StudyDuo
The short answer to how to be a good study partner: try the work before you meet, and ask questions that show where you're stuck. In the session, explain without handing over answers and give feedback your partner can act on. Most of the nine habits below come from research on how students learn in pairs and small groups.
The nine habits at a glance
| Habit | When | What goes wrong without it |
|---|---|---|
| 1. Try the problems first | Before | The session turns into a lecture you could have read |
| 2. Bring pointed questions | Before | "I don't get it" earns the same explanation twice |
| 3. Wait before you help | During | You answer before your partner has finished thinking |
| 4. Explain, then hand it back | During | They can follow your solution but can't produce one |
| 5. Say why, and where you're unsure | During | You recite what you know and learn little yourself |
| 6. Comment on the work, with a next step | During | Feedback stings and can make things worse |
| 7. Own your no-shows | Between | The hour is lost for both of you |
| 8. Use the level gap | Every session | One person tutors while the other coasts |
| 9. Share understanding, not answers | Graded work | You both risk an academic integrity case |
Before you meet
1. Try the problems before the session
Work through the set alone first, and stop on any one problem after about 15 minutes. Mark the step where you got stuck, then move on. Wrong attempts are useful: they're what your partner's explanation will land on.
Trying first has evidence behind it. In a 2021 meta-analysis of 53 studies, Tanmay Sinha and Manu Kapur compared two orders: solving problems before being taught, and being taught first. Problem solving first came out ahead (Hedges' g = 0.36), and further ahead when it was run as designed (0.37 to 0.58). The pattern reversed for skills not tied to one subject, where being taught first did better, and was less clear for children in grades 2 to 5.
2. Bring questions that point at the gap
"I don't get chapter 6" gives your partner nowhere to start. "I follow the setup, but I lose it where the denominator becomes n minus 1" does. Write two or three questions like that while you prepare.
Arthur Graesser and Natalie Person studied tutoring sessions in research methods with college students and in algebra with seventh graders (American Educational Research Journal, 1994). Students asked about 240 times as many questions in tutoring as in class. Once students had some tutoring experience, higher achievement went with the quality of their questions, not how many they asked.
How to be a good study partner during the session
3. Wait before you help
When your partner pauses over a problem, count to five before you say anything. When they finish an answer, pause again before you respond.
Teachers rarely wait that long. Mary Budd Rowe found that teachers typically give students less than one second to start an answer (Journal of Teacher Education, 1986). They also reply within a second of a student stopping.
Kenneth Tobin reviewed the research on wait time (Review of Educational Research, 1987). When the average wait went past 3 seconds, classroom talk changed and higher-level achievement rose in school science. That research was about teachers, but a partner who jumps in does the same thing.
4. Explain the step, then hand the problem back
When your partner is stuck, explain the idea behind the step they missed, then give the problem back and let them finish it. Next, hand them a similar one to solve without help. If they can't, explain again from a different angle.
Noreen Webb, Jonathan Troper and Randy Fall followed six classes of seventh graders through units on decimals and fractions (Journal of Educational Psychology, 1995). Among students who needed help, the best predictor of achievement was what the researchers called constructive activity. That meant solving problems, or explaining how to solve them, with ideas from the help they'd received. The amount of help predicted how much of that work students did, but not their scores directly. An explanation pays off when the listener puts it to use.
5. Say why, and say where you're unsure
When you're the one explaining, give the reason for each step and connect it to something you both covered earlier. Say it out loud when you aren't sure: "I always do this step, but I can't say why it works. Let's check."
Explainers tend to skip that part. Rod Roscoe and Michelene Chi reviewed the research on what peer tutors actually do and found "a pervasive knowledge-telling bias" (Review of Educational Research, 2007). Even trained tutors delivered what they already knew instead of building on it: checking their own understanding, linking new ideas to old ones, working out what follows. As a result, the authors concluded, tutors' own learning rarely reaches its potential.
6. Comment on the work, and say what to change
Feedback helps most when it's about the answer and says what to do next. "Your setup is right, but the sign flips in line 3" does both; "you're bad at algebra" is about the person.
