Re: Most Science Studies Appear to Be Tainted By Sloppy Analysis

[Guest guest]

I read this yesterday and my sense is that the entire article is not quite as contrary as this excerpt, but this is a good reminder about the complexity of this subject, and uncertainty of studies that haven't been vetted by replication. One point worth repeating is that most research is funded often with some goal or intent. While that alone does not negate the research, some bias creep is always a real risk.  One important lesson taught by Walford in his book is the value of critical thinking. Caveat Lector.JROn Sep 15, 2007, at 4:34 AM, bill4cr wrote:Wall Street Journal article:http://online.wsj.com/public/article/SB118972683557627104.html?mod=blog"Take the discovery that the risk of disease may vary between men andwomen, depending on their genes. Studies have prominently reportedsuch sex differences for hypertension, schizophrenia and multiplesclerosis, as well as lung cancer and heart attacks. In researchpublished last month in the Journal of the American MedicalAssociation, Dr. Ioannidis and his colleagues analyzed 432 publishedresearch claims concerning gender and genes. Upon closer scrutiny, almost none of them held up. Only one wasreplicated.Statistically speaking, science suffers from an excess ofsignificance. Overeager researchers often tinker too much with thestatistical variables of their analysis to coax any meaningful insightfrom their data sets. "People are messing around with the data to findanything that seems significant, to show they have found somethingthat is new and unusual," Dr. Ioannidis said."

[Guest guest]

Yes, and always try to read the ENTIRE study not just the abstract.

Look at the data for yourself. Don't rely exclusively on the author's

conclusions.

>

> > Wall Street Journal article:

> >

> > http://online.wsj.com/public/article/SB118972683557627104.html?

> > mod=blog

> >

> > " Take the discovery that the risk of disease may vary between men and

> > women, depending on their genes. Studies have prominently reported

> > such sex differences for hypertension, schizophrenia and multiple

> > sclerosis, as well as lung cancer and heart attacks. In research

> > published last month in the Journal of the American Medical

> > Association, Dr. Ioannidis and his colleagues analyzed 432 published

> > research claims concerning gender and genes.

> >

> > Upon closer scrutiny, almost none of them held up. Only one was

> > replicated.

> >

> > Statistically speaking, science suffers from an excess of

> > significance. Overeager researchers often tinker too much with the

> > statistical variables of their analysis to coax any meaningful insight

> > from their data sets. " People are messing around with the data to find

> > anything that seems significant, to show they have found something

> > that is new and unusual, " Dr. Ioannidis said. "

> >

> >

> >

>