> John Ioannidis is the author of a paper titled
> "Why Most Published Research Findings Are False"
The astute reader might wonder why Ioannidis' paper itself should be accepted as "true" if, as it claims, MOST published findings are false. :)
That astute reader would be absolutely correct - the paper was critiqued in "comments" right off the bat, and the firm proof that Ioannidis was utterly wrong was published in 2013:
"An estimate of the science-wise false discovery rate and application to the top medical literature"
Jager and Leek
Oxford Biostatistics, Volume 15, Issue 1, Pp. 1-12. (2013)
http://biostatistics.oxfordjournals.org/content/15/1/1.full.pdf+html
http://tinyurl.com/heafyhr
Fair warning, the subject matter is statistics and math, slow slogging.
TL;DR - Only about 14% of the papers Ioannidis "reviewed" were not reproducible, and this is not at all surprising given how sloppy some people's recordkeeping is.
> at first it received little attention outside the field of medicine
No surprise, because the paper was solely ABOUT work in the fields of medicine and behavioral psychology, where the results are often far more difficult to quantify than in other branches of science, where quantification can be done with the sort of precision that allows sets of data to be combined, and fall into normal distributions (or not, and reveal themselves to be unexpectedly flawed, or surprisingly revealing of something other than what was thought to be measured).
That paper had perhaps the most unfortunately misleading title ever to gain the attention of the lay public. If not for the title, it would not be the annoyance it is.
It has been cited most often cited as if it were evidence that “science is wrong”. This is gleefully seized upon by alternative medicine promoters, for obvious reasons.
The original Ioannidis paper, and the immediate critique of it may be of interest to some. Here's the links:
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1182327/
http://tinyurl.com/347wmow
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1855693/
http://tinyurl.com/4nr94z7
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1896210/
http://tinyurl.com/j6h5xz4
***********************************************
The BEE-L mailing list is powered by L-Soft's renowned
LISTSERV(R) list management software. For more information, go to:
http://www.lsoft.com/LISTSERV-powered.html
|
