Growth under intervention by means of instructional time expended: Empirical illustrations of applicable models.

Abstract

The empirical literature on academic intervention has increasingly promoted comparative research, moving the field forward in addressing the question, "What works best?" Poncy et al. (2015), Skinner (2008, 2010), and Skinner, Fletcher, and Henington (1996) have suggested that researchers enhance traditional evaluations of learning (behavior change) by using learning rates (behavior change over instructional time [IT]). The precise documentation or control of IT allows researchers to address confounds commonly found in comparative intervention research, most notably variability in intervention duration. The current article demonstrates how recent developments within the field of single-case analysis creates an opportunity to address Skinner's (2008, 2010), Skinner et al. (1996), and Poncy et al. (2015) IT within the context of comparative intervention research. Specifically, a rationale and tutorial on several coding schemes for the generalized linear model is presented that can be used to isolate the effect of IT. The use of these analyses is demonstrated across design frameworks and when considering individual participants, studies, and research synthesis. (PsycINFO Database Record (c) 2019 APA, all rights reserved).