Sensitivity Analysis of the No-Omitted Confounder Assumption in Latent Growth Curve Mediation Models

Davood Tofighi, Yu Yu Hsiao, Eric S. Kruger, David Mackinnon, M. Lee Van Horn, Katie Witkiewitz

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Latent growth curve mediation models are increasingly used to assess mechanisms of behavior change. For latent growth mediation model, like any another mediation model, even with random treatment assignment, a critical but untestable assumption for valid and unbiased estimates of the indirect effects is that there should be no omitted variable that confounds indirect effects. One way to address this untestable assumption is to conduct sensitivity analysis to assess whether the inference about an indirect effect would change under varying degrees of confounding bias. We developed a sensitivity analysis technique for a latent growth curve mediation model. We compute the biasing effect of confounding on point and confidence interval estimates of the indirect effects in a structural equation modeling framework. We illustrate sensitivity plots to visualize the effects of confounding on each indirect effect and present an empirical example to illustrate the application of the sensitivity analysis.

Original languageEnglish (US)
JournalStructural Equation Modeling
DOIs
StateAccepted/In press - Jan 1 2018

Fingerprint

Growth Curve
Mediation
Sensitivity analysis
mediation
Sensitivity Analysis
Confounding
Model
Structural Equation Modeling
Growth curve
Indirect effects
Estimate
Confidence interval
Assignment
confidence
Valid
trend

Keywords

  • correlated augmented model
  • indirect effect
  • latent growth curve
  • Multiple mediation analysis
  • sensitivity analysis

ASJC Scopus subject areas

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

Cite this

Sensitivity Analysis of the No-Omitted Confounder Assumption in Latent Growth Curve Mediation Models. / Tofighi, Davood; Hsiao, Yu Yu; Kruger, Eric S.; Mackinnon, David; Van Horn, M. Lee; Witkiewitz, Katie.

In: Structural Equation Modeling, 01.01.2018.

Research output: Contribution to journalArticle

Tofighi, Davood ; Hsiao, Yu Yu ; Kruger, Eric S. ; Mackinnon, David ; Van Horn, M. Lee ; Witkiewitz, Katie. / Sensitivity Analysis of the No-Omitted Confounder Assumption in Latent Growth Curve Mediation Models. In: Structural Equation Modeling. 2018.
@article{b09a86e86ba64874a4ecfef5eedc305d,
title = "Sensitivity Analysis of the No-Omitted Confounder Assumption in Latent Growth Curve Mediation Models",
abstract = "Latent growth curve mediation models are increasingly used to assess mechanisms of behavior change. For latent growth mediation model, like any another mediation model, even with random treatment assignment, a critical but untestable assumption for valid and unbiased estimates of the indirect effects is that there should be no omitted variable that confounds indirect effects. One way to address this untestable assumption is to conduct sensitivity analysis to assess whether the inference about an indirect effect would change under varying degrees of confounding bias. We developed a sensitivity analysis technique for a latent growth curve mediation model. We compute the biasing effect of confounding on point and confidence interval estimates of the indirect effects in a structural equation modeling framework. We illustrate sensitivity plots to visualize the effects of confounding on each indirect effect and present an empirical example to illustrate the application of the sensitivity analysis.",
keywords = "correlated augmented model, indirect effect, latent growth curve, Multiple mediation analysis, sensitivity analysis",
author = "Davood Tofighi and Hsiao, {Yu Yu} and Kruger, {Eric S.} and David Mackinnon and {Van Horn}, {M. Lee} and Katie Witkiewitz",
year = "2018",
month = "1",
day = "1",
doi = "10.1080/10705511.2018.1506925",
language = "English (US)",
journal = "Structural Equation Modeling",
issn = "1070-5511",
publisher = "Psychology Press Ltd",

}

TY - JOUR

T1 - Sensitivity Analysis of the No-Omitted Confounder Assumption in Latent Growth Curve Mediation Models

AU - Tofighi, Davood

AU - Hsiao, Yu Yu

AU - Kruger, Eric S.

AU - Mackinnon, David

AU - Van Horn, M. Lee

AU - Witkiewitz, Katie

PY - 2018/1/1

Y1 - 2018/1/1

N2 - Latent growth curve mediation models are increasingly used to assess mechanisms of behavior change. For latent growth mediation model, like any another mediation model, even with random treatment assignment, a critical but untestable assumption for valid and unbiased estimates of the indirect effects is that there should be no omitted variable that confounds indirect effects. One way to address this untestable assumption is to conduct sensitivity analysis to assess whether the inference about an indirect effect would change under varying degrees of confounding bias. We developed a sensitivity analysis technique for a latent growth curve mediation model. We compute the biasing effect of confounding on point and confidence interval estimates of the indirect effects in a structural equation modeling framework. We illustrate sensitivity plots to visualize the effects of confounding on each indirect effect and present an empirical example to illustrate the application of the sensitivity analysis.

AB - Latent growth curve mediation models are increasingly used to assess mechanisms of behavior change. For latent growth mediation model, like any another mediation model, even with random treatment assignment, a critical but untestable assumption for valid and unbiased estimates of the indirect effects is that there should be no omitted variable that confounds indirect effects. One way to address this untestable assumption is to conduct sensitivity analysis to assess whether the inference about an indirect effect would change under varying degrees of confounding bias. We developed a sensitivity analysis technique for a latent growth curve mediation model. We compute the biasing effect of confounding on point and confidence interval estimates of the indirect effects in a structural equation modeling framework. We illustrate sensitivity plots to visualize the effects of confounding on each indirect effect and present an empirical example to illustrate the application of the sensitivity analysis.

KW - correlated augmented model

KW - indirect effect

KW - latent growth curve

KW - Multiple mediation analysis

KW - sensitivity analysis

UR - http://www.scopus.com/inward/record.url?scp=85053504939&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85053504939&partnerID=8YFLogxK

U2 - 10.1080/10705511.2018.1506925

DO - 10.1080/10705511.2018.1506925

M3 - Article

AN - SCOPUS:85053504939

JO - Structural Equation Modeling

JF - Structural Equation Modeling

SN - 1070-5511

ER -