How to make fair groups quickly: random, strategic or student choice?
The short answer: for most everyday tasks, groups that are visibly random and changed often are fair, fast and well supported by classroom research in mathematics. Use teacher-chosen groups when you need a specific mix or support, and keep student choice for low-stakes moments. Whatever you choose, change groups regularly so students work with many classmates.
The four ways to form groups
| Method | Strengths | Weaknesses |
|---|---|---|
| Random (cards, draws, a tool) | Fast; no one is "picked last"; nobody can blame you | May put two students together who struggle with each other, or leave one group without a needed skill |
| Teacher-selected | You control mix, language support, behaviour | Slow; students can see the "ability" pattern; you carry the responsibility |
| Student choice | Comfort; fast | Friend groups, social exclusion, and the same students together every time |
| By attainment (similar or mixed) | Targets practice or ensures peers who can explain | Labels students; fixed groups can limit them |
What the research says
Random groups, made visible. Peter Liljedahl studied groups in mathematics classrooms over many years. He reports that when students could see that groups were random, for example by drawing playing cards, they became more willing to work with anyone, social barriers dropped, knowledge moved between students, and enthusiasm rose. In his surveys, a large majority of students in teacher-chosen or self-chosen groups reported being unlikely to offer an idea, which is part of why he recommends randomness. He also recommends re-randomising frequently, roughly every 45 to 75 minutes of class time, so no pattern settles.
Grouping by attainment within a class. A meta-analysis of within-class grouping (Lou and colleagues, 1996) found a small positive average effect of grouping compared with no grouping, about 0.17 standard deviations, and a small advantage for similar-attainment groups over mixed ones, with different effects for lower, middle and higher-attaining students. The Education Endowment Foundation's summary is cautious: the effect is small, and grouping can disadvantage lower-attaining students, especially when groups are fixed. Short, flexible groups for a specific purpose are safer than long-term fixed ones.
Student choice. Social choice tends to reproduce existing friendships and status. Cohen's research on complex instruction describes how status differences between students affect who speaks and who is listened to, and recommends designing tasks and group work to counter it.
Evidence note. Liljedahl's findings come largely from his own classroom studies and surveys in secondary mathematics, and the within-class grouping reviews combine studies from different eras. The direction is consistent, but this is a practical argument for trying visibly random groups more often, not a settled law.
A routine that takes under a minute
- Decide the purpose. Quick talk, problem solving, a project, a station rotation?
- Pick the size. See how many students in a group.
- Form the groups where everyone can see. Cards, a draw, or a projected list. Say once that it is random.
- Check for needs. Quietly swap if two students must be apart or a student needs a language partner.
- Start the task immediately. A visible time limit and a clear first step.
- Change groups at the next task. Do not leave the same group for a whole unit unless the project requires it.
Keeping it fair over time
For a fuller look at grouping students with different needs, see flexible grouping in an inclusive classroom.
- Mix everyone with everyone. Keep a simple tally of who has worked together, or use a tool that does.
- Avoid permanent "ability" teams. Students notice. Change them with each new unit or task.
- Be transparent. "These groups are random, so we all get to work with different people."
- Have a keep-apart list. A short, private list of students who need space from each other.
Try it with ClassMixer
ClassMixer forms groups from your class list in one tap, remembers who has worked together so partners are new each time, lets you add keep-apart rules and swap names by hand, and shows the groups on the projector. The class list stays on your device.
References
- Cohen, E. G. (1994). Restructuring the classroom: Conditions for productive small groups. Review of Educational Research, 64(1), 1–35.
- Education Endowment Foundation. Teaching and Learning Toolkit: Within-class attainment grouping. educationendowmentfoundation.org.uk.
- Liljedahl, P. (2014). The affordances of using visibly random groups in a mathematics classroom. In Y. Li, E. A. Silver, & S. Li (Eds.), Transforming mathematics instruction. Springer.
- Liljedahl, P. (2021). Building Thinking Classrooms in Mathematics, Grades K–12. Corwin.
- Liljedahl, P. Interview: Building thinking classrooms in mathematics. Phi Delta Kappan. kappanonline.org.
- Lou, Y., Abrami, P. C., Spence, J. C., Poulsen, C., Chambers, B., & d'Apollonia, S. (1996). Within-class grouping: A meta-analysis. Review of Educational Research, 66(4), 423–458.