Median-based overlap analysis for single case data: a second study.
IRD is the strongest quick nonoverlap index for single-case graphs.
01Research in Context
What this study did
The authors ran a second test of three quick ways to score single-case graphs. They wanted to know which index best matches what experts see.
They compared PEM, PND, and IRD on a big set of AB graphs. Visual analysts served as the truth check.
What they found
IRD lined up with expert eyes better than PEM or PND. It passed both the math tests and the look test.
PEM beat PND, but IRD still came out on top.
How this fits with other research
Campbell (2004) looked at four older indices and warned that different formulas can tell different stories. Cohn et al. (2007) now give us a clear winner: pick IRD.
Gaily et al. (1998) backed PND meta-analysis, but the new data say IRD is the safer choice today.
Wolfe et al. (2016) show experts barely agree on tough graphs. Using IRD gives you a steady number when eyes disagree.
Why it matters
Next time you need a fast effect score for an AB graph, compute IRD instead of PND. It takes the same two minutes but lines up better with expert judgment and holds up across stats and visual checks. Your reports get clearer and your decisions get easier.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Swap your old PND calculator for IRD on the next single-case graph you score.
02At a glance
03Original abstract
This article takes a further look at the percentage of data points exceeding the median (PEM) analysis method for single-case research data, first presented in this journal by Hsen-Hsing Ma. Ma examined the relationship between PEM and the established percentage of nonoverlapping data (PND) and then applied PEM in a meta-analysis of 61 data sets, correlating their authors' judgments of intervention effectiveness with PEM. The present article covers PEM's historical and statistical context and then applies the new measure in a field test with 165 contrasts between a baseline phase A and a treatment phase B. For comparison, Pearson r , Kruskal-Wallis W, PND, and IRD (improvement rate difference) indices also are calculated and correlated with PEM, and all distributions are examined. Expert visual analysis ratings of the 165 graphs are correlated with all indices. PEM surpassed PND in its validation by other established measures. However, PEM was weaker in distribution shape and visual judgment validation. More strongly validated than either PEM or PND was the new nonparametric measure, IRD.
Behavior modification, 2007 · doi:10.1177/0145445507303452