“This year, hardly any of my friends are in my class.” It is the kind of sentence parents may hear on the first day of school, before anything about teachers, timetables or subjects.
When schools reshuffle classes, the change can create discomfort or resistance, especially when it disrupts established friendships and familiar social dynamics. Families quickly begin to ask: Why were the classes changed? Who decides? Is it a good idea to separate a group that was working well? The answers often come just as quickly: “They’ll make new friends,” “It’s good for them to mix with different people,” or, conversely, “They shouldn’t have been separated.”
There is no single way to build a diverse classroom. Schools may keep groups together, reshuffle them at certain stages, or use different criteria to balance social and educational needs. Changing the composition of a class can disrupt established roles, open up new relationships and create new social dynamics. But it can also weaken a network that was already functioning well.
So the more interesting question is not simply whether classes should be reshuffled, but what happens when we redraw the social map of a classroom.
When “we” changes
The evidence is not evenly distributed. There are hundreds of studies on cooperative learning. There are far fewer on redistributing students between classes. Confusing the two can easily lead to conclusions that are more certain than the evidence allows.
A field experiment conducted at a school in Barcelona involved 131 fourth- and fifth-grade primary school students, comparing them before and after the point at which the school reshuffled its classes. Using resource-allocation games, the researchers found that students tended to be more generous, less retaliatory and more willing to reward “prosocial” behaviour towards those with whom they had shared a classroom. After the reshuffle, new classmates begin to be treated as members of the group, while those who have moved to another class do not necessarily lose that status. The boundary of “us” expands without breaking.
Another study, published this year, followed 380 students from first to eighth grade at three points during the same school year. Students in reshuffled classes maintained slightly fewer stable friendships, although the difference was small. But the study also found that students who had joined the school more recently experienced more rejection and maintained fewer stable friendships when they entered a class whose composition had remained unchanged.
This does not mean that reshuffling classes automatically protects newcomers or that it is always preferable. It does suggest that changing the composition of a group can create a certain degree of instability while also redistributing social opportunities, particularly for those entering an already well-established network of relationships. Starting again comes at a cost. But it also reshuffles the cards.
From “putting people in a group” to genuine cooperation
One of the common misunderstandings, both at school and in many adult organisations, is to assume that putting four people around a table means they are working as a team.
A scientific review published in 2026 brought together fifteen meta-analyses comprising a total of 403 independent studies. Overall, cooperative learning showed a moderate positive effect. However, the results varied depending on study design, and the only one of the fifteen meta-analyses based exclusively on experimental studies did not find this effect. The authors themselves caution that the available evidence is uneven.
Another evidence review adds another relevant piece to the picture: gains associated with cooperative strategies in heterogeneous groups could amount to as much as five months of additional academic progress. The most favourable results appeared when the work had a clear structure, with group goals, individual accountability and interdependence among members.
The way a group is organised therefore matters just as much as its composition.
Mixing students creates possibilities; turning those possibilities into cooperation depends on how shared work is structured, particularly when different rhythms, interests, ideas or ways of communicating emerge.
In a classroom, cooperation depends both on the relationships that are built and on how the work itself is organised. Research on intercultural cooperation distinguishes between a relational foundation, consisting of trust, shared vision and resonance, and more instrumental dimensions related to communication, organisation and urgency. The former generally take longer to develop and can sustain deeper, more durable cooperation. The latter facilitate coordination and more immediate action, but depend on a sufficient relational foundation if cooperation is to be sustained over time. This distinction helps explain why a group may complete an assignment successfully while still not having developed solid cooperation.
What this means in the classroom and at home
For teachers, the composition of the group is only the starting point. Turning diversity into cooperation requires time and careful observation in classrooms that are already complex. When conditions allow, structuring responsibilities, observing who is being left out of conversations, or reviewing how decisions have been made can be just as important as the final outcome of the work.
The study by Ariana Garrote and her colleagues also suggests that particular attention should be paid to students who enter already highly cohesive social groups, where finding their place may be more difficult.
For families, there is less room for action, but there is one thing they can do. When a child says, “I haven’t ended up with my friends,” we can avoid two automatic responses: assuming that separating them is necessarily harmful, or assuring them that the change will be good because they will meet new people. We cannot know either of these things for certain.
It may be more useful to ask: “What exactly are you worried about, and what do you need to find your place in this group?” And if the discomfort, isolation or rejection persists, the conversation should no longer remain only at home: it is also worth discussing it with the student’s teacher or tutor.
The young people who today negotiate who will design the cover of a group assignment will tomorrow work in changing, multicultural and often remote teams. They will have to build agreements with people they did not choose and in environments they will not fully control either. Schools cannot, and probably should not, try to create groups without disagreement.
What they can do is help students learn a much more difficult skill: how to build something with people who do not think exactly as we do.
