Every group project grade has the same complaint waiting behind it: "I did all the work and we got the same grade." Sometimes it's true. Sometimes it's the loudest kid's version of events. Either way, a single shared grade for the whole group is the fastest way to make your best students resent group work and your least motivated students learn that coasting is free.
Why the shared grade breaks down
A rubric built entirely around the final product, the slides, the poster, the model, only measures what got produced, not who produced it. That's fine for judging the deliverable. It's not fine for assigning a grade to four individual students who did wildly different amounts of work to make it happen.
Two grades, not one
- A product grade, shared by the group, scored against the same criteria you'd use for any project: accuracy, completeness, presentation. This part genuinely is a team outcome and should be graded as one.
- A contribution grade, individual, based on evidence you actually collected during the process, not just a vibe at the end. This is the part that fixes the free-rider problem, but only if you build in a way to see contribution as it happens.
Building in evidence of contribution
- A running log, not a memory test. Ask each group to keep a shared doc where they log who did what, updated as they go, not reconstructed the night before it's due. A log built in real time is far harder to fake than a summary written after the fact.
- A short individual reflection at the end. "What did you specifically contribute? What did a teammate do that you couldn't have done yourself?" A student who did nothing struggles to answer this convincingly; a student who did the work usually over-explains it, which is its own tell.
- A quick peer rating, weighted lightly. Confidential, simple (1-5 on effort and reliability), and never the sole input, peer ratings catch patterns a teacher circulating the room might miss, but they shouldn't carry the whole grade since friend groups skew them.
- What you actually saw. Two minutes of checking in with each group mid-project, noting who's talking, who's working, who's on their phone, is worth more than any after-the-fact form. Write it down immediately; don't trust yourself to remember it by Friday.
A quick before/after
Before: one shared grade for the final poster, no way to tell who actually built it.
After: a product grade for the poster itself, plus an individual contribution grade built from a running log, a short reflection, and what you saw during work time, so the kid who carried the group and the kid who coasted don't land on the same number.
Where this gets easier to set up
Our [grading rubric generator](/grading-rubric) can help you build the contribution-specific criteria (not just the product criteria) so it's clear from day one that individual work is being scored separately, and our [PBL unit generator](/pbl-unit) is useful for designing the driving question and milestones in a way that naturally produces checkpoints where you can see who's contributing along the way.
The test of a fair group grade
Ask whether a student who did the real work and a student who free-rode could look at their two grades and see the difference. If the only number they both got is identical, the rubric isn't measuring what actually happened in that group.
