Operant learning principles applied to teaching introductory statistics.
Guaranteed A for a large share mastery plus self-pacing can replace homework and still lift college stats grades.
01Research in Context
What this study did
A professor ran a college stats class like a token economy.
Students could earn an A only if they scored a large share or better on every short quiz.
They moved at their own speed and took quizzes as often as they wanted.
No homework was given.
The report follows the students across one semester.
What they found
Every student hit the a large share mark.
No one dropped the class.
End-of-course surveys showed high satisfaction.
The teacher never assigned homework yet grades rose.
How this fits with other research
Lydersen et al. (1974) tested the same logic with five first-grade boys.
When reading accuracy earned tokens, disruption fell from a large share to near zero.
The pattern matches: reinforce academic work, behavior improves.
Rapport et al. (1982) extended the idea to peer tutoring.
Adolescents earned tokens for praising their tutees.
Both tutors and tutees then read more and stayed on task.
Petursdottir et al. (2019) added function-based planning and fading.
Their elementary students kept the gains even after tokens were removed.
Together these studies show the token-academics link from .
Why it matters
You can swap busywork for clear mastery criteria.
Pick the key skills, set a high but reachable bar, and let learners move when ready.
Try one unit with quiz-until-mastery plus tokens for passing.
Track if homework becomes optional and grades still climb.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Pick one skill module, set a a large share quiz pass rule, and let students retake until they hit it—no extra homework.
02At a glance
03Original abstract
A grade of A was given in an introductory statistics course for meeting a set of contingencies that included no work outside of class (except by request), near-perfect performance on exams following each unit of work in a programmed text, correction of all exam errors, self pacing of work, and the chance to finish the course early. A grade of incomplete was given otherwise. Correlations among performance measures failed to show any meaningful relationships between time taken to finish the course, errors made on exams, and errors made in the programmed text. Responses to a five-part questionnaire were overwhelmingly favorable to the course, but did not vary as a function of grade point average, time taken to finish the course, or number of errors made on exams. The uniformly high level of performance, the students' lack of interest in social contact with the instructor during class, and the absence of drop-outs are all attributed to the contingencies employed, chief among which, according to the instructor's judgment and student rankings, were self-pacing, frequent non-punitive exams and a guaranteed grade of A for near-perfect work at every stage.
Journal of applied behavior analysis, 1970 · doi:10.1901/jaba.1970.3-191