A peer evaluation for a group project is a confidential rating of each teammate's contribution, usually submitted near the end of the project and sometimes midway. Instructors use it to see who did what, and in many courses it adjusts individual grades. To do one well, rate specific behaviors rather than personalities, back every score with a concrete example, and start documenting your own work from the first week.
This guide covers both sides: how to fill one out honestly, and how the numbers usually become grades, so you know what is at stake before you click submit.
TL;DR
- Rate what teammates did (meetings attended, tasks delivered on time, quality of drafts), not whether you like them.
- Keep a simple contribution log all term. It makes your ratings specific and protects your own grade.
- Many instructors convert ratings into an adjustment factor and multiply it by the team grade, so a single low score can move your mark.
- Early, low-stakes check-ins give struggling teammates time to change before the final grade is set.
- Know your syllabus rules: weight, deadline and anonymity differ by course, and the syllabus is where they are written down.
Table of Contents
- What a peer evaluation actually measures
- How to do a peer evaluation, step by step
- How ratings turn into grades
- Does peer evaluation work?
- Peer evaluation when your group has two members, or a free rider
- For instructors and TAs: designing a fairer evaluation
- Our perspective
- Keeping the project, and your grade, on track
- After the evaluation: what to do with the result
- FAQ
- Sources
What a peer evaluation actually measures
Student peer evaluation is a form of assessment in which students judge each other's work against set criteria, typically a predetermined rubric. In a group project it answers a question the final product cannot: who contributed what.
That matters because teamwork assignments often produce one product and one mark. Monash University researchers note that this single mark goes to everyone regardless of contribution, which is the exact complaint students raise. Peer ratings are the most common fix.
Two surprising points from the research are worth knowing:
- Teams with a free rider do not automatically turn in worse work. A study in the Journal of the Scholarship of Teaching and Learning suggests the main problem is fairness in grading, not the quality of the project. This is why your evaluation matters even when the project "turned out fine."
- Students are better at judging substance than polish. Research on peer review of writing found higher agreement among raters on higher-level qualities such as analysis than on mechanics such as grammar and formatting. Trust your read on who drove the ideas.
How to do a peer evaluation, step by step
1. Read the criteria before you rate anyone
Check whether your instructor uses a rating scale (for example 1 to 5 per criterion) or a single overall score. A review of peer evaluation tools describes these as the two most common approaches. Rating scales are more granular, so you will need a reason for each number.
Also confirm the basics in the syllabus: due date, whether you rate yourself, whether comments are anonymous, and what percentage of the project grade depends on the result.
2. Pull out your evidence
Open the shared document's version history, your group chat, meeting notes and task list. You are looking for facts:
- Who attended which meetings, and who sent updates when they could not
- Who delivered their section by the agreed date
- Who revised, formatted or presented
- Who helped a teammate who was stuck
If you kept a contribution log, this takes ten minutes. If you did not, reconstruct it now, then start one for the next project.
3. Score behaviors, not people
A rating of "2 out of 5, because they missed three of five meetings and submitted their section two days after the internal deadline" is useful. A rating of "2 out of 5, because they were annoying" is not, and instructors tend to discount it.
Watch for three common distortions:
- Friendship inflation: giving everyone top marks to avoid conflict.
- Retaliation: scoring down someone you argued with.
- Halo effect: rating the loudest person highest because they talked the most, not because they delivered the most.
4. Write comments that a professor can act on
Good comments name the situation, the behavior and the effect. For example: "Maya wrote the literature review and the first draft of the methods section, and responded to edits within a day." For a low rating: "Jordan did not submit the data table by the agreed deadline, so we rewrote it the night before."
Students often feel unqualified to grade peers. Educators suggest treating reviews as developmental, not just punitive, so include what a teammate did well, and what you wish had gone differently.
5. Rate yourself honestly
If self-evaluation is part of the form, be specific and modest. Overclaiming stands out when every teammate describes the work differently. Underclaiming can quietly cost you points.
6. Submit privately and move on
Most courses treat these as confidential. Do not compare scores in the group chat. If you disagree about contributions, raise it with your instructor or TA rather than in a public thread.
How ratings turn into grades
The math varies by course, but one pattern is common. Some grading models use behavioral anchors to convert qualitative feedback into numbers, such as Excellent equals 100, Satisfactory 75 and Deficient 37.5. Those numbers become an adjustment factor, which is multiplied by the team's base grade.
A simplified example. Your team earns an 88 on the project. Each member's average peer rating is compared with the team average:
| Member | Average rating | Team average | Factor | Individual grade |
|---|---|---|---|---|
| Ana | 4.5 | 4.0 | 1.125 (capped at 1.05 here) | 92.4 |
| Ben | 4.0 | 4.0 | 1.00 | 88.0 |
| Chloe | 3.5 | 4.0 | 0.875 | 77.0 |
This table is an illustration of the method, not a rule your instructor uses. The same IU paper describes a factor of the individual average divided by the team average, capped at a value such as 1.05 so extreme ratings do not distort grades while low contributors are still penalized. Your course may cap differently, or treat peer scores as a separate component worth a set number of points.
The takeaway: a one-point difference in your average rating can matter. Check the syllabus weighting before the evaluation opens, not after.
Does peer evaluation work?
The evidence is encouraging, with caveats.
- It reduces complaints about free riding. In an engineering design course, complaints fell from 26% of students to 7% after structured peer assessment was introduced.
