The first review invitation usually lands in your inbox with a two-week deadline and no instructions beyond a link to the manuscript. Nobody formally teaches you how to peer review a paper: most academics learn by imitating the reviews they have received, good and bad. That is a shaky foundation, because a review is not a verdict handed down from on high—it is a structured professional judgment with conventions you can learn in an afternoon. This guide walks through the whole process, from deciding whether to accept the invitation to choosing a recommendation, so that your first review reads like your fifteenth.
The advice below applies to journal manuscripts, conference submissions, and proceedings papers alike. If you have not yet received an invitation, start with our guide on how to become a peer reviewer, then come back when the first manuscript arrives.
What Your Review Is Actually For
A peer review serves two audiences at once, and confusing them causes most beginner problems. The editor needs a clear, justified recommendation: is this work sound, is it a fit for the venue, and what would it take to make it publishable? The authors need a roadmap: what specifically must change, in what order of importance, and why?
Notice what is not on either list. You are not there to rewrite the paper in your own style, to demand the study you would have run instead, or to demonstrate how much you know. A useful review answers one question above all others: does the evidence presented support the claims made? Everything else—structure, clarity, citations—matters only in service of that question.
Before You Peer Review a Paper: Four Questions to Ask
Accepting the invitation is itself a professional judgment. Run through four checks before you click yes:
- Am I qualified enough? You do not need expertise in every aspect of the paper. If you can competently evaluate the core contribution, accept and tell the editor which parts fall outside your expertise.
- Do I have a conflict of interest? Recent co-authors, close friends, direct competitors on the same question, or anyone whose work the paper attacks or praises at length—if the connection would make a reasonable person doubt your neutrality, decline and say why.
- Can I meet the deadline? A late review delays real people’s careers. If you need an extra week, ask up front; editors almost always prefer a realistic date to silence.
- Is the paper actually in my field? Editors sometimes guess wrong. A quick skim of the abstract before accepting saves everyone time.
One more thing to check: whether the review is anonymized in one direction or both. Many venues, including NIVA’s conferences, anonymize both authors and reviewers; our explainer on how double-blind peer review works covers what that changes about the way you read and write.
How to Peer Review a Paper in Six Steps
1. Read Once Without Taking Notes
Your first pass should mimic an ordinary reader’s experience. Read the manuscript start to finish without stopping to annotate. You are collecting impressions: What is the paper claiming? Did the argument carry you along or lose you? Where did you feel confused, skeptical, or bored? Those reactions are data—the same places will trip up other readers.
2. Let It Sit, Then Read Again With Questions
Wait at least a day if the deadline allows. On the second pass, read slowly with the venue’s review criteria open beside you, and annotate as you go. Now you are verifying rather than absorbing: checking whether the methods match the research question, whether the results support the discussion, and whether the paper delivers what the abstract promises.
3. Evaluate the Core Contribution
Every publishable paper has a center of gravity: a finding, an argument, a method, or a synthesis that justifies its existence. Name it in one sentence. If you cannot, that is your most important comment. If you can, ask whether it is new enough and significant enough for this particular venue—a question about fit, not just quality.
4. Check the Methods and Evidence
This is where your disciplinary training earns its keep. Quantitative work: is the design capable of answering the question, is the sample adequate, are the analyses appropriate? Qualitative and humanities work: are sources handled fairly, is the interpretive framework applied consistently, does the argument engage obvious counterexamples? Flag what is broken, but also distinguish fatal flaws from fixable ones—the difference drives your recommendation.
5. Note Presentation Issues Separately
Typos, awkward phrasing, unclear figures, and formatting slips belong in your review, but in their own minor-issues list. Never let a paper’s surface problems dominate a review of its substance, and never copyedit line by line—a representative handful of examples plus a note that the manuscript needs careful proofreading is enough.
6. Draft, Sleep, Revise, Submit
Write your full report, then leave it overnight. Reread it the next morning asking one question: if my name were attached, would I send this? Soften anything sarcastic, cut anything showy, sharpen anything vague, and submit before the deadline.
Structuring the Written Report
Editors read dozens of reviews a month, and a predictable structure helps them act on yours. The standard shape has four parts:
- Summary paragraph. Two to four sentences restating the paper’s aim and contribution in your own words. This proves you understood the work and gives authors a chance to see where they were misread.
