Difficulty: Intermediate | Prerequisites: key terms from the companion notes (contingent access, noncontingent access, distributed studying, single-subject design, IV/DV). Read those first if these terms are unfamiliar.
These notes walk through three published studies from the Journal of Applied Behavior Analysis, each examining a different way to reduce academic procrastination through behavioural contingencies. Together they form a coherent evidence base: practice quiz schedules, access gating, and grade-point contingencies all shift student behaviour toward earlier, more distributed engagement. This document is a companion to the key-concepts notes. It covers the specifics of each study: who participated, what was manipulated, what was measured, and what the results mean for your exam and for real-world application.
Three JABA studies tested whether structuring when and how students access study materials reduces procrastination. All three found that adding a behavioural contingency (gated quiz access or grade points) shifted study effort earlier and improved performance. The effect was consistent across graduate and undergraduate samples, though students generally preferred the less structured (and less effective) condition.
Reference
Perrin, C. J., Miller, N., Haberlin, A. T., Ivy, J. W., Meindl, J. N., & Neef, N. A. (2011). Measuring and reducing college students' procrastination. Journal of Applied Behavior Analysis, 44(3), 463–474.
Tags: procrastination, practice quizzes, contingent access, noncontingent access, fixed-time schedule, graduate students, distributed studying
11 graduate students (7 doctoral, 4 master's level)
2 male, 9 female
The schedule of online practice quiz delivery, compared across two conditions:
Noncontingent access (baseline): practice quizzes released on a fixed-time 24-hour schedule. One quiz became accessible per day at 6:30 a.m., starting the morning after the class session. All five quizzes accumulated and remained available.
Contingent access (treatment): students had to answer some questions earlier in the week to gain access to additional study questions.
Studying patterns: when during the week students completed practice quizzes (the timing distribution).
Performance on in-class quizzes: percentage correct.
All data collected automatically by the university's computer system server.
Frequency data recorded for practice quiz completions: a quiz counted as completed when the student clicked "submit quiz."
Date and time stamps recorded for each submission.
Quiz scores reported as percentage correct (number of correct responses divided by total possible).
No interobserver agreement (IOA) was reported.
Baseline pattern: participants studied very little at the beginning of the week and crammed toward the end.
Treatment pattern: studying became more evenly distributed across the week as students completed practice quizzes on a rolling basis.
This shift indicated a measurable decrease in procrastination.
Most participants performed better on in-class quizzes during the treatment condition compared to baseline.
The availability of additional practice quizzes earlier in the week encouraged earlier engagement, increased preparedness, and reduced the motivation for last-minute cramming.
40% of participants preferred practice quizzes over other study methods.
80% reported feeling more prepared for the in-class quiz because of the practice quizzes.
90% preferred noncontingent access (the condition that gave them more freedom, even though contingent access often produced better outcomes).
The procedures can be replicated in other school settings. The online delivery mechanism makes it adaptable to most courses that use a learning management system.
Reference
Bird, Z., & Chase, P. N. (2021). Student pacing in a master's level course: Procrastination, preference, and performance. Journal of Applied Behavior Analysis, 54(3), 1220–1234.
Tags: procrastination, contingent access, noncontingent access, preference assessment, single-subject design, hybrid course, graduate students, forced-choice, free-choice
10 participants (9 female, 1 male)
All pursuing a master's degree in applied behaviour analysis
Enrolled in a 13-week hybrid graduate-level course (5 in-person classes, remainder online via video conferencing)
Informed consent obtained; ethical guidelines followed
The type of access to study materials: contingent access vs. noncontingent access (same conceptual manipulation as Perrin et al., applied in a different setting and with a preference component).
Procrastination behaviour: the tendency or frequency with which students delayed studying activities.
Distributed studying behaviour: how evenly study effort was spread across the available time.
Preference for access condition: which schedule students chose when given the option, measured through a choice procedure and an end-of-semester survey.
A single-subject design with three phases:
Random assignment: participants assigned to two groups so each experienced both conditions.
Forced-choice exposure (Weeks 1–2): one group started with noncontingent access, the other with contingent access. In Week 2, they switched. This ensured all participants had direct experience of both conditions.
