Resources for referees

1. Start with a brief summary: An outline of the gist of a paper you are reviewing can be very helpful with clarifying in your own mind the authors’ line of argument. Your summary also gives authors a different perspective on their own work, which can help define their focus, as well as a chance to discover, and then counter, any misunderstandings that may have occurred.

2. Back up your assessments: Where you think the presented work is not novel, it helps editors and authors if you provide references where similar findings have been reported previously. If you have technical concerns, spell out where exactly you think the flaws lie, and what you think has been missed. The authors may be able to address these concerns—but only if they understand them.

3. Keep the tone scholarly, encouraging, and positive: Peer review is first and foremost intended to help improve the scientific literature. That is best done by keeping your comments dispassionate¹. Confrontation, emotive language, or sarcasm are unhelpful; instead, choose a tone you would use with a friend (whether or not you know the authors).

4. Consider and comment on each claim: Most papers present more than one conclusion. If parts of the paper don’t hold up, perhaps others do? As editors, we have the task to balance our editorial requirements for novelty and importance of a piece of work with the comments from a panel of reviewers with different perspectives. With your advice on which claims by the authors are compelling, we can decide whether these claims advance science sufficiently to fit the journal’s aims.

5. Be generous with ideas for improvement—but do not insist: No paper is perfect, and there is always more that could be done. So if you have suggestions for how the authors could improve their paper, do share them—but do not expect or demand that every idea will be implemented. It is the authors’ paper after all.

6. Note where you cannot follow: Most likely, as a reviewer you know more about the subject area than many other readers. So, if you don’t understand, others will run into the same difficulty. Outline what you find confusing. Ask the authors to explain what they mean where the writing is unclear, in particular, where it’s central to the conclusions.

7. Be open about the limits of your expertise: If only part of the manuscript falls within your specialty, let the authors and editors know. It is not always obvious to editors what exactly a prospective reviewer has direct knowledge of, but where we know, we will weight comments accordingly.

8. Do not request citations to your own group’s papers, unless essential: Because you will be most familiar with your own work, it is easy to rely on it in your review beyond its fair share of the literature. Try to counterbalance that tendency by carefully considering what else has been published. Only ask for your own articles to be cited if they are obviously key to the story. In those cases, ideally explain in the notes to the editors why you think this is the case.

9. Judge the science, not the scientist: We strongly recommend that you disregard what you know about the authors. Where they chose double-blind peer review, we suggest that you do not dwell on guessing their identities, and certainly that you do not hold that against the authors. What matters should be the quality of the research, and not who wrote it or which research institution it came from.