Quantitative Techniques and Graphical Representations for Interpreting Results from Alternating Treatment Design
Download the free ATD graph kit and Edgington test to turn your next alternating-treatment data into a clear, defensible picture.
01Research in Context
What this study did
Manolov and colleagues built free tools for analyzing alternating-treatment data. They show how to run Edgington’s randomization test and plug the numbers into ready-made graph templates.
The paper walks through two example ATD sets so you can copy the steps. No new intervention was tested; this is a how-to guide.
What they found
The templates turn messy ATD numbers into clean, decision-ready pictures. Edgington’s test gives a p-value you can report in grant apps or IEP meetings.
All files—Excel sheets, R code, and fill-in graphs—are free to download.
How this fits with other research
Wolfe et al. (2023) also give free software, but for modified Brinley plots that judge replication across kids. Use both tools: ATD templates from Manolov and Brinley plots from Wolfe to cover design and replication checks.
Sunde et al. (2022) supply a visual-inspection checklist for latency-based FAs. Their checklist and Manolov’s templates share the same goal—make visual analysis rules-based instead of eye-balling.
De Los Reyes et al. (2009) meta-analyzed child intervention studies that include ATDs. The new templates let future meta-analysts re-score those older ATD sets with the same metric, tightening future syntheses.
Why it matters
If you run ATDs in classrooms or clinics, you now have a stat test plus court-ready graphs at zero cost. Swap the templates into your next session report and give stakeholders numbers they can trust in under five minutes.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Open the free Excel file, paste last week’s ATD data, and print the auto-generated graph for the parent meeting.
02At a glance
03Original abstract
Multiple quantitative methods for single-case experimental design data have been applied to multiple-baseline, withdrawal, and reversal designs. The advanced data analytic techniques historically applied to single-case design data are primarily applicable to designs that involve clear sequential phases such as repeated measurement during baseline and treatment phases, but these techniques may not be valid for alternating treatment design (ATD) data where two or more treatments are rapidly alternated. Some recently proposed data analytic techniques applicable to ATD are reviewed. For ATDs with random assignment of condition ordering, the Edgington’s randomization test is one type of inferential statistical technique that can complement descriptive data analytic techniques for comparing data paths and for assessing the consistency of effects across blocks in which different conditions are being compared. In addition, several recently developed graphical representations are presented, alongside the commonly used time series line graph. The quantitative and graphical data analytic techniques are illustrated with two previously published data sets. Apart from discussing the potential advantages provided by each of these data analytic techniques, barriers to applying them are reduced by disseminating open access software to quantify or graph data from ATDs.
Perspectives on Behavior Science, 2022 · doi:10.1007/s40614-021-00289-9