Toward cross‐disciplinary translation of the testing effect: A systematic replication
Five quiz questions are enough to raise college students' exam scores and class engagement.
01Research in Context
What this study did
Glodowski and colleagues ran a single-case experiment with college students. They gave three quiz lengths: zero, five, or ten questions before each exam.
The team tracked exam scores and class participation. Each student got all three quiz sizes in a rotating pattern.
What they found
Any quiz beat no quiz. Five questions lifted scores and raised hand-raising as much as ten.
Adding more items past five gave no extra payoff. The testing effect showed up fast and plateaued.
How this fits with other research
Méndez (2024) used the same single-case logic with undergraduates. He showed that rules plus richer payoff, not rules alone, steer choice. Both papers push behavior analysts to test cognitive claims with within-subject designs.
Barnard-Brak et al. (2022) and Manolov et al. (2022) argue we should pick effect measures before data collection. Glodowski et al. followed that advice, giving the field a clean, pre-planned replication of the testing effect.
Young (2019) offers free Monte-Carlo software to judge SCED data. Pairing that tool with Glodowski’s brief quiz protocol could speed up future classroom replications.
Why it matters
If you teach college courses or supervise RBT coursework, swap one review lecture for a five-item quiz. You will likely see the same boost in test scores and participation without extra prep time. The study also shows behavior analysts can export our design standards to education research, making cognitive findings easier to trust and use.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Add a five-question quiz at the start of your next training module and track quiz-to-exam gains.
02At a glance
03Original abstract
The testing effect is a well‐established phenomenon in cognitive psychology that refers to enhanced long‐term retention of information due to active recalling through testing. Following a cross‐disciplinary translation of the testing effect into behavioral principles, we systematically replicated the previous findings in a behavior‐analytic context while evaluating the effects of the number of quiz questions on college students' exam performance and other academic behaviors. Students in an upper‐level behavior analysis course participated. Using a within‐subject experimental design in which the participants served as their own control, we compared their exam performance and academic behaviors, such as class participation and out‐of‐class studying, across three conditions: (a) no quizzes, (b) 5‐question quizzes, and (c) 10‐question quizzes. Quizzes improved exam performance and some academic behaviors, successfully replicating the testing effect in a behavior‐analytic context. However, the number of quiz questions did not influence the improvements. Implications of the results and future research are discussed.
Journal of Applied Behavior Analysis, 2025 · doi:10.1002/jaba.70012