Free-choice preference procedure (Week 3 onward): based on Lockhart et al. (1975). Each week, students chose which condition to use. They could choose the same condition for a maximum of three consecutive weeks, after which a forced-choice exposure to the opposite condition was implemented for one week.
No interobserver agreement (IOA) was reported.
A functional relationship was suggested between the access condition and practice quiz completion.
The contingent access condition generally produced higher percentages of completed practice quizzes (a positive trend).
Individual variability was present: some participants showed consistent differences between conditions, others did not.
Performance outcomes also varied by participant, with some showing higher gains in contingent access and others in noncontingent access.
Overall, the study provides evidence that contingent access reduces procrastination and increases quiz completion among graduate students.
The findings offer practical strategies for educators and practitioners: structuring access to study materials can improve study habits and reduce procrastination. Considering student preferences when designing interventions is important, as the most effective condition may not be the one students prefer.
The procedures can be adapted to undergraduate courses, secondary schools, and online learning environments. Procrastination is not limited to graduate students, and the contingent-access model is compatible with most learning management systems.
Reference
Ryan, C. S., & Hemmes, N. S. (2005). Effects of the contingency for homework submission on work submission and quiz performance in a college course. Journal of Applied Behavior Analysis, 38(1), 79–88.
Tags: homework contingency, points condition, no-points condition, procrastination, homework submission, quiz performance, undergraduate, extrinsic reinforcement
19 undergraduate college students (17 female, 2 male)
Ages 19 to 47
All psychology majors in an advanced research course
Class met three times per week, approximately 110 minutes per meeting
No dropouts during the semester
The points condition: whether a homework assignment carried grade points (scored on a 1–5 scale) or no points. The condition was randomly assigned per chapter per student using a table of random numbers.
Whether students submitted homework assignments in the points vs. no-points condition (submission rate).
Quiz performance (mean quiz grade) in relation to whether the corresponding homework carried points.
10 textbook chapters, each with an associated homework assignment.
Assignments graded on a 1–5 scale in the points condition; scored but not worth marks in the no-points condition.
Students received feedback on all assignments. In the points condition, corrective feedback required students to revise and resubmit to earn the full 5 points.
Quizzes and homework graded against a standardised answer key.
Researchers tracked on-time submission rates across the semester and calculated mean values.
Mean quiz grades compared across points and no-points conditions.
No interobserver agreement (IOA) was reported.
Participants in the points condition submitted a much higher percentage of homework assignments than those in the no-points condition.
In the no-points condition, submissions declined to near-zero levels after the second chapter assignment.
Quiz scores were also higher for chapters where the homework carried points.
A correlation was observed between homework submission and higher quiz grades.
Graded, point-based homework helps students procrastinate less and engage more with their coursework. Assignments without grade consequences tend to be ignored. The implication for students: if a class does not assign points for preparatory work, extra self-discipline is needed to avoid falling behind.
The procedure took place in an urban college setting over a full semester. It could be applied across universities, and the logic extends to any educational context where homework completion is optional or ungraded.
Feature | Perrin et al. (2011) | Bird & Chase (2021) | Ryan & Hemmes (2005) |
|---|---|---|---|
Participants | 11 graduate (7 doctoral, 4 master's) | 10 graduate (all master's ABA) | 19 undergraduate (psych majors) |
IV | Quiz delivery schedule (contingent vs. noncontingent access) | Access type (contingent vs. noncontingent) | Points condition (points vs. no points on homework) |
DV | Study timing, quiz performance | Procrastination, distributed studying, preference | Homework submission rate, quiz scores |
Design | Not specified (server-collected data) | Single-subject with forced- and free-choice phases | Random assignment of conditions per chapter per student |
IOA | No | No | No |
Key finding | Contingent access distributed studying and improved quiz scores | Contingent access increased quiz completion; individual variability in performance | Points condition produced far higher submission rates and quiz scores |
Students often confuse the IV and DV across these studies. The IV is what the researcher changes (schedule type, points condition). The DV is what the researcher measures (study timing, submission rates, quiz scores). If your exam asks you to identify them, trace the manipulation back to the researcher's action.
