An Evaluation of Programmed Procedural Fidelity Errors on Data‐Based Decision‐Making During Precision Teaching
You can let data-decision fidelity slip to 50% and still see tiny short-term math gains, but do not count on it lasting.
01Research in Context
What this study did
Diffley et al. (2025) asked what happens when you cut data-based decision fidelity in half during Precision Teaching.
They used Morningside Math Facts with neurotypical students and tracked short-term gains.
The study was quasi-experimental and looked at math addition and subtraction skills.
What they found
Students still made small, short-term math gains even when fidelity dropped to 50%.
The gains were weak and did not last long.
Larger studies are needed to know if this holds up over time.
How this fits with other research
Clayton et al. (2019) showed a 10-minute BST jump can push DTT fidelity to 97% and keep it there for a month. Diffley et al. flip the question: what if you let fidelity slip on purpose?
Maffei-Almodovar et al. (2017) used BST to teach teachers to make accurate data decisions from graphs. Diffley et al. now test what happens when those decisions are only half-right.
Katechis et al. (2026) trained RBTs to spot fidelity errors with fluency drills. Together these papers form a loop: train, detect, and now test the cost of loosening standards.
Why it matters
If you run Precision Teaching in a busy classroom, you now know you can drop data-based decision fidelity to 50% for a day or two without losing all progress. Do not make it a habit; the gains are small and fade fast. Use this wiggle room for crisis days, then return to full fidelity as soon as you can.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Track your data decisions today; if you must skip half, do it, but plan to return to full checks tomorrow.
02At a glance
03Original abstract
ABSTRACT Research employing parametric analyses of fidelity have demonstrated that some behavior analytic procedures maintain their effectiveness when procedural fidelity is reduced. To date no research has experimentally explored the impact of procedural fidelity errors in Precision Teaching. Thus, the current study aimed to explore the impact of fidelity errors on the “decide” component of the Precision Teaching system. Specifically, the impact of omitting data‐based decision‐making on (a) celeration, (b) bounce or variability, and (c) number of frequency aims achieved. A between groups parametric analysis was employed within which five levels of programmed fidelity (100%, 75%, 50%, 25%, and 0%) were explored. Five mainstream second and third grade classrooms in educationally disadvantaged schools participated. Teachers implemented Precision Teaching and Frequency Building with the Morningside Math Facts Addition and Subtraction curriculum with programmed errors incorporated at differing levels across classrooms. Results suggest that learners can still demonstrate acceptable performance when fidelity on data‐based decision‐making is reduced in the short term. However, more research is warranted with larger sample sizes to ascertain more conclusive findings.
Behavioral Interventions, 2025 · doi:10.1002/bin.70039