Modern statistical practices in the experimental analysis of behavior: An introduction to the special issue

Young (2019) wrote the lead paper for a special issue on better stats for single-case work.

He gives readers a free Shiny app that runs Monte Carlo tests right on your ABAB graphs.

The tool spits out p-values that fit how behavior analysts think about level, trend, and overlap.

The paper is a how-to guide, not an experiment, so no new data are reported.

The takeaway is the app itself: you upload your graph, click run, and get a clear-cut p-value.

Barnard-Brak et al. (2022) updated the conversation in a later special section. They translate even more fancy stats into plain English, building on Young’s 2019 starter kit.

Ninci (2023) turns the same Monte Carlo idea into day-to-day practice rules. She tells BCBAs exactly how many phase changes you need and what to eyeball first, so you don’t mis-read your own graph.

Manolov et al. (2022) give a flowchart for picking the right effect measure before you start. Use their chart first, then plug your data into Young’s app to test it.

Tanious et al. (2020) offer two new consistency numbers for ABAB designs. You can run their CONDAP and CONEFF first, then let Young’s Monte Carlo tool give you a p-value for the same data set.

You no longer have to guess if your treatment graph “looks” better. Download the free app, upload your phase data, and get a p-value that reviewers will accept. Pair it with the newer checklists from Ninci (2023) and Manolov et al. (2022) and you have a full statistical shield against visual bias.