Indexing effects in single-case experimental designs by estimating the percent of goal obtained

Abstract

Single-case experimental designs are often used to examine the efficacy and effectiveness of interventions that are designed to change behaviors so that they are less problematic and more similar to the behavior of typically developing peers. By defining a goal level for the behavior, we were able to develop an effect size based on the percent of goal obtained (PoGO). We discuss methods for determining the goal and methods of estimating PoGO, including those for when there are and are not trends. Applicability is demonstrated using data from single-case experimental design studies of speech production, early literacy, and social skill interventions in the early childhood literature and by comparing PoGO to graphic displays and other summary measures, including Tau-U, NAP, the standardized mean difference, and the log-response ratio. We consider how PoGO could be used in systematic reviews, provide recommendations, and discuss areas in need of research.