Peer reviewing is important, so give yourself time to take a good look at each paper, don't be rude about a fellow academic's work and ensure that any criticism is constructive, advises Harriet Swain
What a load of tripe. Not that you'll tell the author, of course. You'll simply mark the script "Not suitable for publication", then have a good bitch about it with the journal editor.
Don't be surprised if you're not asked to peer review again. Fiona Godlee, editor of the British Medical Journal and chairwoman of the Committee on Publication Ethics, says she looks for a peer reviewer who examines the strengths and weaknesses of a paper in a constructive way and suggests how it could be improved - "not just a demolition job".
And you should not keep the full extent of your loathing to private correspondence with editors. "I would never say anything behind someone's back that I wouldn't say to his or her face," says Peter Cotgreave, director of the Campaign for Science and Engineering, who argues that reviewers should always sign their reviews. "If you are trying to conceal your identity, examine your motives," he cautions.
This doesn't mean you should be mealy-mouthed. "You have to be honest," he says. "Sometimes you get reviews back and you cannot work out whether the referee liked it or not. Sometimes they will reject the paper and you will think, after reading the review, that it had gone down well."
Cotgreave warns that it is easy to write a review that sounds negative, even if you liked the paper, because the tendency is to list all the things that would make it better rather than say why it is good. "Authors don't realise that the reason you've enumerated these things is because there's something there worth taking on," he says.
Susan Hezlet, publisher at the London Mathematical Society, says that some of the worst kinds of reviews are those that say, "This is correct, signed Fred", because they are no help when editors are trying to put a paper in context. "If we are going to reject the vast majority of papers that come to us, we need damn good reasons for why we are going to accept others," she says.
She concedes that if a minimal review is signed "Fred, the Field Medallist", it is likely to be taken more seriously than most, but even then more detail may be required. Reviewers who follow the society's guidance notes to referees are always appreciated, she says, while those who are argumentative, unpunctual or consistently seem to have poor judgment - however good they are at their subject - may not be asked to review again.
Peter Golding, editor of the European Journal of Communication , and chairman of the 2008 research assessment exercise panel for media studies, says editors tend to look for two principle qualities in their reviewers: expertise and authority on the one hand, and common sense and efficiency on the other.
He says that they must understand what journal editors and readers expect and that any comments must be focused and helpful. If they criticise a paper for missing references, they must specify those references. But the review shouldn't make unreasonable demands, such as suggesting a fundamental rethink of the whole research process, or expecting references to works the author couldn't possibly have known about - such as the referee's own unpublished writings.
Hezlet says the first question she usually wants answered by reviewers is whether the paper is tackling a pertinent subject. "Sometimes it may be correct, but is it interesting?" she says. She then expects thorough checking of every calculation in the paper.
Godlee wants peer reviewers to check that papers have passed ethical reviews and then to raise any further ethical concerns they may have. They also need to check notes and tables and consider readability, though not in too much detail. "We aren't looking for them to say a full stop is missing on page x," she says. Any suggestions about whether the paper may need a commentary or needs to be put into context are gratefully received.
She says you should decline the opportunity to peer review if you haven't the time, rather than accepting a manuscript and sitting on it. So much the better if, having declined, you can suggest someone else.
However, Irene Hames, managing editor of The Plant Journal , whose book Peer Review and Manuscript Management in Scientific Journals is due out next autumn, says it is worth maintaining good relations with a journal and saying no only when you have to because you may want your own work published at some stage.
Godlee says it is helpful if reviewers point out where they lack expertise in a particular aspect of a paper rather than misleading editors into thinking everything has been properly checked. If you do feel the need for someone else's opinion, you should always check with the editors first.
Papers are sent out in confidence and it is important to get a proper trail or audit of who has seen them, she says.
Hames adds that you should not reveal to a third party the fact that you have received a particular manuscript for review because it can upset the delicate balance of academic competition. "You sometimes know that the reason someone is submitting is because they have heard someone else has submitted," she says.
And reviewers should not use any work they have been sent for review for their own benefit. First, it would be unethical. Second, until it has been through the full review process, there is a chance that it could be wrong.
Cotgreave says you should still see reviewing as beneficial to your work and try to learn something from the papers you see. And, he adds, you should always read the whole paper "even if it's obvious after the first two lines that you're going to pan it. Do it properly."
Further Information: How to Survive Peer Review by Fiona Godlee and Tom Jefferson, Elizabeth Wager and Liz Wager. BMJ Books, 2002 Association of Learned and Professional Society Publishers: www.alpsp.org
Be an expert
Think about getting peer review training
Be honest but be constructive
Deliver on time