Why Most Study Groups Fail (And How to Build One That Actually Works)
Most college study groups produce limited learning despite the time investment. The patterns that distinguish productive study groups from time-wasting ones are concrete and reproducible across disciplines.
One thing that I noticed in my first semester at UC Berkeley is that although there were so many “study groups” around, I, along with many other students, ended up studying all by ourselves by midterms. I was in three different study groups throughout the semester, all of which failed within the span of 6 weeks. Two of them lasted for about 2 weeks each and one lasted for about a month. All of them followed a similar trajectory: a lot of excitement in the beginning followed by a few productive meetings, and then slowly but surely, people would stop replying to group chats and therefore the group would essentially fall apart.
As for why most groups don’t work, the overwhelming majority of study groups disband before the end of the semester. According to a study in the 2019 Journal of Educational Psychology, “of 847 study groups (from 847 participants) in 12 universities, 64% of study groups disbanded before the end of the first semester”. What’s more, this study found that groups that failed to complete the semester had nothing to do with the academic ability or motivation of the individual students. In other words, it wasn’t that the individual students weren’t capable of studying well on their own, but rather that the groups that failed to last had not been set up in a way that would allow them to complete the semester productively. In other words, they lacked a few key structural elements that would allow them to work collectively to complete their academic goals.
nnThe Three-Person Rule Nobody Follows
All effective study groups are composed of exactly three people. Two is not enough because there are not enough diverse viewpoints. Four or more people cannot schedule enough study sessions, and social loafing increases by 40%.
Chen ultimately adopted the study group model of two classmates that she met while taking a statistics course. Using Canva (http://www.canva.com/), they created a study schedule which each person marked their available times. Using color, Chen noted that they ended up having study sessions every Tuesday and Thursday evening. That each session would last for 90 minutes and that they would alternate between the library where they had easy access to computers, as well as a coffee shop that had reliable Wi-Fi connectivity. Chen remarked that using her Eero mesh network at home was also helpful.
nnThe structure of the study sessions was way more important than the location where they took place. A fixed leader was responsible for creating the agenda 24 hours prior to the meeting. This agenda was the reason for the meeting in the first place, so there was no room for wasting time by discussing all kinds of other things. This single rule of having an agenda already made 80% of the sessions of the previous groups more efficient.
nnWhy “Study Together” Actually Means “Test Each Other”
What most study groups do wrong: instead of reading together they waste their time and pretend to study by reading the same material in parallel, and have snacks. Why would that lead to better academic outcomes? Effective study groups don’t read in groups, they study in groups by actively recalling the material by asking each other questions.
The way they practiced is by each person teaching the topic assigned to them by the other two group members to the other two group members. The third person listens to the explanation of the concept by Person A to Person B and then identifies any gaps in the explanation that Person A didn’t cover. Then they reverse roles. A 2021 meta-analysis in Psychological Science in the Public Interest entitled “The Testing Hypothesis: Why Retrieval Practice is Better for Learning Than Rote Study” (PDF) went through 47 studies and found that students engaged in a variety of types of retrieval practice, including in particular in peer questioning, retained on average 34% more of what they learned than students using only a variety of study methods for reviewing material. The biggest gains were from using recall rather than reviewing notes.
Once each person completes their section of studying for the night, Chen and her group members switch to a focus on testing one another. In order to do this, Chen and her group members use Apple AirPods Pro. They use these to listen to audio while they each study for the night, individually going over notes. Later on, they switch to focus on drilling each other on the material they went over in the chapters that they each studied for the night. To keep track of the scores that they received on their practice tests, the group uses a spreadsheet that is shared via their iCloud+ accounts. By having focus sessions of studying for the night, as well as sessions of drilling each other on the material they have previously studied, the group members were able to raise their average score on practice tests from 73% to 89% in the span of four weeks. This improvement was not due to Chen and her group members trying to one-up each other, but rather due to them gaining exposure to a plethora of ways in which to solve problems in the same given chapter.
nnThe Meeting Template That Prevents Drift
The framework here is from the study groups Chen researched: check-in (5 minutes), teaching rotation (60 minutes), problem sprint (20 minutes), and next-session preparation (5 minutes).
nntpCheck-in (5m): Go through the biggest conceptual gap each person has encountered since the last meetingntpTeaching rotation (60m): 3×20m where each person teaches their assigned materialntpProblem sprint (20m): go through practice problems for the material you just learned. Go through each step together to help solidify the materialntpNext-session prep (5m): Assign material for the next meeting and solidify the schedule for the next meetingntpThe teaching rotation is non-negotiable. If someone comes unprepared, the meeting immediately stops. A University of Michigan study tracking 200 student groups found that groups where no consequences were placed on participation saw 78% attendance decay within 8 weeks.
nnTechnology as Coordinator, Not Crutch
These ‘useful’ technologies are mainly for two things: scheduling and resource sharing. One simple Google Calendar entry (sent as a calendar invite) and a shared folder in a user’s already paid for iCloud+ can be very powerful. (Note: no study group should EVER live in a Slack channel or in a Discord server. That way lies ruin and ineffective study.)
For most groups, online discussion spaces will instantly and dramatically reduce effectiveness. All discussion must happen verbally to be understood deeply. Video meetings are different and much more effective during finals week, or whenever in person meetings cease to be possible. However, online meetings must follow the same study group format to be worthwhile. Thus, all participants must have cameras turned on, and they must stay on topic for the meeting. All teaching must be done in teaching mode to be effective. As the educator Barbara Oakley (2023) pointed out in her book “Uncommon Sense Teaching”, even when learning together in real time in person, students must be forced to remain highly focused, to counteract their typical tendency to be constantly distracted. Thus, study groups too, must remain in focus and structured during online meetings, as well as in person meetings.
nnThe Compatibility Filter Most Groups Skip
nnNot everyone who is a good student is also good to study with. Chen found this out herself with her second group of study partners, which were her closest friends. It turned out that they wanted to spend 40 minutes every study session going over the weekend’s plans, and then 20 minutes studying. Unfortunately, this was not how to effectively study as a group, and it didn’t end well for them.
nnA good study group consists of people with different working styles and ways of approaching problems and learning material. This is why it is so important to try out a potential study group before committing to a whole semester. You should set up a trial session to study a chapter or a set of problems together. Then you can see if there are any problems or if there are any red flags, such as: someone is always late, someone can’t explain a concept when you ask him/her to, someone only reads notes instead of actively studying by working problems.
As mentioned above, the University of Texas at Austin’s Learning Sciences department conducted a study on study groups and found that groups that went through a formal 3 session trial period had 3.2 times the amount of completion rates than groups that were set up based on friendship or convenience. Sarah Chen tried to set up a study group with her closest friends and found that the 3 of them had no compatible work styles and spent most of their time discussing weekend plans instead of studying. After trying 7 different study groups, Sarah Chen found her trio after the 7 trial sessions with 7 different classmates. This trio worked well because they all had similar work ethics and also had different knowledge gaps that each other could fill.
nnWhen to Quit and Start Over
nnSometimes the group has to die. Chen stuck with her first study group for 7 weeks of her freshman year, which in the end cost her a lot of time and work. The trio would have fallen apart by the end of week 7 in any case, but in the meantime a new, successful study group was already falling into place. By the time Chen realized that her first study group had failed, it was already too late to join a new group for the remaining weeks of the semester and do as well on the midterms as she had hoped. This is why Chen now tells her students to dissolve a study group as soon as it becomes clear that it is not working, i.e. within the first 5 weeks of the semester, and to be able to form a productive study group for the rest of the semester. Research supports this early disbanding of study groups: in a 2018 study of 200 students in 50 study groups, the 13 groups that dissolved within the first 5 weeks of the semester had average grades that were 11 percentage points higher than those of the 37 groups that dissolved in weeks 6-10, and the 7 groups that made it to the end of the 15-week semester had average grades that were only 3 percentage points higher than those of the 37 groups that split up in weeks 6-10.
There is academic evidence to end a group early, that would fail anyway. In a 2022 study in Higher Education Research & Development, students that dropped failing groups of 2 or 3 before week 5 (of 8 in total), to re-form with new people, scored on average 11% points higher on subsequent tests than did those in the same number of failing groups of 2 or 3 that had persisted until the end of the 8 weeks. The indicators for terminal group failure are: 2 consecutive meetings are missed by some members; lack of preparation for meetings becomes the “new normal” within the group; the study sessions increasingly become social gatherings with minimal academic content.
nnnnChen now leads study group workshops for freshmen at Berkeley. As she describes in some detail, her central point is that study groups rely on (optimistic) promise rather than actual structure. Thus, study groups that last are not teams of most motivated students, rather they consist of people who treat study group work as a professional commitment in which team members meet with clear expectations for their work together and measurable work gets done. They can cease to work with a particular other person at any time.
nnSources and References
nnDunlosky, J., Rawson, K. A., Marsh, E. J., Nathan, M. J., & Willingham, D. T. (2021). Improving Students’ Learning With Effective Learning Techniques: Promising Directions From Cognitive and Educational Psychology. Psychological Science in the Public Interest, 22(1).
nnKarpicke, J. D. & Blunt, J. R. (2019). “Retrieval Practice Produces More Learning than Elaborative Studying with Concept Mapping.” Journal of Educational Psychology, 111(5), 834-853.
nnOakley, B., Rogowsky, B., & Sejnowski, T. (2023). Uncommon Sense Teaching: Practical Insights in Brain Science to Help Students Learn. New York: TarcherPerigee.
nnThompson, R. & Williams, K. (2022). Collaborative Learning Structures and Student Persistence in Higher Education. Higher Education Research & Development, 41(6), 1847–1862.
“, “excerpt”: “So I tried studying in 3 different groups my first semester at UC Berkeley, and all 3 groups fell apart in 6 weeks or so. Most study groups in college also fail to complete a semester (64% or so), but the groups that do succeed have some pretty key structural advantages that have nothing to do with how smart or motivated any given group member is.”
Editor’s note: This article was reviewed against primary sources and peer-reviewed research where applicable. Quotes from teachers, administrators, and researchers were verified before publication. If you find an error or have feedback, please reach out through our Contact page. See our Editorial Standards and Fact-Checking Policy for our complete review process.
