Tuesday, December 23, 2008

Nuts and Bolts of Peer Review

Peer review is one of the important cornerstones of academic paleontology--this process attempts to ensure that manuscripts considered for publication contain good science from start to finish. It certainly ain't perfect, but until someone proposes a practical, consistently-applicable alternative, peer review (when properly implemented) is a pretty effective "gatekeeper" of the literature.

Of course, peer review relies on us--the scientists. If you're at all interested in academic paleontology, you have been, are, or will be involved in this process at some level. Personally, I've been very fortunate to take part in peer review at both the "giving" and "receiving" end. In this series of posts, I'll talk about my approach to reviewing manuscripts (which I've now done for a number of journals and other publishers) and receiving reviews of my own papers. The intent is not to say that my philosophies are perfect (or that I follow them perfectly myself)--they'll certainly change as I gain experience as a researcher and a reviewer. Instead, the purpose is to provide some insight into the process for those who are relatively new to the field.

Why Participate in Peer Review?
From a pragmatic perspective, you have to play the game if you want to get into the literature (in most cases). From a scientific perspective, peer review ostensibly separates the wheat from the chaff (or at least lets the chaff blow over to another journal). Like it or not, peer review is something every scientist has to face.

Personally, I have found the process to be quite rewarding. As an author, the feedback I receive from reviews is invaluable--and no matter how much it hurts sometimes, the reviewers are usually right. They catch my awkward sentences, inappropriate analogies, and convoluted analyses. Sometimes they'll even suggest new angles that greatly increase the scientific value of the paper. Often, it involves more work--but my papers are always better for it. Sure, it's never fun to have mistakes pointed out, but I'd much rather this happen at the manuscript stage than in a public rebuttal on the pages of a journal or blog.

As a reviewer, I just have fun with it! First, I won't deny that it's a bit of an ego boost to be asked to review a paper. Beyond this, it's very gratifying to be able to use some of my (obscure) research skills to contribute to the scientific process. Science involves community--and peer review is an important civic duty. Also, it's kinda fun to learn about the latest breaking research months before it appears officially in print.

Why Me (or Her, or Him) as a Reviewer?
Journals usually find their reviewers through two sources--from the authors or from an informal "reviewer pool." As an author, you always have the option to suggest a list of potential reviewers (whether in the cover letter or in a specific part of the online submission form). This also means you can suggest the exclusion of reviewers. If you think that Professor X will give an unfair review no matter how good your paper is, it's perfectly within your rights to request he or she be excluded as a reviewer (and there's no need to say why). But, recognize that it's also perfectly within the journal's rights to ignore your suggestion. However, my general sense is that journal editors will try to respect authors' wishes whenever possible. So, this is probably the most control you (as an author) have over the process (aside from writing good quality papers). It goes without saying that one shouldn't abuse this privilege--don't try and pack the review panel with "easy" reviewers (good editors will see right through this), or exclude someone just because you think they might be "tough." An editor once told me that authors don't realize that the referees suggested by the authors are often the toughest critics of manuscripts!

The second source of reviewers for journals is from a "reviewer pool"--the informal list of experts whom editors think would know something about the manuscript in question. The easiest way to get into this pool is to publish your own quality work. Once you're known as an expert in hadrosaur hindlimb biomechanics, the odds are pretty good that you'll get papers to review on hadrosaurs, hindlimbs, biomechanics, or any combination of the three.

Edited volumes present a special case. Typically, the editor(s) for such a volume will draw from the contributing authors as reviewers for each others' papers. If you're the sort of person who likes reviewing papers (or whom the editors learn does a good job at reviewing), this can be a lot of fun, and/or a lot of work.

What Happens
The first step in the process, for a reviewer, is an email from the journal editor. Often, this comes as a form letter with the paper title, list of authors, and perhaps the abstract. The reviewer is given a choice--will you accept the responsibility of reviewing, or decline?

A reviewer may choose to decline for one of several reasons. Perhaps he or she doesn't have time at the moment to review the manuscript properly. Perhaps the manuscript is so completely out of the expertise of the reviewer that it's not worth the effort. Or, maybe the requested reviewer had a non-negligible role in the research, and it thus wouldn't be appropriate to review the manuscript.

If the reviewer has seen the manuscript before, this requires some care. Perhaps it was as a reviewer for another journal from which the paper was rejected. Perhaps the reviewer looked over the paper for the authors before they submitted the final version. Either way, it's usually a good idea to let the editors know. These aren't deal-breakers (manuscripts can change substantially between drafts, after all), but editors usually appreciate knowing this sort of information.

Accepting the invitation to review a paper is a grave responsibility, with lots of unwritten obligations. As a reviewer, you promise to provide a fair, thorough, and timely report on the manuscript. The Golden Rule applies here--would I, as an author, want to receive a sloppy evaluation delivered a year after initial submission, in which the referee clearly hadn't bothered to read the paper? Hopefully the answer is obvious.

Coming soon. . .reviewing text, figures, analyses, and much more!

No comments: