Peer assessment asks students to evaluate each other's work against shared criteria, and done with structure it becomes one of the most reliable ways to deepen learning. This guide covers the methods you can use, a simple form you can copy, and examples you can show your students.

Peer assessment asks students to evaluate each other's work against shared criteria. Done casually it can feel like a popularity contest. Done with structure it becomes one of the most reliable ways to deepen learning, because judging someone else's work forces you to understand the standard you are judging against. This guide covers what peer assessment is, the methods you can use, a simple form you can copy, and worked examples, so you can run it with confidence.
Peer feedback is the comments students give to help work improve. Peer assessment goes a step further and applies criteria to reach a rating or a score. The two often run together, and the strongest activities combine honest comments with a clear rubric. Peer to peer assessment is the same idea seen from the student's point of view, where the responsibility for judgement is shared across the group rather than held only by the instructor. That shift, from treating criteria as something done to them to something they apply themselves, is where the learning sits.
The evidence is encouraging. A meta-analysis of 54 studies by Double, McGrane and Hopfenbeck (2020) found peer assessment improves academic performance compared with no assessment and with teacher assessment alone. Nicol, Thomson and Breslin (2014) showed why, finding that producing a review engages students in repeated acts of evaluative judgement, because they have to apply criteria and explain their reasoning. The practical lesson is to treat reviewing as the learning activity, not as a chore on the way to a grade. Our piece on the impact of self and peer grading on student learning explores how that plays out in practice.
Peer assessment for students earns its place in a few specific situations. It works well on drafts, where a second reader before final submission catches problems while they are still cheap to fix. It works on group projects, where peers see contributions an instructor never witnesses. And it works on open tasks such as essays, presentations, and designs, where there is no single correct answer and exposure to different approaches is part of the learning.
A reliable method is mostly about removing ambiguity before students start. Define the criteria first and share a rubric so students know what they are looking for. Model what good looks like with one or two examples. Decide on anonymity, since anonymous peer review tends to produce more honest and less personal comments. Run a calibration round where everyone assesses the same sample piece and then compares notes, because that single step prevents most early confusion. Then open the real activity with enough time for students to read carefully rather than rush.
A good peer assessment form keeps students focused on reasoning rather than only scoring. For each criterion, ask for three things.
CriterionRatingOne strength and one suggestionClarity of argumentNot yet / Developing / StrongWhat worked, plus one concrete changeUse of evidenceNot yet / Developing / StrongWhat worked, plus one concrete changeStructure and flowNot yet / Developing / StrongWhat worked, plus one concrete change
The suggestion column matters most. A score tells the author where they landed, but the suggestion tells them how to move. You can adapt this same shape for essays, presentations, lab reports, or group contributions by swapping the criteria while keeping the three parts. Building the criteria is faster with AI rubrics as a starting point you then refine.
It helps to show students the difference between a comment that helps and one that does not. Instead of "good intro," a stronger version reads, "Your opening framed the problem before the solution, which pulled me in. Add one sentence on why it matters now so the stakes land earlier." Instead of "confusing," try, "I lost the thread in the third paragraph where two claims competed. Splitting them into separate points would make each easier to follow." For a full set of templates students can copy, see peer feedback examples for students.
Three problems account for most failed peer assessment. The first is vague comments, which you fix with criteria, sentence frames, and prompts that ask for reasoning. The second is the popularity effect, where students reward each other rather than the work, which you fix with anonymity and a rubric that anchors judgement to evidence. The third is logistics, which is what a Peer Review tool is for, since it distributes work, protects anonymity, and gathers scores without manual effort.
If you want to run peer assessment without building the workflow by hand, our Peer Review tool handles the heavy lifting inside your LMS. To start from a proven design rather than a blank page, the activity onboarding students to peer review is a good first low stakes run, and developing critical writing skills through peer review and reflection suits written work. And to pair assessment with structured reflection, reflection on learning objectives with the SOLO rubric helps students judge their own progress against the course outcomes.
Pair this with peer feedback examples for students for the language to share, and see the full hub on peer review. For the research and a semester one roadmap with ready to use templates, our ebook A guide to authentic peer assessment in higher education is the natural next read, or join the live discussion in our webinar Assessment Rebooted: peer review and authentic learning in a post-AI classroom.