Your first peer review: a short guide

Nobody is taught how to review a manuscript; most of us learn by doing it. This guide covers the steps that matter most, in the order you will meet them.

1. Before you accept

Read the abstract in the invitation and ask yourself three questions:

• Do I know this area well enough to judge the method and the conclusions?
• Can I finish within the deadline, usually 21 days?
• Is there anything that could make me less than impartial? This includes a close collaborator, a competitor, or a manuscript I think I recognize.

If any answer is no, decline. You can add a short reason and suggest a colleague. Declining promptly is a real help to the editor.

2. First read: the whole paper

Read the manuscript once from start to finish without taking notes on details. Aim to answer two questions:

• What question does the paper ask, and is it worth asking?
• Does the paper, as a whole, answer it?

Your most useful comments usually come from this read.

3. Second read: the method and the evidence

Now read slowly. Check that the design fits the question, that the sample and procedures are described well enough to repeat, and that the analysis suits the data. Check that each conclusion follows from the findings.

Look up a few key references, especially where a claim depends on them. You do not need to check every citation, and you do not need to correct language line by line.

4. Write the report for the authors

A clear structure helps the authors act on your comments:

• Summary: two or three sentences on what the paper does, in your own words. It shows the authors you understood their aim.
• Major issues: numbered, each with what the problem is, why it matters, and what would solve it.
• Minor issues: in order, with line numbers from the PDF.

Be specific and kind. “The sampling is unclear” is hard to act on. “Lines 112–118 do not say how the 26 participants were selected or why recruitment stopped there” is easy to act on.

Do not ask the authors to cite your own work unless it is truly needed. Do not try to work out who they are.

5. Answer the questions and choose a recommendation

Our form asks five questions, each answered yes, partly, or no:

• Are the topic and aim clear, and is the contribution to the field evident?
• Is the method appropriate and adequately described?
• Are the findings and discussion consistent, and are the conclusions supported?
• Is the literature current and sufficient, and are the sources cited correctly?
• Are the language, style, and structure clear?

Then choose a recommendation:

• Accept: ready as it is, or nearly so.
• Minor revision: sound, with changes the editor can check without another full review.
• Major revision: potentially publishable, but the method, analysis or argument needs substantial work. You may be asked to review it again.
• Reject: the problems cannot be fixed by revision, or the paper does not fit the journal.

The recommendation is advice to the editor; the decision is the editor’s. Make sure your comments explain your choice. A “major revision” with only two minor comments leaves the authors confused.

6. The confidential note to the editor

Use this box for what the authors should not read: a suspicion of plagiarism or duplicate publication, an ethics concern, or a limit to your own expertise (“I could not judge the statistical model”). It is not the place for a harsher verdict than the one you give the authors.

7. Confidentiality and AI tools

The manuscript is shared with you for this review only. Do not pass it on, and do not upload it or your report to AI tools, because that would break the confidentiality of the review. The assessment must be your own.

8. After you send

The editor combines your report with the others and writes to the authors. If you said you would review the revised version, you will receive it together with the authors’ response to your comments. Your certificate will be ready on your dashboard as soon as you send the report.

Your first review will take longer than later ones. A report that is clear, fair, and specific is more useful than a long one.

Would you like to review for our journals? Tell us your topics on the Review for our journals page: https://mjournals.org/get-involved/reviewers