"Noncontingent" does not mean "no quizzes." It means quizzes are delivered regardless of the student's behaviour. All students in the noncontingent condition still had full access to practice material.
The Bird and Chase (2021) study is sometimes mistaken for a group design because it used random assignment. The random assignment was only to determine the order of condition exposure. The study itself is a single-subject design: each participant's behaviour is analysed individually over time.
Ryan and Hemmes (2005) did not test whether students learned more from doing homework. They tested whether the presence of grade points changed submission behaviour and correlated quiz performance. The homework content was the same in both conditions.
⚠️ Be ready to compare and contrast the three studies on any dimension: participants, IV, DV, design, or findings. The comparison table above is your quick reference.
⚠️ All three studies lack IOA. If asked to critique the studies, this is a valid methodological limitation to raise.
⚠️ The preference-performance split in Perrin et al. and Bird and Chase is a strong discussion-question topic. Be prepared to explain why students might prefer the condition that produces worse outcomes (immediate reinforcement of freedom vs. delayed reinforcement of better grades).
⚠️ Ryan and Hemmes (2005) is the clearest demonstration of extrinsic reinforcement driving behaviour: remove the points, and the behaviour collapses. This is a textbook illustration of operant conditioning principles applied to a real educational setting.
⚠️ Know that all three studies argue their procedures are generalisable beyond the original sample. If asked whether the findings apply to other settings, the answer is yes, with the caveat that the original samples were small and specific (graduate or upper-level undergraduate students).
Q: Compare the independent variables across the three studies. What do they have in common?
A: All three IVs involve manipulating the consequences or conditions attached to study-related behaviour. Perrin et al. and Bird and Chase both compare contingent vs. noncontingent access to practice quizzes. Ryan and Hemmes compare a points vs. no-points condition on homework. The common thread is that each study tests whether adding a behavioural contingency changes when and whether students engage with preparatory material.
Q: Why is the absence of interobserver agreement (IOA) a limitation in these studies?
A: IOA verifies that the data are being measured reliably by confirming that independent observers produce consistent records. Without it, there is no independent check on the accuracy of the behavioural measurements. In Perrin et al. and Bird and Chase, data were collected automatically by a server, which reduces (but does not eliminate) the concern. In Ryan and Hemmes, human grading was involved, making the absence of IOA more notable.
Q: A student argues that noncontingent access is better because 90% of participants preferred it. How would you respond using evidence from these studies?
A: Preference and effectiveness are different measures. While 90% of participants in Perrin et al. preferred noncontingent access, the contingent-access condition generally produced more distributed studying and better quiz performance. Bird and Chase found similar results. Preference tells you what feels easier or more comfortable; performance data tell you what produces better academic outcomes. An exam answer should distinguish between these two dependent variables.
Q: How does Ryan and Hemmes (2005) illustrate the concept of extrinsic reinforcement?
A: The points condition served as an extrinsic reinforcer for homework completion. When points were available, submission rates were high. When points were removed, submissions collapsed to near zero. This demonstrates that the behaviour (completing homework) was maintained by the external consequence (grade points), not by intrinsic motivation or interest in the material.
Q: Could the procedures from these studies be combined into a single intervention? What might that look like?
A: An instructor could release practice quizzes on a contingent-access schedule (requiring early engagement to unlock later material) and attach grade points to their completion. This would combine the scheduling contingency from Perrin et al. and Bird and Chase with the reinforcement contingency from Ryan and Hemmes, potentially maximising both the distribution of study effort and the motivation to complete preparatory tasks.
procrastination interventions, academic procrastination studies, JABA procrastination, Perrin Miller Haberlin Ivy Meindl Neef 2011, Bird Chase 2021, Ryan Hemmes 2005, contingent access study materials, noncontingent access baseline, practice quiz intervention, points condition homework, no-points condition, distributed studying research, massed vs distributed practice, single-subject design education, forced-choice exposure, free-choice preference procedure, homework submission rates, quiz performance contingency, extrinsic reinforcement education, self-management project, behaviour analysis graduate course, Psychology II FIU, applied behaviour analysis interventions, study habits intervention, LMS quiz delivery, contingency management education