Monday, June 14, 2004

Statistical Flaws Revealed In Top Journals' Papers

Two of the world's top scientific journals, Nature and the British Medical Journal, have been found guilty of routinely publishing figures that do not add up.

The evidence comes from Spanish researchers who found a surprisingly large number of statistical errors in the two journals. They warn that the same may well be true of other mainstream journals.

"It's a warning to be more careful," says Emili García-Berthou of the University of Girona, who discovered the discrepancies. "Our concern is that these kinds of errors are probably present in all numerical results and all steps of scientific research - with potentially important consequences."

As well as warning researchers and editors to be more careful with data, they also urge the publication of raw data online. "If we had that, we could check the results," García-Berthou says. "Some journals already publish supplements online, but it's rare, and I think it should become commonplace."

1. The Weekend Australian, 12 June 2004, p18
2. NewScientis - Statistical flaws revealed in top journals' papers - 28 May 04
3. http://www.nature.com/nsu/040531/040531-3.html