Why group work hides who did the work

Group work is meant to teach collaboration. The grade it usually produces teaches something quieter: that effort and reward can come apart. When five names share one document, the finished file cannot say who wrote it, who revised it, or who never appeared. The mark is real. The story behind it is missing.

The shared mark is the root of the problem

A single mark given to every member of a group is fast to award, and that is most of its appeal. But a review of the group-work assessment literature concludes that a shared mark rarely produces good learning behaviour, frequently leads to freeloading, and is reasonably perceived by students as unfair. ¹ UK sector guidance makes the same point from the other direction: a shared mark can both inflate a weak member and deflate a strong one, regardless of what each actually did. ²

This is not a fringe complaint. When students were asked what worried them most about collaborative coursework, free-riding was the single greatest concern across every discipline surveyed, and the sharpest frustration was receiving the same grade as someone who had not contributed. ³

The disengagement is predictable, not random

It is tempting to read a quiet group member as simply lazy. The research describes something more structured. The classic finding is social loafing: people put in less individual effort on a collective task as the group grows, even once you rule out coordination problems. ⁴ A meta-analysis of 78 studies found the effect robust and general, and, importantly, found that it shrinks under specific conditions. ⁵

Two of those conditions matter here. People free-ride when their own contribution starts to feel dispensable to the result. ⁶ And capable members will deliberately pull back rather than be the one carrying a teammate who is coasting, an effect named the sucker effect. ⁷ One disengaged member, in other words, can quietly lower the effort of the people who were willing.

Students are already judging the fairness

Faculty sometimes assume students accept group marks as part of the deal. A recent validated instrument shows they do not. Students judge the fairness of group assessment along three distinct dimensions, and the primary one is grade congruence: whether the marks people received actually matched what they contributed. ⁸ When the mark and the work come apart, students notice, and they remember.

Why "just add peer marking" is not the whole answer

The usual fix is to ask students to rate each other. Done well, peer assessment is a genuinely useful signal, and we treat it as one. But it is not a clean instrument on its own. Peer scores drift upward when a grade is attached, friends are scored more generously even when a rubric is in place, and the size of the bias is larger among weaker contributors. ^{9 10} Peer marking is part of a fair process. It should not be the only thing the process can see.

What changes when the work is visible

The opposite of a black box is not surveillance. It is a record. If each contribution is attributed as it happens, the conditions that suppress effort weaken, the people who carried the project can be recognised for it, and a disputed grade can be explained instead of defended. That is the premise Dwixel is built on, and the rest of this series covers how to do it without crossing into monitoring, and how to turn it into a grade you can stand behind.