The relationship between visual analysis and five statistical analyses in a simple AB single-case research design.
Stats and eyes can clash in AB graphs—check autocorrelation and use IRD or BC-SMD to stay safe.
01Research in Context
What this study did
The team showed 35 made-up AB graphs to 45 behavior analysts.
Each expert judged whether the graph showed a real change.
The study then ran five common stats on the same graphs.
They wanted to see which numbers matched the experts’ eyes.
What they found
Visual calls and the five stats often disagreed.
Autocorrelation in the data made the mismatch worse.
The paper warns: check for autocorrelation before trusting any number.
How this fits with other research
Howard et al. (2019) later ran the same test on multiple-baseline graphs.
They found IRD and BC-SMD now line up best with visual calls.
This extends the 2006 warning into newer, better indices.
Lanovaz et al. (2017) add a fix: collect at least 3 baseline and 5 treatment points.
Doing so keeps false positives low when you use the dual-criteria method.
Why it matters
Your eyes can still over- or under-call an effect.
Run a quick autocorrelation check first.
If the data are sticky, lean on IRD or BC-SMD, not older stats.
Add more data points when you can.
These two moves make your visual call safer and clearer.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Before you judge any AB graph, paste the data into a free IRD calculator and glance at the autocorrelation value.
02At a glance
03Original abstract
This study explored some practical issues for single-case researchers who rely on visual analysis of graphed data, but who also may consider supplemental use of promising statistical analysis techniques. The study sought to answer three major questions: (a) What is a typical range of effect sizes from these analytic techniques for data from "effective interventions"? (b) How closely do results from these same analytic techniques concur with visual-analysis-based judgments of effective interventions? and (c) What role does autocorrelation play in interpretation of these analytic results? To answer these questions, five analytic techniques were compared with the judgments of 45 doctoral students and faculty, who rated intervention effectiveness from visual analysis of 35 fabricated AB design graphs. Implications for researchers and practitioners using single-case designs are discussed.
Behavior modification, 2006 · doi:10.1177/0145445503261167