« Grilling the Data | Main | How not to measure temperature, part 31 »

Maybe they need a statistical analysis class

From Slashdot.org The Wall Street Journal has a sobering piece describing the research of medical scholar John Ioannidis, who showed that in many peer-reviewed research papers 'most published research findings are wrong.' The article continues: 'These flawed findings, for the most part, stem not from fraud or formal misconduct, but from more mundane misbehavior: miscalculation, poor study design or self-serving data analysis. [...] To root out mistakes, scientists rely on each other to be vigilant. Even so, findings too rarely are checked by others or independently replicated. Retractions, while more common, are still relatively infrequent. Findings that have been refuted can linger in the scientific literature for years to be cited unwittingly by other researchers, compounding the errors.'

Posted by Anthony Watts on September 19, 2007 06:59 PM |

Comments

Here's the link to the article written by John P. A. Ioannidis: "Why Most Published Research Findings Are False".

http://medicine.plosjournals.org/perlserv/?request=get-document&doi=10.1371/journal.pmed.0020124&ct=1

Interesting article.

Posted by: Paul S | September 18, 2007 03:33 PM

Findings that have been refuted can linger in the scientific literature for years to be cited unwittingly by other researchers, compounding the errors.'

Reading this reminded me of an article I read a while back in the National Post by Steve McIntyre and Ross McKitrick about their presentation last year to the NAS WRT Mann's work that lead to the infamous Hockey Stick graph. Does this sound familiar:

The NAS panel drew attention to other recent studies claiming that the 20th century was warmer than the Medieval Warm Period. We've attempted to replicate these other studies as well, only to run into one obstacle after another in identifying data and methods -- similar to the problems that led to the original congressional questions about the Mann study. In one case, the authors even refused to identify the sites from which data was collected for their study!

Despite these pointless obstacles, we know enough about the "other studies" to be confident that none of them meets the methodological standards now recommended by the panel. In fact, somewhat remarkably, two of the most recent studies even continue to use Mann's discredited principal components series. (emphasis added)


Posted by: Stan Needham | September 18, 2007 05:51 PM

Post a comment

(If you haven't left a comment here before, you may need to be approved by the site owner before your comment will appear. Until then, it won't appear on the entry. Thanks for waiting.)

Site Meter