Peer Review Week: An analysis of peer review style and quality

This week celebrates the first ever Peer Review Week; a collaborative concept from ORCID, Wiley, Sense About Science and ScienceOpen, to highlight and celebrate the invaluable role peer review plays in scientific and medical publishing.

Here at BMJ Open we are, of course, advocates of open peer review and as such are pleased to be publishing a timely research article by our friends at Biomed Central.

The paper, from Maria Kowalczuk et al., is a retrospective analysis of the quality of referee reports from author-suggested and non-author-suggested reviewers in open or single blind peer review journals.

Their objective was to elucidate whether reviews from peers suggested by authors would show bias in quality and decision recommendation compared to reviewers selected by other means. They also aimed to assess whether open review vs. single blind review had an impact on quality and recommendation. To achieve this, the study looked at 200 reviewer reports submitted in 2010-2011 to BMC Microbiology, 200 submitted to BMC Infectious Diseases, and 400 that were submitted to the Journal of Inflammation, these journals use single blind peer review, open peer review and a combination of the two, due to policy change, respectively. Comparisons were made by assessing the quality of report (using the Review Quality Instrument), by analysing the editorial recommendation made, and with author surveys. After statistical analysis of the data, they could conclude that the reports from reviewers suggested by the authors were of comparable quality but were more likely to suggest publication.

They also conclude that the open peer review reports were of a slightly higher quality than those using single blind review. These findings are in line with those from the randomised trial conducted by The BMJ, after they became one of the first journals to use open peer review in 1999, and similar to another study by the British Journal of Psychiatry.

As BMJ Open is open access with fully open peer review, we are always happy to see further research demonstrating the success of this model.  As we use a combination of both author suggested and non-author suggested reviewers (with an in-house filtering process), we couldn’t help but agree when we spoke to the paper’s authors and they said the following,  ‘It is reassuring that reviewers suggested by the authors provided reports of as good quality as reviewers found by editors using other means. Author-suggested reviewers tended to recommend acceptance of the manuscript more often than other reviewers, which highlights the important role of the editor in making the final decision on the manuscript’ .

The full text of the paper, Retrospective analysis of the quality of reports by author-suggested and non-author-suggested reviewers in journals operating on open or single-blind peer review models– Kowalczuk et al., can be found here: http://bmjopen.bmj.com/content/5/9/e008707.full