Group projects, and the setup that keeps them working

Three features of a student group project generate almost all of its problems. The mark is shared, so the link between effort and reward is broken for everyone in it. Nobody holds authority, so no instruction can be given, only requested. And membership is fixed, so the usual response to an unreliable colleague, which is to stop working with them, is unavailable. Each of these is normal in the design of the assessment, and none of them is anyone's fault, which is why treating a struggling group as a moral situation tends to make it worse.

The consequence is that effort quietly converges on the level of the least engaged member, unless something prevents it. Someone notices they are doing more, reduces their input to feel less exploited, and the standard drops. Alternatively one person absorbs everything, delivers a decent mark and resents it for the rest of the term. Both outcomes are common, and both are produced by the same missing mechanism: nothing in the group makes individual contribution visible while there is still time to change it.

So the useful question is not who is reliable. It is what a group can install in place of authority. Visible allocation, a written record, internal deadlines and a normal escalation route do the work that a manager would do elsewhere. None of it requires anyone to be confrontational, and all of it is easier to set up before there is a problem to solve.

More is decided in the opening conversation than in any later one, and it is routinely spent on introductions and a vague division of topics. The mistake is to leave it with a shared understanding rather than a shared document. Understanding degrades and no one notices it degrading. A page that everyone can open does not, and writing it takes a fraction of the time the meeting was going to use anyway. It also sets a tone nothing else can set later, which is that this is a group that keeps track.

None of this needs to be formal, and it should not read as a contract. A short shared note with names, pieces and dates does the job. What it changes is the nature of every later conversation. Instead of asking someone to do more, you are pointing at something the group wrote together, which is a much smaller thing to say and a much smaller thing to hear. The difference between those two conversations is the difference between a request that lands and one that starts an argument.

Almost every serious group dispute is a disagreement about what was agreed. Two people leave the same conversation with different versions, both are honest, and neither can prove anything. A record removes that category of argument entirely. It does not need to be detailed: who is doing what, by when, decided when, is enough to settle nearly all of it, and it takes a minute at the end of a meeting.

Keep it in one place everyone can see, and add to it as the meeting closes rather than from memory afterwards. A short summary posted to the group, listing what was decided and who took what, also creates a moment where anyone who disagrees can say so immediately. Silence after a written summary is worth considerably more than agreement in a room, because the person who was quiet has now had a chance to object in writing.

There is a second function that matters later. If the situation reaches the point where a tutor becomes involved, a group with dated notes is describing a sequence of events, while a group without them is offering competing impressions of each other. That difference decides how the conversation goes, and it is created long before anyone imagines needing it.

Uneven effort is the standard failure, and it is usually raised too late, in one exasperated message, after resentment has been accumulating privately. By then the conversation is about character, and character is not negotiable. Raised early, the same subject is administrative: a piece is outstanding, and the group needs to know when it is coming. The difference is timing more than wording, though wording helps a great deal.

What keeps it survivable is talking about the work rather than the person, and leaving a route open that does not require anyone to admit anything. People who are behind are often behind for reasons they have not said, and a message that assumes bad faith forces a defence rather than an answer. Ask about the deliverable, offer a smaller version of it, and put a new date in writing.

If the disagreement has already become personal, the general repair techniques are in the guide on resolving conflicts with other students, and they apply here without modification. What is specific to group work is the deadline. Ordinary conflicts can be left to cool; this one has a submission attached, so the group needs a working arrangement rather than a resolution, and those are different targets.

Disappearance is a different problem from underperformance and needs a different response. Someone who is present and slow can be prompted. Someone who has gone silent cannot be, and the group's instinct is to keep waiting because chasing feels aggressive. Waiting is the expensive option. Do two things at once: continue as though the missing piece will not arrive, and make a calm, dated attempt at contact through more than one channel.

Assume nothing about the reason. Silence looks identical whether someone has withdrawn from the course, is unwell, has a family situation, has lost access to a device, or has concluded the work will get done without them. That is precisely why a group should not diagnose it and should not phrase messages as if it knows. One short message naming the outstanding piece, giving a date, and saying what the group will do otherwise is both fair to the person and usable later if the tutor asks what was tried. Where you have reason to think someone is struggling rather than avoiding, supporting a friend who is struggling is a more useful frame than a group message.

Tell the tutor sooner in this case than in any other. A member who has gone entirely quiet is information the department may need for reasons that have nothing to do with your project, and passing it on is not a complaint. It also protects the remaining members, because a group that reports a disappearance early is asking for guidance, while a group that reports it at submission is asking for an exception.

Groups now routinely contain people in different countries, working in a second or third language, trained in academic systems that expect different things. Much of the friction this produces gets read as attitude when it is really translation. Someone who writes brief, direct messages may be economising in a language they are still working in. Someone who never contradicts the group in a meeting may come from a system where disagreeing in front of others is discourteous, and may say the same thing willingly in writing. Adapting to a new academic culture takes time that a project deadline does not allow for.

The mechanisms that fix this are dull and effective. Put decisions in writing so that nobody depends on catching a spoken point at speed. Use asynchronous work as the default and meetings as the exception, since a slot that suits everyone across a spread of time zones is often one nobody is at their sharpest for. Confirm understanding by restating rather than by asking whether everyone understood, because the answer to that question is always yes. And where a member is working in an additional language, the practical guidance on handling language barriers applies to the group as much as to the individual.

Students avoid involving a tutor because it feels like informing on someone. Look again at what is actually happening: a group is telling the person who set the task that its allocation has stopped working, and asking how to proceed. Tutors deal with this constantly, and they generally prefer to hear it while the work can still be adjusted rather than at the point where a mark is being contested. Nothing about it requires naming a villain.

Escalate as a group where possible, bring the record rather than the feelings, and ask a question rather than make a demand. What the tutor needs is the allocation, the dates, what was attempted and what is outstanding. What they can offer varies by institution: a change of weighting, an individual component, a conversation with the member concerned, or a confirmation of what the group is entitled to do on its own. Asking early keeps all of those available, and asking at submission usually leaves only the last.

Group work is also the closest thing in a degree to how most jobs actually run, which is why the habits are worth building rather than merely surviving. Written allocation, visible deadlines, early escalation and the ability to raise a problem without attacking a person are the same skills that make someone straightforward to work with later, and they carry into early career more directly than much of the content of the course does.