Effect sizes as result interpretation aids in single‐subject experimental research: description and application of four nonoverlap methods

Abstract

Single‐subject experimental research (SSER), one of the most commonly used research methods in special education and applied behaviour analysis, is a scientific, rigorous and valid method to evaluate the effectiveness of behavioural, educational and psychological treatments. However, studies using single‐subject experimental research designs are often excluded from meta‐analyses of evidence‐based practices due to the lack of methodological consensus on the type of effect size indices to be used to determine treatment effect. To promote the use of effect size indices as an adjunct to visual analysis, this article describes four nonoverlap methods (PND, IRD, PEM‐T and Tau‐U) and demonstrates their application to data obtained from studies employing different SSER designs. Advantages and disadvantages of each method are highlighted and considerations for selecting the most appropriate method are provided for researchers and practitioners.

Systematic Review of Functional Analysis and Treatment of Elopement (2000–2015)

“…Effect Sizes There is considerable debate regarding the validity of measures of effect size (e.g., standardized mean difference [SMD], regression-based techniques, overlap methods) with single-subject data (Olive & Franco, 2008;Olive & Smith, 2005;Parker, Vannest & Davis, 2011;Rakap, 2015;Wolery, Busick, Reichow, & Barton, 2010). However, we chose SMD over proposed alternatives (e.g., overlap methods) for several reasons.…”

The Impact of a Mindfulness-Based Intervention on Happiness: A Reflection on the Relevance of Integrating a Positive Psychology Framework within Mindfulness Research in Youth

“…In this study, primary hypotheses were tested using a single case modeling strategy allowing comparisons between A-B-A phases, and their statistical significance. Specifically, statistical analyses were conducted using the Tau-U index, which is a regression technique using a non-overlap approach, while controlling for monotonic trends in the baseline (Parker et al 2011;Rakap 2015). Specifically, the Tau-U index allows for analyses of variations in scores within and between each treatment phase.…”

Teaching Play Skills to Visually Impaired Preschool Children: Its Effect on Social Interaction

“…In recent years, single-subject studies have been supported by non-parametric analyses so that their visual analysis may be easily interpreted through more objective results (Kratochwill et al, 2010;Laushey et al, 2009;Rakap, 2015). In this study, the percentage of non-overlapping data (PND) was found to support the efficiency of the independent variable.…”

“…PND was obtained for each subject by drawing a straight line starting from the top most data point of the baseline level toward the implementation stage and then by dividing the data points of the implementation stage resting above this line by the total number of data points. The resulting number was then multiplied by 100 (Kratochwill et al, 2010;Rakap, 2015).…”