Date of Award
Spring 2026
Document Type
Open Access Thesis
Department
Psychology
First Advisor
Sarfaraz Serang
Abstract
Randomized control trials (RCTs) are considered the gold standard experimental design in intervention research. Randomized pretest, posttest, follow-up (RPPF) designs are a special case of RCT in which participants are randomized to a treatment or control group and measured at a baseline measurement occasion, then measured at two subsequent timepoints to assess the initial efficacy of an intervention and the maintenance of its effect. Several methods, including univariate linear models and structural equation models (SEMs), have been proposed to analyze data from this type of design, including the autoregressive latent change score model (ARLCSM) which was developed to address some of the limitations associated with other SEM approaches. Beyond treatment effects, however, a researcher may be interested in the circumstances under which a treatment is effective – a question which can be addressed through moderation analysis. The present paper proposes two variations of the ARLCSM to assess moderated treatment effects and evaluates their performances compared to another popular approach for assessing moderation, analysis of covariance (ANCOVA), via a simulation study. Simulation results suggest, across conditions, both variations of the ARLCSM are more powerful than the ANCOVA in detecting moderated treatment effects and have acceptable levels of type I error rate and parameter bias. The models were then fit to a real empirical dataset to illustrate the ARLCSM’s utility in applied research contexts.
Rights
© 2026, Annabelle Henley Reese
Recommended Citation
Reese, A. H.(2026). An Autoregressive Latent Change Score Model to Assess Moderation in Randomized Pretest, Posttest, Follow-up Designs. (Master's thesis). Retrieved from https://scholarcommons.sc.edu/etd/8944