were large, and N < 200 and the effects were medium. An empirical example from psychology illustrated a Bayesian analysis of the single mediator model from prior selection to interpreting results.

AB - Bayesian methods have the potential for increasing power in mediation analysis (Koopman, Howe, Hollenbeck, & Sin, 2015; Yuan & MacKinnon, 2009). This article compares the power of Bayesian credibility intervals for the mediated effect to the power of normal theory, distribution of the product, percentile, and bias-corrected bootstrap confidence intervals at N ≤ 200. Bayesian methods with diffuse priors have power comparable to the distribution of the product and bootstrap methods, and Bayesian methods with informative priors had the most power. Varying degrees of precision of prior distributions were also examined. Increased precision led to greater power only when N ≥ 100 and the effects were small, N < 60 and the effects were large, and N < 200 and the effects were medium. An empirical example from psychology illustrated a Bayesian analysis of the single mediator model from prior selection to interpreting results.

KW - Bayesian statistics

KW - power

KW - single mediator model

KW - small sample sizes

UR - http://www.scopus.com/inward/record.url?scp=85018668544&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85018668544&partnerID=8YFLogxK

U2 - 10.1080/10705511.2017.1312407

DO - 10.1080/10705511.2017.1312407

M3 - Article

AN - SCOPUS:85018668544

SP - 1

EP - 18

JO - Structural Equation Modeling

JF - Structural Equation Modeling

SN - 1070-5511

ER -