TY - JOUR
T1 - Comparing Models of Change to Estimate the Mediated Effect in the Pretest–Posttest Control Group Design
AU - Valente, Matthew J.
AU - Mackinnon, David
N1 - Funding Information:
This research was supported in part by the National Institute on Drug Abuse grant number R37 DA09757.
Publisher Copyright:
Copyright © Taylor & Francis Group, LLC.
PY - 2017/5/4
Y1 - 2017/5/4
N2 - Models to assess mediation in the pretest–posttest control group design are understudied in the behavioral sciences even though it is the design of choice for evaluating experimental manipulations. The article provides analytical comparisons of the four most commonly used models to estimate the mediated effect in this design: analysis of covariance (ANCOVA), difference score, residualized change score, and cross-sectional model. Each of these models is fitted using a latent change score specification and a simulation study assessed bias, Type I error, power, and confidence interval coverage of the four models. All but the ANCOVA model make stringent assumptions about the stability and cross-lagged relations of the mediator and outcome that might not be plausible in real-world applications. When these assumptions do not hold, Type I error and statistical power results suggest that only the ANCOVA model has good performance. The four models are applied to an empirical example.
AB - Models to assess mediation in the pretest–posttest control group design are understudied in the behavioral sciences even though it is the design of choice for evaluating experimental manipulations. The article provides analytical comparisons of the four most commonly used models to estimate the mediated effect in this design: analysis of covariance (ANCOVA), difference score, residualized change score, and cross-sectional model. Each of these models is fitted using a latent change score specification and a simulation study assessed bias, Type I error, power, and confidence interval coverage of the four models. All but the ANCOVA model make stringent assumptions about the stability and cross-lagged relations of the mediator and outcome that might not be plausible in real-world applications. When these assumptions do not hold, Type I error and statistical power results suggest that only the ANCOVA model has good performance. The four models are applied to an empirical example.
KW - longitudinal mediation
KW - mediation
KW - pretest–posttest design
KW - two waves
UR - http://www.scopus.com/inward/record.url?scp=85011838352&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85011838352&partnerID=8YFLogxK
U2 - 10.1080/10705511.2016.1274657
DO - 10.1080/10705511.2016.1274657
M3 - Article
AN - SCOPUS:85011838352
SN - 1070-5511
VL - 24
SP - 428
EP - 450
JO - Structural Equation Modeling
JF - Structural Equation Modeling
IS - 3
ER -