A simpler approach for mediation analysis for dichotomous mediators in logistic regression

Hani Samawi, Jingxian Cai, Daniel F Linder, Haresh Rochani, Jingjing Yin

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

Mediation is a hypothesized causal chain among three variables. Mediation analysis for continuous response variables is well developed in the literature, and it can be shown that the indirect effect is equal to the total effect minus the direct effect. However, mediation analysis for categorical responses is still not fully developed. The purpose of this article is to propose a simpler method of analysing the mediation effect among three variables when the dependent and mediator variables are both dichotomous. We propose using the latent variable technique which in turn will adjust for the necessary condition that indirect effect is equal to the total effect minus the direct effect. An intensive simulation study is conducted to compare the proposed method with other methods in the literature. Our theoretical derivation and simulation study show that the proposed approach is simpler to use and at least as good as other approaches provided in the literature. We illustrate our approach to test for the potential mediators on the relationship between depression and obesity among children and adolescents compared to the method in Winship and Mare using National children health survey data 2011–2012.

Original languageEnglish (US)
Pages (from-to)1211-1227
Number of pages17
JournalJournal of Statistical Computation and Simulation
Volume88
Issue number6
DOIs
StatePublished - Apr 13 2018
Externally publishedYes

Fingerprint

Mediation
Mediator
Logistic Regression
Logistics
Direct Effect
Simulation Study
Obesity
Survey Data
Latent Variables
Categorical
Health
Health Surveys
Logistic regression
Necessary Conditions
Dependent
Direct effect
Simulation study
Indirect effects

Keywords

  • Causal inference
  • childhood obesity
  • indirect effect
  • latent variables
  • logistic regression
  • mediation analysis
  • national children health survey

ASJC Scopus subject areas

  • Statistics and Probability
  • Modeling and Simulation
  • Statistics, Probability and Uncertainty
  • Applied Mathematics

Cite this

A simpler approach for mediation analysis for dichotomous mediators in logistic regression. / Samawi, Hani; Cai, Jingxian; Linder, Daniel F; Rochani, Haresh; Yin, Jingjing.

In: Journal of Statistical Computation and Simulation, Vol. 88, No. 6, 13.04.2018, p. 1211-1227.

Research output: Contribution to journalArticle

Samawi, Hani ; Cai, Jingxian ; Linder, Daniel F ; Rochani, Haresh ; Yin, Jingjing. / A simpler approach for mediation analysis for dichotomous mediators in logistic regression. In: Journal of Statistical Computation and Simulation. 2018 ; Vol. 88, No. 6. pp. 1211-1227.
@article{ac0e3d747e324bc6bc0731ed104c09a4,
title = "A simpler approach for mediation analysis for dichotomous mediators in logistic regression",
abstract = "Mediation is a hypothesized causal chain among three variables. Mediation analysis for continuous response variables is well developed in the literature, and it can be shown that the indirect effect is equal to the total effect minus the direct effect. However, mediation analysis for categorical responses is still not fully developed. The purpose of this article is to propose a simpler method of analysing the mediation effect among three variables when the dependent and mediator variables are both dichotomous. We propose using the latent variable technique which in turn will adjust for the necessary condition that indirect effect is equal to the total effect minus the direct effect. An intensive simulation study is conducted to compare the proposed method with other methods in the literature. Our theoretical derivation and simulation study show that the proposed approach is simpler to use and at least as good as other approaches provided in the literature. We illustrate our approach to test for the potential mediators on the relationship between depression and obesity among children and adolescents compared to the method in Winship and Mare using National children health survey data 2011–2012.",
keywords = "Causal inference, childhood obesity, indirect effect, latent variables, logistic regression, mediation analysis, national children health survey",
author = "Hani Samawi and Jingxian Cai and Linder, {Daniel F} and Haresh Rochani and Jingjing Yin",
year = "2018",
month = "4",
day = "13",
doi = "10.1080/00949655.2018.1426762",
language = "English (US)",
volume = "88",
pages = "1211--1227",
journal = "Journal of Statistical Computation and Simulation",
issn = "0094-9655",
publisher = "Taylor and Francis Ltd.",
number = "6",

}

TY - JOUR

T1 - A simpler approach for mediation analysis for dichotomous mediators in logistic regression

AU - Samawi, Hani

AU - Cai, Jingxian

AU - Linder, Daniel F

AU - Rochani, Haresh

AU - Yin, Jingjing

PY - 2018/4/13

Y1 - 2018/4/13

N2 - Mediation is a hypothesized causal chain among three variables. Mediation analysis for continuous response variables is well developed in the literature, and it can be shown that the indirect effect is equal to the total effect minus the direct effect. However, mediation analysis for categorical responses is still not fully developed. The purpose of this article is to propose a simpler method of analysing the mediation effect among three variables when the dependent and mediator variables are both dichotomous. We propose using the latent variable technique which in turn will adjust for the necessary condition that indirect effect is equal to the total effect minus the direct effect. An intensive simulation study is conducted to compare the proposed method with other methods in the literature. Our theoretical derivation and simulation study show that the proposed approach is simpler to use and at least as good as other approaches provided in the literature. We illustrate our approach to test for the potential mediators on the relationship between depression and obesity among children and adolescents compared to the method in Winship and Mare using National children health survey data 2011–2012.

AB - Mediation is a hypothesized causal chain among three variables. Mediation analysis for continuous response variables is well developed in the literature, and it can be shown that the indirect effect is equal to the total effect minus the direct effect. However, mediation analysis for categorical responses is still not fully developed. The purpose of this article is to propose a simpler method of analysing the mediation effect among three variables when the dependent and mediator variables are both dichotomous. We propose using the latent variable technique which in turn will adjust for the necessary condition that indirect effect is equal to the total effect minus the direct effect. An intensive simulation study is conducted to compare the proposed method with other methods in the literature. Our theoretical derivation and simulation study show that the proposed approach is simpler to use and at least as good as other approaches provided in the literature. We illustrate our approach to test for the potential mediators on the relationship between depression and obesity among children and adolescents compared to the method in Winship and Mare using National children health survey data 2011–2012.

KW - Causal inference

KW - childhood obesity

KW - indirect effect

KW - latent variables

KW - logistic regression

KW - mediation analysis

KW - national children health survey

UR - http://www.scopus.com/inward/record.url?scp=85041011658&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85041011658&partnerID=8YFLogxK

U2 - 10.1080/00949655.2018.1426762

DO - 10.1080/00949655.2018.1426762

M3 - Article

VL - 88

SP - 1211

EP - 1227

JO - Journal of Statistical Computation and Simulation

JF - Journal of Statistical Computation and Simulation

SN - 0094-9655

IS - 6

ER -