Covariances between regression coefficient estimates in a single mediator model

Davood Tofighi, David Mackinnon, Myeongsun Yoon

Research output: Contribution to journalArticle

27 Citations (Scopus)

Abstract

This study presents formulae for the covariances between parameter estimates in a single mediator model. These covariances are necessary to build confidence intervals (CI) for effect size measures in mediation studies. We first analytically derived the covariances between the parameter estimates in a single mediator model. Using the derived covariances, we computed the multivariate-delta standard errors, and built the 95% CIs for the effect size measures. A simulation study evaluated the accuracy of the standard errors as well as the Type I error, power, and coverage of the CIs using various parameter values and sample sizes. Finally, we presented a numerical example and a SAS MACRO that calculates the CIs for the effect size measures.

Original languageEnglish (US)
Pages (from-to)457-484
Number of pages28
JournalBritish Journal of Mathematical and Statistical Psychology
Volume62
Issue number3
DOIs
StatePublished - Nov 2009

Fingerprint

Coefficient Estimates
Regression Estimate
Mediator
Regression Coefficient
Sample Size
Effect Size
Confidence Intervals
Standard error
Mediation
Type I error
Model
Estimate
Confidence interval
Coverage
Simulation Study
Calculate
Numerical Examples
Necessary

ASJC Scopus subject areas

  • Psychology(all)
  • Statistics and Probability

Cite this

Covariances between regression coefficient estimates in a single mediator model. / Tofighi, Davood; Mackinnon, David; Yoon, Myeongsun.

In: British Journal of Mathematical and Statistical Psychology, Vol. 62, No. 3, 11.2009, p. 457-484.

Research output: Contribution to journalArticle

@article{bdf0e127e24b412a83e1c1be091e290b,
title = "Covariances between regression coefficient estimates in a single mediator model",
abstract = "This study presents formulae for the covariances between parameter estimates in a single mediator model. These covariances are necessary to build confidence intervals (CI) for effect size measures in mediation studies. We first analytically derived the covariances between the parameter estimates in a single mediator model. Using the derived covariances, we computed the multivariate-delta standard errors, and built the 95{\%} CIs for the effect size measures. A simulation study evaluated the accuracy of the standard errors as well as the Type I error, power, and coverage of the CIs using various parameter values and sample sizes. Finally, we presented a numerical example and a SAS MACRO that calculates the CIs for the effect size measures.",
author = "Davood Tofighi and David Mackinnon and Myeongsun Yoon",
year = "2009",
month = "11",
doi = "10.1348/000711008X331024",
language = "English (US)",
volume = "62",
pages = "457--484",
journal = "British Journal of Mathematical and Statistical Psychology",
issn = "0007-1102",
publisher = "Wiley-Blackwell",
number = "3",

}

TY - JOUR

T1 - Covariances between regression coefficient estimates in a single mediator model

AU - Tofighi, Davood

AU - Mackinnon, David

AU - Yoon, Myeongsun

PY - 2009/11

Y1 - 2009/11

N2 - This study presents formulae for the covariances between parameter estimates in a single mediator model. These covariances are necessary to build confidence intervals (CI) for effect size measures in mediation studies. We first analytically derived the covariances between the parameter estimates in a single mediator model. Using the derived covariances, we computed the multivariate-delta standard errors, and built the 95% CIs for the effect size measures. A simulation study evaluated the accuracy of the standard errors as well as the Type I error, power, and coverage of the CIs using various parameter values and sample sizes. Finally, we presented a numerical example and a SAS MACRO that calculates the CIs for the effect size measures.

AB - This study presents formulae for the covariances between parameter estimates in a single mediator model. These covariances are necessary to build confidence intervals (CI) for effect size measures in mediation studies. We first analytically derived the covariances between the parameter estimates in a single mediator model. Using the derived covariances, we computed the multivariate-delta standard errors, and built the 95% CIs for the effect size measures. A simulation study evaluated the accuracy of the standard errors as well as the Type I error, power, and coverage of the CIs using various parameter values and sample sizes. Finally, we presented a numerical example and a SAS MACRO that calculates the CIs for the effect size measures.

UR - http://www.scopus.com/inward/record.url?scp=70449393647&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=70449393647&partnerID=8YFLogxK

U2 - 10.1348/000711008X331024

DO - 10.1348/000711008X331024

M3 - Article

C2 - 18652738

AN - SCOPUS:70449393647

VL - 62

SP - 457

EP - 484

JO - British Journal of Mathematical and Statistical Psychology

JF - British Journal of Mathematical and Statistical Psychology

SN - 0007-1102

IS - 3

ER -