Four failure modes

Group work fails for predictable reasons.
All are design problems, not student problems.

Group work fails for four predictable reasons: tasks that don't require collaboration, groups too large for individual accountability, insufficient preparation time before group activity begins, and no mechanism for converting group discussion into individual understanding.

The instinct when group work fails is to blame the students. This diagnosis is almost always wrong. Students behave rationally in group work: if the task can be completed without genuine collaboration, they will let one student do it alone. If there is no accountability for individual contribution, they will free-ride. These are not character failures — they are rational responses to a poorly designed task.

📊The social loafing research
Bibb Latané's social loafing studies showed that individual effort in group tasks decreases as group size increases — a group of 4 produces roughly 50% of the effort of 4 individuals working alone on the same task. Social loafing is not a personality trait — it is a structural response to diffuse accountability. The fix is structural: tasks designed so that individual contribution is both necessary and visible.
Latané, B., Williams, K. & Harkins, S. — Many hands make light the work, Journal of Personality and Social Psychology, 1979
Fix 1: Tasks that require collaboration

The group can only succeed if individuals
contribute something the others don't have.

The most important design decision for group work is whether the task genuinely requires multiple contributors. A task designed so that each member holds a necessary piece — of information, perspective, or skill — cannot be completed without genuine collaboration.

1
Jigsaw — each member becomes expert in different content
Free-riding is structurally impossible

Divide the total content into N parts (where N = group size). Each student becomes the class expert on their part before the group convenes. The group task requires every member's content to be combined — no member has enough information alone.

Design requirement
Each content section must be genuinely necessary to complete the group task. If students can produce a reasonable answer without one section, that section's member can free-ride. Test the task: could the group complete it if one member were absent? If yes, redesign.
2
Gallery walk debate — each group produces an argument; all groups challenge each other's
Both producer and critic roles require genuine engagement

Groups produce written arguments on posters. The group task then requires them to identify weaknesses in other groups' arguments. Every group is both producer and critic — and both roles require genuine engagement with the content.

Why this works
Knowing your argument will be challenged by other groups changes how you construct it. Groups that know they must defend their work produce more carefully reasoned arguments than groups that know only the teacher will read them.
3
Design challenge with constraints — each group has a different constraint
Each group's constraint knowledge is unique

Give groups the same problem but different constraints: one group has a budget limit, one has a time limit, one has a material limit. The solutions will differ, and comparing them generates productive discussion. Each group's constraint knowledge is unique.

The discussion it generates
"Your solution works with unlimited budget — but ours had to cost under €50. Here's what that forced us to prioritise..." The constraint differences create genuine comparison that no single group could generate alone.
Fix 2: Group size and preparation

Three is usually optimal.
And never start a group before individuals have thought.

Most teachers default to groups of 4–5. Research consistently suggests that groups of 3 produce higher individual contribution and better outcomes for most school tasks. In a group of 3, there is no majority coalition that can outvote a minority — every member must be persuaded rather than outvoted, which increases genuine discussion relative to vote-taking.

Before any group activity begins, every student must have had individual thinking time. 2–3 minutes of individual processing before the group convenes. Without this step, the first confident speaker dominates — because they are the only one who has formed a view. With it, every student arrives at the group with a position.

💡The exit mechanism is equally important
Group discussion must convert to individual written understanding before the lesson ends. Not a group report — individual writing. A student who contributed actively to a discussion but produces nothing individual has not demonstrated personal understanding. The exit question is individual, not collaborative.