A Comparison of Single-Case Effect Measures Using Check-In Check-Out Data.

The team looked at every Check-In Check-Out single-case study they could find. They pulled out seven common ways to measure how well CICO works.

They ran the same numbers through each formula. The goal was to see if one method would call CICO "huge" while another called it "tiny."

All seven formulas agreed CICO helps. Yet the size of the benefit jumped from "small" to "large" just by picking a different ruler.

That means two BCBAs could read the same kid’s chart and walk away with opposite ideas about how strong the intervention is.

Sottilare et al. (2023) is one of the very CICO studies inside this meta. Their classroom data look great, but the new paper shows the same data could be sold as either a medium or a huge win depending on the formula you pick.

McIntyre et al. (2017) warned us that CICO evidence needs tighter standards. Andrews et al. (2024) answers that call by showing the standards must include picking one effect measure and sticking to it.

Veenman et al. (2018) found small benefits for other classroom programs. Their work used group designs, so the size jump seen here may be a single-case issue, not a CICO issue.

If you write an evaluation report, pick one effect measure and name it. Tell the IEP team whether you used Tau-U, PND, or another ruler so they know why your number differs from last year’s. When you read someone else’s CICO study, flip to the method section first. If they used a ruler that tends to inflate size, take the "large" claim with a grain of salt. Consistency beats sparkle.