Research communities are unanimous in acknowledging the value of peer review, but there’s a growing desire for more transparency in the process. As part of that, researchers want to see how publishing decisions are made, and they want greater assurance that referees and editors act with integrity and without bias.
For many journals, including Nature, peer review has typically been single-blind — that is, authors do not know who is reviewing their paper. At the same time, the contents of peer-review reports, and correspondence between authors, reviewers and editors, are kept confidential.
This prevents readers from seeing the often fascinating and important discussions between authors and reviewers, which are crucial in shaping and improving research and checking its integrity. Keeping these debates confidential also helps to reinforce perceptions that the research paper is the last word on a subject — when the latest finding is often simply a milestone along the scholarly journey.
Our authors have told us they want change. In a 2017 survey of Nature referees, 82% agreed that standard peer review ensures high-quality work gets published. But 63% said publishers should experiment with alternative models, and more than half said peer review could be more transparent — and expected publishers to do more to make it so.
Along with many journals, we have begun to open up the peer-review process. Four years ago, Nature invited referees to be acknowledged in papers — with the consent of both author and reviewer. Around 3,700 Nature referees have chosen to be publicly recognized, and around 80% of the journal’s papers have at least one referee named.
Beginning this week, authors of new submissions to Nature will be offered the option to have anonymous referee reports published, along with their own responses and rebuttals, once a manuscript is ready for publication.
Those who agree to act as reviewers should know that their anonymous reports — and their anonymized correspondence with authors — might be published. Referees can also choose to be named, should they desire.
In making this change, Nature is following seven other Nature Research journals. They are: Nature Biomedical Engineering, Nature Cell Biology, Nature Ecology & Evolution, Nature Human Behaviour, Nature Immunology, Nature Microbiology and Nature Structural & Molecular Biology. And in taking this step, we’re joining the pioneering efforts of The EMBO Journal and BMC journals — and, more recently, Nature Communications, which has been publishing reviewer reports since 2016.
We will report back as the trial progresses, but the experience of Nature Communications has been positive. In 2018, the overwhelming majority (98%) of the journal’s authors who had published their reviewer reports told us they would do so again. Last year, almost 70% of research in the journal was published with referee reports and authors’ responses. This figure represented 62% of papers in chemistry, 58% in physics, 74% in Earth sciences and 74% in life sciences.
But our surveys have shown that not all researchers support visibility for reviewer reports. One concern is that, even with anonymity, reviewers might be less critical — perhaps expecting reciprocal treatment for their own papers. Or some could spend many hours striving to produce a ‘perfect report’, adding to the burden of peer review. The experience from Nature Communications negates this concern. The vast majority of our reports are already written in a professional and constructive manner — greatly enhancing the integrity of our research papers — so we do not want our reviewers to change what they do now.
Research published in the Nature Research journals is subjected to high levels of scrutiny before it is considered acceptable for publication — and as editors, we see our authors’ tremendous effort to make their manuscripts technically robust. We feel these efforts deserve to be seen. Published peer reviews are intended to advance scholarly discussion about a piece of research — dialogue that continues after publication − and it is important that our readers and the research community at large can benefit from such discourse. We are pleased to help make that happen.
Nature 578, 8 (2020)