A beginner’s guide to peer review: Part Three

Previous posts in this series covered how to decide whether to accept an invitation to peer review and how to begin the peer review process. This post will cover how to write a good peer review report.


If you followed the checklists in previous posts in this series, you should have a list of what you think was done correctly or well and what concerns you. You should be able to use your expertise to determine whether each concern falls under the ‘major essential’, ‘minor essential’, ‘discretionary’ or ‘confidential’ category.

When you write your report, don’t just state that something is wrong or flawed. Explain why and what, if anything, can be done to address your concern.

This will make a much more useful report for both the editor and authors and will help to inform your final recommendation.                                                                                                                                                   

How to make a recommendation

This is a crucial part of the peer reviewer report. This is where you give your ‘bottom line’ on what you think should happen next.


– The journal does not want to publish research that is scientifically flawed.

– The journal wants to publish research that fits its scope and threshold.

Reject, accept or revise?

If you think the study is methodologically sound, fits the journal scope and just needs a few revisions, the recommendation is easy. Similarly, if you’ve spotted a major flaw in the methodology that cannot be addressed without redoing the entire study, you can confidently reject as unsound.

You often have to decide between ‘reject’ or ‘needs major revisions’ and this is a distinction that can be difficult to make.

However, manuscripts rarely fit neatly into one of these two categories. You often have to decide between ‘reject’ or ‘needs major revisions’ and this is a distinction that can be difficult to make.

If you asked yourself, ‘What is my concern?’ ‘Why is it a concern?’ and ‘Can my concern be addressed?’ during the review process, it will help you to decide what to recommend.

For example

  1. Is the study basically sound, but needs revision such as some additional information, reorganization or clarification?
    If yes, you can recommend revisions.
  2. Is the study limited in some way, for example conducted correctly but too small, can only be applied to a specific population, cannot conclude as much as the authors do? Can this be addressed by rewriting the conclusions or highlighting the limitations?
    If yes, you could recommend these revisions, but let the editor know if you are not sure whether a revised version of the manuscript will still meet the threshold of the journal.

A note on extra experiments

Authors are usually given a short time (2 to 4 weeks) to revise their manuscript. So, if you think that further experiments need to be done, consider whether the authors can realistically complete the necessary studies within that time. If you don’t think they can, explain to the editor, what is wrong, what needs to be done and how long you think it will take. The editor can then decide whether to reject the manuscript or allow the authors extra time to revise.

How to re-review

If you recommended major revisions, it is likely you will be asked to look at the manuscript again to judge whether the authors have revised adequately. The authors will have written a ‘point by point’ response to your report. They may have rebutted some of the concerns you raised.

  • Do focus on how well your original comments have been addressed, either with changes or thoughtful replies.
  • Don’t take rebuttals of your concerns personally. Consider whether they are valid. Did you misunderstand what was done? Has the authors’ explanation removed any need for concern?
  • Don’t needlessly raise new points. This is unfair to the authors and reflects a lack of thoroughness in your original review. If you do spot something that you missed in your first review that you think should be addressed, do raise it, but acknowledge your own mistake in missing it first time around.
  • Don’t ask for further experimentation. It is unfair to ask for further experimentation at this stage of peer review. This should have been raised at an earlier stage.

How to raise concerns about ethics and suspected misconduct

Make sure you are familiar with standards for research and reporting in your field. The journal policies should have all the information you need to know about the standards the journal expects the authors to adhere to.

Make sure you are familiar with standards for research and reporting in your field.

It is usually appropriate to raise your concerns in the non-confidential section of the peer review report. For example, if there is no mention of ethics committee approval in the manuscript and you think the study clearly requires ethics approval, you can ask the authors about this in the main report.

However, you may spot or suspect something that could constitute misconduct such as plagiarised text or an undeclared competing interest. If you suspect any sort of misconduct, always raise your concerns in the confidential comments to the editor.

Be as specific as possible and provide evidence if you can, so that the editor can take the matter forward. There are standard procedures for how journal editors and publishers deal with allegations of misconduct as recommended by the Committee on Publication Ethics.

Peer reviewer misconduct

Finally, don’t forget that there is such a thing as peer reviewer misconduct

The Committee on Publication Ethics have created guidelines for peer reviewers which provide more information on how to be an ethical peer reviewer.


If you are an early career researcher and have felt unconfident about peer reviewing, we hope this series of blogs has encouraged you to take the plunge.

Please do give us feedback on whether you think the series has been useful and on any aspects on which we should expand, by either commenting below or emailing us at blogging@biomedcentral.com.

View the latest posts on the Research in progress blog homepage


Martin Černý

I think a good review does not contain only flaws of the paper, but also what is good about the paper. I also think that asking yourself “Why should this paper be published?” before asking “Why this paper should NOT be published?” leads to a better view of the research. In any case, starting the review with the strong points of the paper is useful. At the very least it signals that you are not just a grumpy guy who takes joy in showing other people what they did wrong, which in turn makes the authors more open to your recommendations.

Comments are closed.