Saturday, April 27, 2013

False Positivism .. Gone Rogue: How one man got away with mass fraud by saying ‘trust me, it’s science’

Diederik Stapel
One of the great unanswered questions about the Stapel affair, however, is how he got away with such blatant number-fudging, especially in a discipline that claims to be chock full of intellectual safe-guards, from peer review to replication by competitive colleagues. How can proper science go so wrong?

The answer, according to a growing number of statistical skeptics, is that without release of raw data and methodology, this kind of research amounts to little more than “‘trust me’ science,” in which intentional fraud and unintentional bias remain hidden behind the numbers. Only the illusion of significance remains.

GETTY IMAGES/THINKSTOCK
GETTY IMAGES/THINKSTOCK
S. Stanley Young and Alan Karr of the US National Institute of Statistical sciences, for example, point to several shocking published claims that were not borne out by the data on which they were based, including coffee as a cause of pancreatic cancer, Type A personality causing heart attacks, and eating breakfast cereal increasing the odds that a woman will give birth to a boy.

“The more startling the claim, the better,” they wrote in a recent issue of the journal Significance. “These results are published in peer-reviewed journals, and frequently make news headlines as well. They seem solid. They are based on observation, on scientific method, and on statistics. But something is going wrong. There is now enough evidence to say what many have long thought: that any claim coming from an observational study is most likely to be wrong – wrong in the sense that it will not replicate if tested rigorously.”

Victor Ivrii, a University of Toronto math professor, described the problem similarly on his blog: “While Theoretical Statistics is (mainly) a decent albeit rather boring mathematical discipline (Probability Theory is much more exciting), so called Applied Statistics is in its big part a whore. Finding dependence (true or false) opens exciting financing opportunities and since the true dependence is a rare commodity many “scientists” investigate the false ones.”

http://news.nationalpost.com/2011/12/30/how-one-man-got-away-with-mass-fraud-by-saying-trust-me-its-science/ 

No comments:

Post a Comment