- Major issues. Numbered points that affect the validity or significance of the work. For each: what the problem is, where it occurs, why it matters, and—when possible—what would fix it.
- Minor issues. A numbered list of smaller items keyed to page or line numbers: unclear sentences, missing citations, figure labels, terminology slips.
- Confidential comments to the editor. Anything the authors should not see: suspicions of plagiarism or data problems, honest limits of your expertise, or context about your recommendation.
Numbering matters more than it seems. Authors respond point by point in their revision letters, and numbered comments make that exchange orderly instead of chaotic.
Choosing a Recommendation
Most venues offer four options, and first-time reviewers routinely agonize over the boundary between them. This table maps the usual logic:
| Recommendation | When it fits | What your comments must do |
|---|---|---|
| Accept | Sound, significant, and clearly written; only trivial fixes remain | Confirm the contribution and list any final polish items |
| Minor revisions | Sound core; fixes are clear, bounded, and need no re-review of substance | Give a precise, finite list the authors can simply work through |
| Major revisions | Promising core, but validity or framing problems need real work and a second look | Explain what must change and what success would look like |
| Reject | Fatal flaw in design or evidence, or a fundamental mismatch with the venue | Justify the decision respectfully and point toward a better path |
Two habits keep recommendations honest. First, make sure your comments and your recommendation tell the same story—a glowing review with a reject verdict, or a brutal review with minor revisions, leaves the editor guessing. Second, remember the recommendation is advice, not a ruling; editors weigh multiple reviews and sometimes overrule all of them.
Getting the Tone Right
The reviewer’s golden rule is simple: write the review you would want to receive on your own hardest day. Critique the work, never the author. Replace verdicts with observations—instead of declaring the analysis sloppy, note that the analysis omits a control for class size and explain why that matters. Specific criticism is kinder than vague criticism, because it can actually be acted on.
If you have ever drafted a response letter yourself—our templates for responding to peer review comments show the author’s side of this exchange—you know how long a single dismissive sentence can echo. Praise honestly where praise is earned; a review that names what works is more credible when it names what does not.
Mistakes First-Time Reviewers Make
- Reviewing the paper they wish existed. Evaluate the study the authors ran, not the different study you find more interesting.
- Demanding new experiments reflexively. Ask for more data only when the paper’s own claims require it, and say which claim fails without it.
- Burying the verdict. If the paper has a fatal flaw, say so early and plainly instead of hiding it at comment fourteen.
- Copyediting instead of reviewing. Forty grammar corrections and no comment on the method helps no one.
- Trying to sound like a reviewer. Borrowed severity is the most common beginner voice. Plain, direct, specific sentences are what experienced reviewers actually write.
Where First-Time Reviewers Can Start
Journals are not the only venues that need reviewers, and they are rarely the easiest place to begin. Conference review is shorter, faster, and more forgiving: you evaluate an abstract or a short paper rather than a forty-page manuscript. Organizations that publish reviewed conference proceedings depend on a steady supply of early-career reviewers, and NIVA’s journals likewise welcome graduate students and faculty who want to build reviewing experience across disciplines. Each review you write teaches you something about your own writing—usually the exact weakness you keep flagging in others.
Frequently Asked Questions
Plan on four to eight hours for a full journal manuscript: two reads, note-taking, and a written report. Conference abstracts and short papers take far less, often under an hour each. Speed comes with experience, so budget generously for your first few reviews.
Yes, and you should when you lack the expertise, have a conflict of interest, or cannot meet the deadline. Decline promptly so the editor can move on, and suggest an alternative reviewer if you know a suitable colleague.
Almost never. Peer review is volunteer service to the scholarly community, though it counts as professional service on your CV, builds your reputation with editors, and sharpens your own research and writing. A few publishers offer small honoraria or discounts, but that remains the exception.
Four parts: a short summary of the paper in your own words, numbered major issues that affect validity or significance, numbered minor issues keyed to pages or lines, and confidential comments to the editor. End with a recommendation that matches the tone of your comments.
Reviewing sharpens your judgment; presenting puts it to work. When your own research is ready for an audience, submit a proposal to NIVA’s next virtual conference—faculty register for $150, students for $70, and NIVA membership is free.