Publish or perish?

Should scientists be judged on where they published rather than what they publish? No, says Stephen Curry...

Stephen - Caffeinated and back in my office, I gossip for 10 minutes with a colleague before making a start on reading a manuscript that I've been asked to review by an academic journal.

This is the peer review process and it's supposed to provide quality control. My eye stumbles along the first few lines of stilted prose in the paper's introduction and my heart sinks. This is going to be a long haul.

I steel myself by remembering why I do this. I'm not paid or credited for the hours it takes to review the 20 or 30 papers and grant applications that land on my desk each year. It's a quid pro quo. After all, I expect others to review my papers, though I hope I don't give them as much grief as these guys seem to have in store for me. My editorial hand stops on every other line of the manuscript to correct or comment in exasperation.

The peer review process isn't perfect but it works well enough most of the time, providing a useful opportunity for authors to clarify or improve their papers before publication. But there are growing concerns about the inability of peer review to catch all errors. Some are just honest mistakes, but in other cases, the reviewers are fooled by outright fraud! Infamously, some authors have even created fake email addresses so that they would be sent their own papers for review. And the bad behaviour is becoming more common: between 1975 and 2012, there's been a 10 fold increase in the numbers of scientific articles retracted after publication. In over 70% of cases the retractions were due to scientific misconduct of one type or another.

It's unfathomable to me that researchers would write up results they know not to be true. But scientists are human too and this is the dark side of the ubiquitous pressures to publish. It is a complex problem, but at least it seems to be attracting more attention. As a community we need to recalibrate our incentives to reward good, reliable science and not simply publication in top-tier journals.

But back to the manuscript at hand which, after a few tortuous hours, I have now finished. My initial assessment? Bloody awful. But I know I need to wait a day or two before writing my report: time for my inner nit-picker to calm down.