Ever wondered you are given weeks for reviewing but it takes rather months to get the decision for your own submissions and whether there’s something amiss with the reviewing process? In this essay, Michael Hochberg and others, all journal editors, point out two habits of authors that lead to overexploitation of the resource ‘reviewer’.
Firstly, authors tend to initially submit their manuscripts to high-ranking journals for that they aren’t really suited. Of course, it’s tempting to send a manuscript to Nature, Science or PNAS just on the off-chance they like it but this means that a manuscript is reviewed more often than necessary. This problem isn’t eased by the fact that not all scientists are equally willing to review manuscripts (some of them rather prefer to spend writing even more). In this context a proposal by Hauser and Fehr to punish authors for late and sloppy reviews or refusing to review at all is interesting. Basically, if a reviewer turns in his/her review, say, one week late than the decision on his/her submission will be delayed by twice the time, i.e. two weeks, on purpose.
The secondly issue identified by Hochberg et al. Is that authors do not always take the previous reviewers’ into account and just submit more or less the same version to another journal. Here, they propose a possible solution: asking the authors to sign that they “… confirm that should our study have been previously submitted to another journal, [they] have taken all reviewers comments into account in revising our manuscript for submission to…,” (Page 3, 3rd paragraph). I wonder whether this would help much because it’s hardly possible to check on this and hence the temptation to cheat is big. If ignoring previous reviewers’ comments really is a problem then a more serious approach is necessary and it could look as follows: after final rejection the authors are given the opportunity to reply to the reviewers’ (and editor’s) comments. After submission to another journal all this will be forwarded to it and be made accessible to the editor (of course, this would require that journal exchange information about submissions but this should be a minor problem in the ‘age of the internet’). If the editor wants, he can check whether and how the manuscript has changed and may base his decision to send it out for review on this information. Because this information is not made accessible to the new set of reviewers they won’t be biased in their judgements. Of course, the editor’s decision can be biased by the earlier rejection but if authors clearly ignored important, earlier comments he can save ‘his/her’ reviewers time by rejecting it straight away. Many journals nowadays do this anyway and these rejections (without review) are not always well justified…
ME Hochberg, JM Chase, NJ Gotelli, A Hastings & S Naeem (2009) The tragedy of the reviewer commons. Ecology Letters 12, 2-4. doi: 10.1111/j.1461-0248.2008.01276.x
M Hauser & E Fehr (2007) An incentive solution to the peer review problem. PLoS Biology 5, e107. doi:10.1371/journal.pbio.0050107
2 Comments
These issues and ideas are logistically complicated, but probably worth pondering.
Another issue here is in that many of publishers are making *awfully* lot of money out of publishing (and rejecting) manuscripts, and the reviewers are basically serving these publishers for free.
Perhaps one should discuss also whether the pro bono basis for reviewing needs to be revised. Just 1-2 decades ago the pace and amount of review work was quite different what it is today, as are the demands put on us by universities and funding bodies.
Just a quick calculation. Say that you do 5 reviews per month (quite average review load, I would guess), this sums up to 60 mss per year. If you do your work properly, you spend easily 4h per ms. That is 240 h or 30 8-hour working days (=10 full days) of the year just for reviewing. What would your employer say if you would tell that over one month of your working time is spend on working for a international publishing house for free?
Maybe my employer would say that I should only review for non-commercial journals, like the ones run by/for societies, e.g. J. Evol. Biol.?
And here another suggestion how to reduce overburdening of reviewers:
If everybody reviews at least twice (since there’re two reviewers per ms) as many manuscripts as he/she is 1st author on (within a given time period, say a year or two) the burden of reviewing is equally distributed among us.
Implementing this would require some central data base, into which all refereeing activity is entered. When a manuscript is submitted to any journal, this journal checks with the data base whether the 1st author has reviewed enough mss and if not sends the ms back or keeps it on hold until he/she has done his/her ‘duty’.