According to a headline in International Business Times, “Fraud and errors in scientific studies skyrocket”. That’s an attention-grabbing verb, but it isn’t borne out by the facts. The headline refers to an article in the Wall Street Journal entitled “Mistakes in scientific studies surge”*.
The WSJ based its article on a report that it commissioned from Thomson Reuters, which showed “a steep rise” in retraction notices, from 22 in 2001 to 339 in 2010. One does wonder how a steep rise turned into a surge, which turned into a skyrocket, but that could be put down to journalistic hyperbole.
What about the substance of the complaint? Well, 339 is clearly a much bigger number than 22 – an increase of 1540%, in fact, which Thomson Reuters compare with a 44% increase in published articles over the same period. But in the context of how many articles are covered by ISI, the percentage of articles being retracted has increased from something like 0.0002% to something like 0.003%, which is hardly evidence of systemic failure.
Clearly, such figures may indicate not so much an increase in transgression but an increase in detection, in the establishment of codes of practice and ethical guidelines, in the use of plagiarism-detecting software, such as CrossCheck, and in due process. As the full version of the article says, “the increase could indicate [that] journals have become better at detecting errors. They point to how software has made it easier to uncover plagiarism.”
Our own retraction figures have shown an increase in retraction requests from 24 in 2010 to 91 in 2011 (September to September), i.e. from something like 0.02% to 0.06% of our total article output, but some of this is catch-up, with transgressions from earlier in the last decade now coming to light. Most retractions are due to plagiarism and (usually inadvertent) dual publication, not misleading science or fraud and misconduct. But of course that doesn’t make for such lurid headlines ….
*full article available to subscribers to WSJ only