That second kind tends to backfire. Avraham Kluger and Angelo DeNisi pooled 607 effect sizes in a 1996 meta-analysis (Psychological Bulletin). Feedback improved performance on average (d = 0.41), yet over a third of feedback interventions made performance worse. Feedback did less the more it pulled attention away from the task and toward the self.
How much the feedback tells you matters as well. Benedikt Wisniewski, Klaus Zierer and John Hattie pooled 435 studies with more than 61,000 participants and compared types of feedback (Frontiers in Psychology, 2020):
| Feedback type | Effect size |
|---|---|
| A reward or a penalty | |
| Right or wrong, plus the correct answer | |
| The answer, plus how to get there and check yourself |
Between partners, that's the gap between "wrong, it's 12" and "wrong, it's 12: you added before you multiplied, so do the brackets first and check by putting 12 back in."
Reliability, levels and graded work
The rest of how to be a good study partner is about the partnership itself: showing up, a gap in level, and graded work.
7. Own your no-shows, and handle theirs calmly
If you can't make it, cancel as soon as you know and offer a new time in the same message. Send whatever you prepared, so your partner can still use it. If you miss without warning, apologize once, briefly, and turn up early next time.
If your partner doesn't show, wait ten minutes, then do the planned work alone, so the slot still produces something. Send one neutral line: "Missed you today. I did set 4, notes attached. Still on for Thursday?" After a second unexplained miss, ask whether the time still works for them or a different format would. The rules worth agreeing on at the start are in our guide to finding a study partner.
8. Make a level gap work for both
Partners rarely know the material equally well, and the gap itself isn't the problem. Noreen Webb, Kariane Nemer and Stephen Zuniga compared high-achieving students doing group science assessments in same-level groups and in mixed groups (American Educational Research Journal, 2002). In same-level groups, the strong students all did well. In mixed groups, some did just as well and some didn't, and the quality of the group's work together explained much of the difference.
So work on how you work together:
- The partner who's ahead asks "What did you try?" before explaining anything.
- The partner who's behind explains each solved problem back in their own words.
- The partner who's ahead takes the hardest problem in each set and thinks it through aloud, so they're learning too.
9. Share understanding, not finished answers
On graded work, the line between studying together and copying is set by each course, not by the two of you. MIT's academic integrity handbook tells students not to assume they know the collaboration policy, and to ask the instructor (Collaboration). In one of its examples, a math course lets others explain their solutions to you, but write-ups must be done independently, without looking at anyone else's. Another course requires you to name your collaborators.
A safe default: talk through the approach together, close the shared notes, and write your own version alone.
How to be a good study buddy on video
Online, the same habits need a few adjustments:
- Keep your camera on during the talking parts, so your partner can see when you're lost.
- Share your screen or hold the page up, so your partner sees the exact line.
- Put formulas and links in the chat, where they stay after the call.
- Mute during silent work, and say so when you step away.
On StudyDuo, a 1:1 study session is a video call for two. Post a time on the board, now or within the week, and whoever joins comes into the video with you; talk with the mic or use the chat. Both of you need a verified photo, and sessions are for people 16 and older. For the solo prep in habit 1, Focus Room never closes.
These habits are about how to be a good study partner, not what to study together. For that, see 8 methods for a study session in a pair, from teaching back to arguing before you check the key. If your sessions are turning into a small group, Study Alone or in a Group? explains why effort can drop when work is pooled.
FAQ
What if my partner keeps asking me for the answers?
Give the next step instead of the answer, then ask what they would do after it. If they want answers to hand in, point to the course's collaboration policy: on graded work, both of you can be held to it.
How do I give feedback on a partner's essay?
Mark it against the grading rubric, not your own taste. Name one thing that works and the two changes that would raise the grade most, each with the line it refers to.
How do I say our sessions aren't working?
Describe the sessions, not the person: "Our last few sessions ran long and we finished fewer problems. Can we try a fixed agenda next week?" Offer one concrete change and try it for two sessions before deciding anything.
Can I be a good partner if I'm shy?
Yes. Most of the work is preparation and listening, not talking. Write your questions down before the session and read them out, or type them in the chat on a video call.
