Statistical inference in behavior analysis: Useful friend.
Simple stats can sharpen your single-case graphs and open doors to grants and wider journals.
01Research in Context
What this study did
The author wrote a think-piece, not an experiment. He asked one question: can numbers and graphs live together?
He looked at how single-subject researchers judge effects by eye. Then he showed how small, simple stats can make those judgments sharper.
What they found
Stats are not the enemy. A quick confidence interval or effect-size can back up what you already see in the graph.
Using these tools helps you win grants and get published outside behavior-analysis journals.
How this fits with other research
Kyonka et al. (2019) counted every article in JEAB from 1992-2017. They found slow growth in stats like NHST and error bars, just as the 1999 paper urged.
Davis et al. (1994) showed that basic and applied journals almost never cite each other. Adding stats can bridge that gap by speaking a language both sides understand.
Bechtel (2005) also pushed for better methods, but focused on mid-level theory instead of numbers. Both papers want the field to grow up without losing its roots.
Why it matters
You already plot data. Add one extra step: drop a large share confidence bands around your phase means or compute an effect-size like Tau-U. It takes five minutes in free software. Reviewers trust the numbers, families understand the graph, and insurance reviewers see a format they recognize. Keep the visual analysis you love—just let stats serve as your friendly witness.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Compute one Tau-U for your last AB graph and add the value to the caption.
02At a glance
03Original abstract
Single-subject and statistical inference are virtually identical. With both techniques change is inferred when variability across conditions is sufficiently large to accommodate variability within conditions, replication is the final arbiter of whether change is likely to occur by chance, a large effect size is preferred to a small consistent difference, there are similar threats to internal validity, and generalizability of results is valued. Knowing how to use statistical inferential procedures would make behavior analysts more methodologically sophisticated. It would also help them to critically evaluate research in other areas of psychology, obtain research grants, and publish their research in diverse outlets, which would help others to see behavior-analytic work.
The Behavior analyst, 1999 · doi:10.1007/BF03391987