Your email brings an invitation from the editor of a reputable journal in your field to review a manuscript. You’ve never done a review before. Even if you have yourself published a peer-reviewed journal article, the whole review process is pretty much a black box. What to do?

Herewith some thoughts drawing on my five years experience as English Co-editor of the Canadian Journal of Political Science.

First off, you should be pleased to be asked. Aside from giving you a good excuse to put off sorting out that tricky data interpretation issue in Chapter Four of your thesis/thesis revision, recognize that you’re being asked to contribute to an integral element of the academy: peer review. More significantly, appreciate that you are being recognized as an expert whose opinion on weighty academic matters is valued by a journal editor in your field.

Fine. You’re suitably chuffed, as the Brits say. But how to proceed? A key precept to bear in mind at all times: review unto others as you would have others review unto you. When you submit a paper to a journal, you expect to receive insightful, constructive, professional and timely reviews of your work. Strive to do exactly that in your review.

Let’s begin with timely. Editors don’t expect to hear back from you agreeing or declining to do the review instantaneously. But it’s reasonable to expect your response within two or three days. Most invitations to review include the paper’s abstract. It should be fairly clear from the abstract as to whether this is something you can handle. Don’t be modest – the editor asked you for a reason (unless there’s some mix-up: “Oh, you’re not that Jane Jenson?”). Once you agree, you’ll likely get an email with instructions for downloading the manuscript. If, having done so, you realize that you can’t do the review – it really isn’t in your area of expertise; you have a conflict of interest because you know who the author is and have worked with her; and so on – it’s fair ball to tell the editor you have to bow out. If you really don’t think you can’t deliver a review in the assigned time, or close to it, say so and withdraw.

It’s good that you’ve been timely in letting the editor know you’ll do the review. More important is delivering the review in timely fashion. Different journals set different deadlines for reviews, typically four to six weeks. What happens if you miss the deadline? If you’re just a few days late, no one will get excited. However, if the deadline is weeks past and you have yet to deliver, in addition to various automated reminders, you will have a frustrated, annoyed editor on your case. That’s not going to harm your career … is it? For me, the real issue is that you’ve failed to live up to an important commitment you made, in formal terms to the editor but in real terms to the author.

You’ve read the paper once or twice. I can’t advise on important issues such as strength of conceptual/theoretical framework, nature and quality of data and sophistication/originality of analysis. That’s why you’ve been asked to provide an assessment. You’re the expert. However, here are some things to consider in putting your review together.

  • Be aware of the journal’s remit and audience. Some journals, like CJPS, are omnibus journals that will publish articles in any subfield of the discipline, though even with CJPS some submissions may be so specialized that they would be better placed in a journal dedicated to a geographic region or a narrow subfield. The editor will have made a preliminary assessment of the paper’s fit for the journal before sending it to you (rather than rendering the dreaded ‘desk reject’). Still, ask yourself whether this is a paper that the journal’s readers will find of interest and value.
  • Many journals provide reviewers with templates to structure their reviews. These can be helpful and you should make use of them … but you don’t have to follow them precisely.
  • You don’t have to write a thesis (presumably the one you’re working on, or revising for publication, is enough) but you do have to provide the editor with a reasoned explanation for your recommendation, which should be, whatever the editor’s decision, helpful to the author. Avoid “I recommend publishing this paper; it’s good and we need more of this kind of article out there” or “I recommend rejecting this paper; it’s not very good”. Neither is helpful to the editor – and yes, over the years I received a few like that – and indeed is a disservice to the author who is looking for constructive criticism.
  • Don’t just say the paper has weaknesses (or strengths). Demonstrate them and, if appropriate, offer suggestions as to how to deal with them. Be as concrete as you can in making your constructive criticisms. Don’t just say ‘the paper needs a better theoretical framework’; offer some ideas as to what that might entail.
  • Be reasonable. Only so much can be done in a short (8/9000 word) article. The editor is not likely to offer much extra space for authors to deal with criticisms. Focus on the paper’s main ideas, approaches and arguments rather than possible tangents the author could explore.
  • Relatedly, avoid the temptation to apply the ‘is this the article I would have written?’ criterion. You didn’t write it. Accept that others will have different takes on particular subjects, including those near and dear to your heart (or thesis).
  • Don’t show off. You may have read that just-published article in the Journal of Seriously Obscure Political Studies while the author hasn’t cited it, but mention it only if it really is relevant to the paper. If the author seems to have missed significant literature, say so, but don’t list off sources just to show how well read you are.
  • Use civil, professional language. Don’t be nasty. Even if the paper is patently subpar, convey that without offensive language. Bad enough for an author to have a paper rejected without being belittled in offensive, condescending terms. A few – happily, very few – times I had to tone down or edit reviewers’ comments (‘this submission reads like an undergraduate paper’). Many journals offer reviewers the opportunity to submit comments to the editor that the author won’t see; again, don’t be nasty but if you feel the need to emphasize a point to the editor, that’s an option.
  • What about the writing? Is it clear and coherent? If the submission is riddled with typos and grammatical glitches say so but you don’t have to enumerate them. Pointing out an egregious example or two could be in order.
  • Point out factual errors or clear misinterpretations; no need to comment on them.
  • Be honest in your recommendation. I don’t mean you shouldn’t recommend rejection lest the article scoop your work. Journals typically, as CJPS does, offer reviewers three choices: accept (which can include minor revisions), reject, and revise and resubmit. The vast majority of reviews I received as editor recommended ‘R&R’ even when the text of the reviews were quite negative, perhaps because most of my reviewers were Canadian and wanted to be the stereotypical ‘nice’ Canadian. Most submissions that came across my virtual desk had the potential, given enough work, to be publishable, but editors (and reviewers) don’t have the time to nursemaid every weak but potentially publishable manuscripts to fruition or the space to publish them. So unless you see genuine potential for successful revision, and you offer guidance for revision, better to recommend rejection.
  • Understand that if the editor decides to offer the author an ‘R&R’ you will likely be asked to assess a revised manuscript, even if – perhaps especially if – you recommended rejection. Here it’s perfectly okay to be very brief: ‘the author has adequately/effectively responded to all/most of my main criticisms and I now recommend publication’. If you don’t think the revisions are adequate, a somewhat more fulsome commentary will be needed.

Doubtless I could mention other things. What’s that? How long should my review be? No standard answer. I’d guess that most reviews I received at CJPS were in the range of 200-400 words and that was fine. Certainly you don’t have to go on at great length, though if you believe you have useful comments/advice by all means don’t feel constrained by word limits.

Finally, use the experience to learn. Most journals, CJPS included, send reviewers the editor’s decision on a submission, including all reviewers’ comments (save the ‘editor only’ comments). This is a very useful way to learn about the process and to compare your assessment against those of other reviewers.

Graham White is professor emeritus of political science at the University of Toronto and adjunct research professor of political science at The University of Western Ontario.