Peer Review Pairing Generator

What this generator does

Produces a directed review ring rather than a set of partners. The pair generator makes mutual pairs, where two people work together; here the allocation is one-way, so the person you review is not the person reviewing you. That difference matters in assessment, because reciprocal pairs invite reciprocal generosity.

How to use this tool

1. Enter the names of everyone taking part, one per line.
2. Choose how many rounds you need — one for a single review, more for a course that repeats it.
3. Press Make pairs to see who reviews whom.
4. Print the ring or copy it into your feedback sheet.

Understanding the controls

Names

One per line. Three is the minimum, because with two people any allocation is necessarily reciprocal — there is nobody else to review.

Rounds

How many separate review cycles to allocate. Each round avoids every partnership used in the earlier ones, so a student never reviews the same classmate twice until the possibilities run out.

Seed

Fixes the allocation so it can be reproduced. Helpful when you print the ring, lose it, and need exactly the same allocation back rather than a new one that contradicts what students were told.

Worked examples

Five students, one round

A ring such as A reviews B, B reviews C, C reviews D, D reviews E, E reviews A.

Twelve students, three rounds

Thirty-six allocations with no partnership repeated — each student reviews three different classmates.

Ten students

Four rounds are possible before anyone must review a classmate a second time.

Common use cases

• Allocating peer marking of essays or lab reports
• A writing group where each member critiques one other member's work
• Code review practice in a computing class
• Portfolio feedback rounds across a term, with a different reviewer each time
• A workshop where every participant gives structured feedback to one peer

How this generator works

Everyone is arranged into a single closed loop and each person reviews the next person around it. Because it is one loop rather than an arbitrary matching, four properties hold at once and without any retrying: everyone reviews exactly one person, everyone is reviewed by exactly one person, nobody reviews themselves, and no two people review each other. For later rounds the tool searches for a loop that reuses none of the earlier partnerships, so repeat pairings are avoided for as long as they can be.

Randomness and fairness

The loop is drawn fresh each time from the browser's cryptographic randomness, so no student can predict who will be reviewing their work — which matters more here than in most tools, because a predictable allocation invites arrangements between friends. Adding a seed trades that unpredictability for reproducibility, so only use one when you actually need the same ring back: a seeded ring is produced by a deterministic, non-cryptographic generator, and anyone who learns the seed and the class list can work out the whole allocation.

For how randomness is produced across the whole site, see how Generate Random works.

Limitations and good to know

• The number of repeat-free rounds is smaller than people expect. Each ring uses up one partnership per person, so with ten students only four fresh rounds exist, not nine — the tool states the limit before you generate.
• There is no way to keep two particular students apart from this page, so if a pairing would be unwise you will need to swap two allocations by hand afterwards.
• Review quality is not addressed at all. Allocating who reviews whom does nothing to ensure the feedback is useful, which is what a structured feedback sheet or a rubric is for.
• Everyone is treated as equally able to review anyone. Mixed-attainment pairing, where stronger writers review those who need most help, is a deliberate teaching choice this tool cannot make.

Common mistakes

Asking for more rounds than the class size allows

Check the stated limit before generating. Past it, partnerships repeat, and the tool flags exactly which ones so you can decide whether that matters.

Assuming the allocation is mutual and printing only half of it

Every student has two roles here — reviewer and reviewed — and they involve different people. Print the whole ring so both directions are visible.

Regenerating after telling students their allocation

Every regeneration produces a completely different ring. If you may need the same one back, set a seed before the first run.

Practical tips

• Use initials or candidate numbers rather than full names if the printed ring is going to be handed round the class.
• Generate all your rounds for the term in one go — the no-repeat guarantee only works across rounds produced together.
• If an absence breaks the ring, ask the reviewer of the missing student to review the person that student was due to review, which keeps the loop closed.

Privacy and your data

Names typed here are processed in your browser and never sent anywhere. Nothing is written to the address bar, so the allocation cannot leak through a shared link, and nothing is stored, so the list disappears when the tab closes. The analytics event records only the number of rounds — never a name. If you would rather not type real names at all, initials or register numbers work just as well.

Frequently asked questions

Why is this not just a pair generator?

Peer review is one-way. In a mutual pair each person reviews the other, which is exactly the arrangement that produces mutually flattering feedback. A ring separates the two roles.

How many rounds can I run without repeats?

About half the class size — precisely, the whole number part of one less than the class size, divided by two. Ten students gives four rounds.

What if someone is away?

Close the gap by having their reviewer take over their allocation, which keeps everyone with exactly one piece of work to review.