China
Africa
Peer review
This page is part of Taylor & Francis and Sense about Science's research integrity toolkit.
Peer review is an essential process: experts assess the quality, validity (whether it was conducted in line with norms and standards), and originality of research before it is published.
It plays a crucial role in maintaining research standards and preventing the dissemination of flawed or fraudulent research. However, several challenges can hinder the effectiveness of peer review.
Sense about Science collaborated with editors and researchers to produce a nuts and bolts guide to peer review, which opens up the black box of peer review, sharing editor insights, the different types of peer review, and some of the challenges that can hinder its effectiveness.
One key issue is the prevalence of bias, such as favoritism towards certain research topics, an ideological slant, authors, or institutions, which can lead to unfair evaluations.
[I’ve seen] weaponizing of the peer review process where there's overlap in a peer reviewer's work, so they give extra negative feedback or suggest further experiments to delay the process so they can get their research out first.
There are other challenges, for example, in open peer review, early career researchers in particular can be hesitant to criticize research papers/practices of more established researchers in their field because of perceived potential backlash. While in double blind peer review (where reviewer and author are both anonymous), there is an increased onus on editors to address potential biases.
Another issue is inconsistency in the quality of reviews, as the process can be subjective and not all reviewers are equally skilled in identifying flaws or providing constructive feedback. Researchers may take on their first peer review under a mentor, but the level of support offered is variable.
Learning to review journal papers is a key skill, but there isn't much formal support on how to do it. I found it particularly difficult to judge what level of detail was needed, or what elements to focus on, as peer reviews are not typically accessible or published... nor did I have any first-hand experience of being on the receiving end. I can imagine the experience of reviewing, and quality of the review itself, varies dramatically between PhD students and depends heavily on the support from their supervisor.
These issues can result in flawed research being published, which undermines trust in the scientific process. It is key that you are aware of these challenges and work to ensure transparency, fairness, and objectivity in the peer review process, both as authors and reviewers.
Reviewing is a role that is integral to the research process, so it is important that early career researchers get involved in the process early on:
- Contact a journal editor or ask a senior colleague to recommend you – journal editors are always looking out for new reviewers, especially those with expertise in areas under-represented in their pool of contacts. If there's a journal that you read regularly, email the editor directly or ask a colleague to pass on your details.
- Look out for calls for reviewers – some journals circulate calls for reviewers, for example if the journal is new or expanding its scope into a different area.
- Register with a journal's publisher – some publishers invite aspiring reviewers to add their details to a reviewer database.
- Find a mentor – ask a senior colleague, with experience in reviewing, whether you could work with them on a review.
- Raise your visibility on researcher networking sites – academic networking sites, such as ResearchGate or Academia.edu, are places to build a profile where editors looking for new reviewers can find you.
- Sign up for a certified training course that supports researchers in their understanding of peer review. Details of courses can be found on page 25 in Sense About Science's nuts and bolts guide.
AI in peer review
Reviewers must not upload text or images from an unpublished manuscript into a generative AI tool, as this may infringe upon intellectual property rights.
Generative AI may only be used to improve the language of a peer review report.
Built with Shorthand