- It can be reliable. Carefully designed peer ratings can approach the validity of a single instructor's marks, with agreement scores above 0.40 on most rubric criteria.
- It improves with practice. Programs that run feedback frequently over at least six months tend to produce better feedback quality than a one-off form. Most course projects are shorter than that, which is one reason first-time evaluations feel awkward.
And the caveat: a form is only as good as the evidence behind it. Two raters with no notes can describe the same teammate very differently.
Peer evaluation when your group has two members, or a free rider
In a two-person team, each person's rating of the other largely decides the outcome, so a disagreement is a stalemate. Stick to dated facts and offer to share your log with the instructor if they ask. If the course allows it, ask in advance how ties or conflicting ratings are handled.
With a free rider, avoid waiting until the final form. Instructors are advised to run a formative peer review early in the project, which doubles as an early warning for struggling members. If your course does not have one, you can create the same effect yourself:
- In week one, write down who owns which tasks and the internal deadlines.
- At the midpoint, hold a ten-minute check-in against that list.
- If someone falls behind, message them privately and specifically: what is missing, by when, and what help would unblock them.
- If nothing changes, email your instructor with your log before the deadline, not after grades post.
Instructors use this documentation too. A Cornell course, for example, explains that private peer evaluations exist because individual grades will be assigned and each member's contribution has to be assessed fairly.
For instructors and TAs: designing a fairer evaluation
If you are a student leader or TA, a few design choices matter most:
- Use behavioral anchors. Define what a 5 and a 2 look like. Rubrics with described levels make ratings more consistent, and this OSCE feedback example from medical education shows how behavioral criteria and scales translate observations into scores.
- Train students first. Show a sample of a good and a poor comment, and explain how the ratings will be used.
- Evaluate more than once. A midpoint check lets teams repair problems while it still counts.
- Cap the adjustment. A cap limits how far one rating can swing a grade.
- Pair ratings with instructor judgment. Peer scores are evidence, not a verdict. Cornell's teaching center recommends rubrics shared in advance and regular feedback so groups can gauge their progress.
Our perspective
We build a planner for college students, and what we see again and again is that group-project trouble is a tracking problem before it is a people problem. Nobody wrote down who owned what, nobody noticed a slipping deadline, and by the time the evaluation arrives the memory of week three is gone.
So we suggest treating the peer evaluation as the last step of a process that starts on day one. Put the project deadline and your internal milestones on the same calendar as your other classes. Keep your own contribution notes next to your syllabus details. And watch your grade picture: the evaluation may adjust a project score, but that score sits inside a course grade with its own weights. A strong participation or project component can be worth more than students assume, as we show in our piece on tracking participation grades.
One more point from experience: students underestimate how much a single group project can shift a class average. Before you decide how hard to push on a disputed rating, work out the actual stakes. A grade goal projection shows what you need on remaining work to hit your target, which turns a heated argument into a concrete number.
— Haden
Keeping the project, and your grade, on track
Group work usually lands in the middle of a crowded stretch: midterms, other deadlines, work shifts. Semester Flow reads your syllabi, extracts due dates, exams and grading policies, and syncs them to Google Calendar, so a project deadline and the checkpoints you set sit beside everything else. It also tracks your GPA and course grades in real time and flags classes that are slipping, which helps you judge how much a group project weighs in your term.
You can start on the free trial. The Pro plan is $7.99 per month, and a Parent plan is $15 per month. Details are on the pricing page. If you manage academics for a chapter or club, see how organizations use Semester Flow.
After the evaluation: what to do with the result
When grades post, look at three things:
- The weighting. How many points did the peer component affect?
- Your comments. If instructors share anonymized feedback, note one habit to keep and one to change.
- Your course standing. Run the new grade through your syllabus weights to see where you now stand.
If the result looks wrong, email your instructor promptly with your log and a short, factual request to review it. Do not argue about individual teammates. A midterm grade check routine and a structured end-of-term reflection are good places to record what you would do differently next time.
FAQ
Should I be honest if my friend did less work?
Yes, and specific. Rate the behavior with dated facts, and acknowledge what they did contribute. Inflated ratings for friends shift the cost onto teammates who did more, and they weaken the evaluation for everyone.
Are peer evaluations anonymous?
It depends on the course. Many are confidential from teammates but visible to the instructor, and some tools anonymize comments. Read your syllabus or ask your instructor before you write anything you would not want attributed to you.
How much can a peer evaluation change my grade?
That is set by your instructor. In some courses it is a small participation component, and in others the ratings produce a factor multiplied by the team grade. One published approach caps the factor at a value like 1.05 while still penalizing low contributors, but your course may differ.
What if I think my teammates will rate me unfairly?
Keep a contribution log with dates and links to your work, and tell your instructor early if conflict is building. Documentation is far more persuasive than an explanation after the fact.
How often should groups do peer evaluations?
More than once is better. Guidance for instructors recommends an early formative review so teams can correct course before final grades are set.
Sources
- Responding to student feedback: individualising teamwork scores based on peer assessment, Monash University
- Cooperative learning and peer evaluation: the effect of free riders on team performance
- Peer assessment, Teach HQ, Monash University
- Best Practices in Peer Assessment, PMC
- Individual Contributions to the Group, Center for Educational Innovation
- Peer Evaluations, Cornell Computer Science
- How to evaluate group work, Cornell Center for Teaching Innovation
- Peer assessment in small groups, University of Northern Iowa