Is Peer Review Broken? |
This past week I found myself asking this question quite a few times. What is going on with the peer review process? Is anyone actually reviewing the papers getting into journals anymore?
This is due to some recent experiences I’ve had with papers published in both the larger highly reputable journals and smaller niche journals that has left me wondering. I review papers for Current Issues in Molecular Biology and I have run the peer review process for them so I understand what is expected as an editor or a reviewer. The journal is counting on me to ensure that a high quality paper is approved and nothing less. As an editor running the peer review process and selecting reviewers of the paper, it is my responsibility to make sure I choose people who will be fair, unbiased, and have the time to actually read the paper.
I am incredibly busy, but this is a responsibility I take seriously because not only does a poorly written paper reflect badly on the journal and on the authors, it reflects on the editor.
How did this get published?
Recently I came across a paper that was published in a popular journal for microbiologists despite the fact that it had no control experiments, the conclusions didn’t match the data (see the previous article on the importance of reading a paper thoroughly), and the entire study missed the opportunity to make any scientific contribution to the field by focusing on the wrong points. How did this make it through peer-review, I asked myself?
Letters can be written to the authors and the editor, however, it doesn’t correct the problem. How does a scientific study gone awry get published anyway? How does every reviewer miss such obvious flaws in a paper?
More examples…
Recently I read a paper from a smaller niche journal for microbiologists that a scientist/friend asked me to comment on. We both came to the same conclusions: the paper was garbage. The authors used the incorrect name for their organism, making one up that doesn’t exist and mislabeled a figure so that the results section did not match the figure legend, making interpretation confusing. Furthermore, they left a key piece of information out of their methods (how much bacteria they used to inoculate the soil), so the data obtained was uninterpretable, and to cap it all they misinterpreted their very own data in the conclusions.
How does this get published? Was it reviewed at all? A paper of this low quality brings down the journal and its editors, and affects the whole field.
It’s not just microbiology journals!
The popular science and general methods type journals are not immune to this problem. Several months ago I read an article in a very popular magazine that used qPCR for their study and attempted to use the MIQE guidelines to validate their work. It was great to see people attempting to use the guidelines.
However, upon opening the attached spreadsheet in the Supplemental Data, I was shocked to see that most of it was empty. The authors actually wrote “NA” in almost every field of the MIQE checklist and didn’t provide any information on yields, purities, etc. How many people do you think took the time to go to the Supplemental Data and actually see the checklist? I’ll hazard a guess that it is very few since no one else noticed the alarming lack of information contained in their “supplemental data”.
I know we are all busy and I know that we want to help our colleagues to get published, but there needs to be more measures in place that ensure that the standards are high.
Here’s what I suggest
My suggestion is that if the journal allows the authors to choose reviewers for the paper, the editor running the peer-review process should use only one name from their list and the others are not. Chances are that the reviewers suggested by the authors are going to be very lenient (after all, they are going to ask for the return favor when they publish their next paper) and the one with nothing to gain by approving the paper will give a more accurate assessment.
And editors need to stay objective to the body of work regardless if the authors are friends or collaborators. It is just like referring someone for a job- it reflects on us as editors if we recommend a piece of work that really is not up to par. No journal wants to publish papers that need erratum published later, or worse, no one takes the time to let the journal or authors know that an article needs erratum and instead the community decides that that the journal is too low tier to even publish there or that the people who publish there are those who get rejected everywhere else. And of course, no one wants to cite a paper with obvious flaws so the citation index continues to go down.
Ok, so maybe it’s not ALL broken…
I know that the peer-review process does work successfully and that there are many journals making sure that their review process is stringent. It’s probably no coincidence that these journals usually have the highest citation indexes as well.
My advice to all the readers out there preparing papers for submission to journals is to step back and look at your data as objectively as you can. I know you have a theory or model you are trying to prove and want the data to support you. So knowing that, step back and read it again from an outside point of view. What other interpretations could there be? What else could be going on? Let the data tell you what is going on. Be open to other interpretations of the data and write it up objectively.
Not only will you have a stronger paper that more people will cite, but, it will mean that you were the first to think of it and when other people follow your lead and build on your work, they will have to say you suggested it first.
And don’t be disappointed if you need to make revisions to the paper when it comes back for review. Thank the reviewers for taking the time to read your work carefully and know that you will be able to re-submit a much stronger and more citable paper in the next round.
To leave a reply you now have to register on Bitesize Bio. We're sorry for the inconvenience, but it's free, only takes a few seconds, and it will enable you to view our seminars for free, ask questions from the professional community, and take part in the lively community of Bitesize Bio
Alejandro
Just as Keith Robison once said: “If your results sound amazing, take a deep breath & think of several tests that could debunk them. Then do those ten tests. If they survive, go to bed & think of another batch of tests”
http://amontenegro.blogspot.com/2010/02/quotes-from-science-blogosphere.html
John Mackay
Hmm. . interesting strategy. Be sure to mention in introduction that qPCR work was done taking into account the MIQUE guidelines – and then blank the checklist
anon
There seem to be many issues surrounding publications. An increasing number of reviews in our weekly lab meetings from a major publisher beginning with N are becomming coffee table magazine quality and are frequently trashed by the presenting post doc. I am only very new to the academic scientific world however I have also heard many colleagues commenting on reviewers delaying publications as competing labs rush to finish similar papers.
Does the post “credit crunch” era of dwindling grant funds pose the threat of major journals becoming members only clubs promoting each others work and locking out the smaller or newer labs? Who watches the watcher?
Jode
I’ve often thought that one way to tighten up the system would be to publish the names of the reviewers* with the paper. You want to let a buddy sneak in a stinker? Well now your reputation is on the line too. I haven’t figured out how to deal with the dramatic drop-off in the number of willing reviewers once my plan is implemented, though…
*If they recommended publication – sometimes the editor overrides us.
Ranga
And did any of you guys know an international perspective here? The same reviewers would sit all night with coffee in their hands to trash the manuscript if the paper was from another country (exclude western Europe). I’ve seen it personally, with excellent work being rejected and mauled just because they don’t want science from other countries being recognized. In one example, a paper with almost 15 figures (must have taken like 5 years) was published in a popular journal (starts with N) with a paper with just 2 figures (which had no controls, quick and dirty conclusion; must have taken like 6-8 months), clearly showing the partiality that exists in the peer review process. And the scientific world judges you from those publications.
Maria
@Ranga
The thing is you don’t know how good the 15-figure paper was when it was first submitted. It might have been the reviewers saving the authors from themselves (and the journal from a retraction) by ensuring they had enough evidence to match their assertions. For the 2-figure paper, how much did they have in supplementary info?
Richard
It is not just garbage papers. There are many garbage journals. I stopped reading papers from some journals this year because they almost always publish low quality stuff. I think that major journals get peer-review correctly. However, they often reject good papers if reviewers did not bother read the paper and just reject it based on abstract. I recently got one of low quality bad reviews. The reviewer complained that I did not reference the paper that actually was between my references! And he/she continued, if I recall correctly this results are known…. WTF?
Ranga
@Maria.
You are right. It might have been that the reviewers made the authors do years of experiments to prove their point. A very good thing for science indeed. But they don’t do it when they know the authors personally. That’s where its bad. And with regard to the 2 figure paper, even though my knowledge might be limited to effectively critique a paper, I believe my professors (most of them) will not be wrong (these papers were discussed in a departmental journal club and the 2 figure paper was thrashed and trashed)