An R Shiny App for Sensitivity Analysis for Latent Growth Curve Mediation

Eric S. Kruger; Davood Tofighi; Yu Yu Hsiao; David P. MacKinnon; M. Lee Van Horn; Katie Witkiewitz

doi:10.1080/10705511.2022.2045203

An R Shiny App for Sensitivity Analysis for Latent Growth Curve Mediation

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

Psychology

Research output: Contribution to journal › Article › peer-review

1 Scopus citations

Abstract

Mechanisms of behavior change are the processes through which interventions are hypothesized to cause changes in outcomes. Latent growth curve mediation models (LGCMM) are recommended for investigating the mechanisms of behavior change because LGCMM models establish temporal precedence of change from the mediator to the outcome variable. The Correlated Augmented Mediation Sensitivity Analyses (CAMSA) App implements sensitivity analysis for LGCMM models to evaluate if a mediating path (mechanism) is robust to potential confounding variables. The CAMSA approach is described and applied to simulated data, and data from a research study exploring a mechanism of change in the treatment of substance use disorder.

Original language	English (US)
Pages (from-to)	944-952
Number of pages	9
Journal	Structural Equation Modeling
Volume	29
Issue number	6
DOIs	https://doi.org/10.1080/10705511.2022.2045203
State	Published - 2022

Keywords

Behavior change
confounding
latent growth curve mediation modeling
mediation
sensitivity analyses

ASJC Scopus subject areas

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

Access to Document

10.1080/10705511.2022.2045203

Cite this

@article{94913f0a6f39428fbebac8e15bf9d7f1,

title = "An R Shiny App for Sensitivity Analysis for Latent Growth Curve Mediation",

abstract = "Mechanisms of behavior change are the processes through which interventions are hypothesized to cause changes in outcomes. Latent growth curve mediation models (LGCMM) are recommended for investigating the mechanisms of behavior change because LGCMM models establish temporal precedence of change from the mediator to the outcome variable. The Correlated Augmented Mediation Sensitivity Analyses (CAMSA) App implements sensitivity analysis for LGCMM models to evaluate if a mediating path (mechanism) is robust to potential confounding variables. The CAMSA approach is described and applied to simulated data, and data from a research study exploring a mechanism of change in the treatment of substance use disorder.",

keywords = "Behavior change, confounding, latent growth curve mediation modeling, mediation, sensitivity analyses",

author = "Kruger, {Eric S.} and Davood Tofighi and Hsiao, {Yu Yu} and MacKinnon, {David P.} and {Lee Van Horn}, M. and Katie Witkiewitz",

note = "Funding Information: We would like to thank local agronomists and extension agents in each country for their help in data collection. We acknowledge the Swiss Agency for Development and Cooperation for financial support to conduct the CORIGAP survey (Grant 681 number 7F-08412.02 to A.M.S.) and also the Bill and Melinda Gates Foundation for their support through the CGIAR Excellence in Agronomy 2030 Incubation Phase (Grant number INV-005431 to P.G.). We also thank the China Postdoctoral Science Foundation (2020M682439 to S.Y.), the China Scholarship Council (201706760015 to S.Y.), the Belt and Road Center for Sustainable Rice Production, the Indonesian Agency for Agricultural Research and Development (IAARD), the Office of Global Engagement at the Institute of Agriculture and Natural Resources (IANR) at the University of Nebraska-Lincoln (UNL) and the UNL Daugherty Water for Food Global Institute. Funding Information: This work was supported by NIAAA R01 [grant number AA025539 to Katie Witkiewitz and Davood Tofighi: MPIs] and NIDA [grant number R37DA09757 to David P. MacKinnon: PI]. Publisher Copyright: {\textcopyright} 2022 Taylor & Francis Group, LLC.",

year = "2022",

doi = "10.1080/10705511.2022.2045203",

language = "English (US)",

volume = "29",

pages = "944--952",

journal = "Structural Equation Modeling",

issn = "1070-5511",

publisher = "Psychology Press Ltd",

number = "6",

}

TY - JOUR

T1 - An R Shiny App for Sensitivity Analysis for Latent Growth Curve Mediation

AU - Kruger, Eric S.

AU - Tofighi, Davood

AU - Hsiao, Yu Yu

AU - MacKinnon, David P.

AU - Lee Van Horn, M.

AU - Witkiewitz, Katie

N1 - Funding Information: We would like to thank local agronomists and extension agents in each country for their help in data collection. We acknowledge the Swiss Agency for Development and Cooperation for financial support to conduct the CORIGAP survey (Grant 681 number 7F-08412.02 to A.M.S.) and also the Bill and Melinda Gates Foundation for their support through the CGIAR Excellence in Agronomy 2030 Incubation Phase (Grant number INV-005431 to P.G.). We also thank the China Postdoctoral Science Foundation (2020M682439 to S.Y.), the China Scholarship Council (201706760015 to S.Y.), the Belt and Road Center for Sustainable Rice Production, the Indonesian Agency for Agricultural Research and Development (IAARD), the Office of Global Engagement at the Institute of Agriculture and Natural Resources (IANR) at the University of Nebraska-Lincoln (UNL) and the UNL Daugherty Water for Food Global Institute. Funding Information: This work was supported by NIAAA R01 [grant number AA025539 to Katie Witkiewitz and Davood Tofighi: MPIs] and NIDA [grant number R37DA09757 to David P. MacKinnon: PI]. Publisher Copyright: © 2022 Taylor & Francis Group, LLC.

PY - 2022

Y1 - 2022

N2 - Mechanisms of behavior change are the processes through which interventions are hypothesized to cause changes in outcomes. Latent growth curve mediation models (LGCMM) are recommended for investigating the mechanisms of behavior change because LGCMM models establish temporal precedence of change from the mediator to the outcome variable. The Correlated Augmented Mediation Sensitivity Analyses (CAMSA) App implements sensitivity analysis for LGCMM models to evaluate if a mediating path (mechanism) is robust to potential confounding variables. The CAMSA approach is described and applied to simulated data, and data from a research study exploring a mechanism of change in the treatment of substance use disorder.

AB - Mechanisms of behavior change are the processes through which interventions are hypothesized to cause changes in outcomes. Latent growth curve mediation models (LGCMM) are recommended for investigating the mechanisms of behavior change because LGCMM models establish temporal precedence of change from the mediator to the outcome variable. The Correlated Augmented Mediation Sensitivity Analyses (CAMSA) App implements sensitivity analysis for LGCMM models to evaluate if a mediating path (mechanism) is robust to potential confounding variables. The CAMSA approach is described and applied to simulated data, and data from a research study exploring a mechanism of change in the treatment of substance use disorder.

KW - Behavior change

KW - confounding

KW - latent growth curve mediation modeling

KW - mediation

KW - sensitivity analyses

UR - http://www.scopus.com/inward/record.url?scp=85127227472&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85127227472&partnerID=8YFLogxK

U2 - 10.1080/10705511.2022.2045203

DO - 10.1080/10705511.2022.2045203

M3 - Article

AN - SCOPUS:85127227472

SN - 1070-5511

VL - 29

SP - 944

EP - 952

JO - Structural Equation Modeling

JF - Structural Equation Modeling

IS - 6

ER -

An R Shiny App for Sensitivity Analysis for Latent Growth Curve Mediation

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Teacher’s Corner: An R Shiny App for Sensitivity Analysis for Latent Growth Curve Mediation

Cite this

An R Shiny App for Sensitivity Analysis for Latent Growth Curve Mediation

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Datasets

Teacher’s Corner: An R Shiny App for Sensitivity Analysis for Latent Growth Curve Mediation

Cite this