Inside scientific publishing
Scientific publishing is undergoing profound changes. Here and in future issues of EMBOencounters, we would like to discuss some of the important topics in publishing, including open access, ethics, or the merits of impact factors. The first commentary looks at ways of enhancing the quality and transparency of peer review.
Most scientists have strong opinions about scientific publishing. Concerns focus on two main issues: the speed and efficiency of publishing, and the transparency and fairness of the peer review
process. Inadequate reviews, lack of transparency in editorial decisions, unreasonable expectations for additional experimentation and multiple rounds of submission and review are amongst the most frequent sources of irritation. EMBO publishes four peer-reviewed journals – The EMBO Journal, EMBO reports, Molecular Systems Biology, and EMBO Molecular Medicine. The editors of the four journals have been working on finding solutions for these concerns and have implemented practical changes in the editorial process to address them (Table 1). I will comment on quality and transpar- ency here but I encourage readers to consult other commentaries that have been written by editors of EMBO journals.1,2
Quality of referees’ reports. As scientists, we are both authors and referees, but we sometimes act in an inconsistent way: when reviewing a manuscript, we forget how we would like referees to analyse and comment on our own papers. Conversely, when we receive referees’ reports, the smallest critical remark makes us think ‘they are out to get us’.
Even if we review papers with the greatest diligence, relevant points may escape us. We may miss important insights, or fail to detect flaws. I believe referees are often concerned about whether they have made the right recom- mendation and most scientists like the procedure by which some journals inform the referees of their final decisions and allow them to see the other referees’ reports. The EMBO journals have taken the exchange of referees’ reports one step further, to a place where it has a direct effect: As soon as all reports have been received, the editor sends them, anonymously, to all three referees who then have one day to consider the others’ views and decide if they need to alter their own review. This allows extreme opin- ions to be scrutinized at an early point, mistakes and errors to be detect- ed, and helps the editor to get back to the author with balanced decisions. More than a third of referees respond, either criticizing the remarks of the other reviewers or acknowledging that they missed key points noted by a colleague.
Transparency of decision making. Commentators and bloggers have suggest- ed that making referees’ reports public would benefit science.2,3 Indeed, The EMBO Journal introduced such a mechanism in 2008.1 The publication of the ‘peer review process files’, which includes the anonymous referees’ remarks, the editorial decision letter, and referees’ and authors’ comments on revisions, is now implemented at the four EMBO journals. The publi- cation of the peer review files for rejected papers is not a realistic option: very few authors would agree to having the rejection history of their papers publicly available since it might compromise review of their submissions elsewhere.
Many journals allow the reviewer to make ‘confidential comments for the editor’ that are not passed on to the author. The ‘confidential comments’ box appears to encourage referees to be secretive, and the EMBO journals have therefore eliminated it. I believe that all comments relevant to a deci- sion should be communicated to authors but serious concerns, for example
ethical standards or data integrity, should be discussed directly with the editors. Finally, the editorial process is a dialogue. The telephone numbers of editors at EMBO are publicly available and the journals encourage authors to contact staff directly on any matters that they wish to discuss.
It is clear that the solutions discussed here do not cure all ills. However, we should acknowledge that the peer review process works, that it remains the only proven mechanism of quality control in the sciences, and that it depends on the remarkable goodwill of referees. We must look for ways to improve the process but we must not forget that it also needs our undivided support.
Maria Leptin EMBO Director
REFERENCES 1. Rørth P, EMBO J, 28, 1–3 (2009). 2. Pulverer B, Nature, 469, 29–31 (2010). 3. Ploegh H, Nature, 472, 391 (2011).
Table 1 | Concerns and some solutions for scientific publishing.
Decisions are made behind a veil of anonymity
Editors hide behind referees 3
2. Quality of refereeing
E.g., superficial referee reports, bias, conflict of interest
Manuscripts pass through multiple rounds of revision. Manuscripts are passed from one journal to the next which wastes the time of researchers and referees
4. Unreasonable requests
Referees ask for too much, editors do not intervene, supplementary data proliferate
Paper is not accepted because a similar study is published elsewhere while the paper is under review
Response: Change in editorial procedure
➔ ‘Transparent peer review’: Referee reports and editorial correspondence are published together with the paper 1,2
➔ Discontinued ‘confidential comments for the editor’
➔ Appeals process, including expert arbitration where necessary
➔ Referees cross-comment on each others’ remarks
➔ Transparent peer review provides an incentive for referees to write constructive reports
➔ Manuscript transfer to other journals with review files and referee identities
➔ Eliminated need for unnecessary re-formatting at submission
➔ Clear instructions to referees to review the manuscript under scrutiny and not a new phase of the project
➔ Currently testing structured referee reports
➔ Manuscripts are not considered scooped between day of submission to the times of revision and final decision: publication of similar data by competitors during this period does not prohibit acceptance of manuscripts
2 EMBOencounters | Winter 2011|2012 | firstname.lastname@example.org