Maria Kowalczuk,1 Michelle Samarasinghe2
Anecdotal evidence from editors suggests it is more difficult to recruit reviewers for journals that use fully open peer review compared with single- or double-blind peer review. The aim of this study was to determine whether there is a difference in the proportion of reviewers who agree to undertake peer review of manuscripts for journals that use different peer review models in different subject areas.
Retrospective analysis of BioMed Central and SpringerOpen journals that use the 3 different peer review models in biomedicine, chemistry, clinical medicine, computer science, earth science, engineering, health sciences, life sciences, mathematics, and physics. We calculated the proportion of invited reviewers accepting an invitation to review or re-review a manuscript per journal per month between June 1, 2001, and July 1, 2015, for single-blind and open peer review journals and between February 1, 2011, and July 1, 2015, for double-blind peer review journals.
The proportion of accepted invitations was 49% overall (N = 498 journals), 60% for the 40 journals implementing double-blind peer review, 53% for the 388 single-blind peer review journals, and 42% for the 70 open peer review journals (Table). Within the 4 subject areas for which data for all 3 peer review models were available—biomedicine, clinical medicine, health sciences, and life sciences journals—the proportion of reviewers accepting invitations to review was lowest for open peer review and highest for the double-blind model. A pairwise proportion test showed statistically significant differences between the proportions of reviewers who agreed to open peer review for the clinical medicine (45%), biomedicine (41%), health sciences (38%), and life sciences (31%) journals. An analysis ofnsingle-blind peer review data available for each field showed the proportion for mathematics was higher than for the other subject areas that did not exhibit differences.
A smaller proportion of invited reviewers agreed to peer review for journals operating under open and single-blind peer review models compared with journals that use double-blind peer review. As a result, for journals with open and single-blind peer review models, a higher number of reviewers need to be invited. However, these journals have operated under their respective models for many years, so the discrepancy between these proportions does not seem to be detrimental to the success of a journal.
1BioMed Central, London, UK, firstname.lastname@example.org; 2Nature America, New York, NY, USA
Conflict of Interest Disclosures:
Maria Kowalczuk is employed by BioMed Central (part of Springer Nature), and Michelle Samarasinghe is employed by Nature America (part of Springer Nature).