From Percentages to Precision: Using Response Rates to Advance Analyses of Procedural Fidelity

In some domains of behavior analysis, summarizing data as a percentage is nearly ubiquitous. This is certainly the case when behavior analysts report data about procedural fidelity (the extent to which procedures are implemented as designed); fidelity data were reported solely as percentage in 423 of 425 recent studies published in the Journal of Applied Behavior Analysis. In this article, we critically examine the use of percentage, especially in the context of analyzing procedural-fidelity data. We demonstrate how exclusive reliance on percentage can obscure important nuances in fidelity data and how adding response rate as a metric offers a more precise understanding. To illustrate our points, we include reanalyzed data from a recent evaluation of procedural fidelity in public schools. We conclude with practical recommendations for adopting rate as a metric in the analysis of procedural-fidelity data, thereby building on contributions of notable behavior analysts like Henry Pennypacker, who prioritized continuous, dimensional approaches to measurement.

The online version contains supplementary material available at 10.1007/s40614-025-00433-9.