Analysis of Covariance, Moderated Regression, Logistic Regression, and Mediation

This chapter examines ANCOVA, regression designs with interactions, logistic regression, and indirect effects (mediation). For ANCOVA, the chapter highlights differences in power comparing ANOVA and ANCOVA approaches. Regression designs with interactions are presented using several estimation procedures with a focus on the challenges associated with detecting interactions. Logistic regression approaches cover use of categorical and continuous predictors as well as designs with multiple predictors. Mediation approaches cover designs with a single mediator and multiple mediators. Additional issues include reliability influences on detection of regression interactions. For each analysis, the chapter presents examples using R functions from the pwr2ppl package.