Science: A year-end wad of fraud, falsified data, and other award-winning tenure strategies ...
I just got my e-mail notice of the December edition of the "Outside the Cover" of The Scientist, and it made for a curious reading experience:
Bad news at Cell
Improper citation, disregard for antecedent research, and shoddy experimentation - those are just a few of the allegations levied against a recent Cell paper. Is this paper emblematic of a larger problem in scientific publishing?(Um, it's a big topic, but let's start with peer review and work backward, okay? The skinny: Do third rate minds get together to suppress first rate minds who challenge their "findings"?)
And at JEB (Journal of Experimental Biology):
In the first retraction in its 85-year history, the journal calls the authors' reuse of images a case of outright fraud, not a careless error as claimedNature to retract plant study
A highly cited paper that identified a long-sought receptor critical for mediating plant response to stress is being retracted after researchers were unable to reproduce the results("But we ran out of Fairy Dust, you see ... " (?))
A Texas stem cell researcher falsified data by mucking around with her results in Photoshop (Another shoo-in for the Muddy Waters award?)
"Officials have halted enrollment in more than 600 human research studies due to shoddy paperwork" (Too bad Miss Grimstone, Secretary (and don't you ever forget it!), retired twenty-nine years ago today ... )
Journal of Evolutionary Psychology: Several prominent evolutionary psychologists have been accused of fraud, in appropriating hundreds of Just-So Stories from other evolutionary psychologists. These academics have hired a lawyer who is defending them on the grounds that they cannot be guilty of fraud because their work is 100% speculative fiction. They can, he admits, be sued for plagiarism. But plagiarism must be proved, and he says there is not enough proof. The urban legends in question have been floating around in the pop science culture for decades, under various guises. His clients also can't be nailed for libel, he advises, because one cannot libel the dead - let alone persons who probably never existed.
(Note: I linked these blog posts for your convenience, but you must register to read them.)
Okay, sure, I made that last one up. The others you can read about elsewhere. I hope this is merely a year end review of regrettable moments, rather than a Krakatoa of corruption just beginning to blow off ...
But I was hardly reassured to read this, sent recently by a friend:
In a 2005 article in the Journal of the American Medical Association, epidemiologist John Ioannidis showed that among the 45 most highly cited clinical research findings of the past 15 years, 99 percent of molecular research had subsequently been refuted. Epidemiology findings had been contradicted in four-fifths of the cases he looked at, and the usually robust outcomes of clinical trials had a refutation rate of one in four.The wonderful thing about science is that it is self-correcting? Oh, come on!
The revelations struck a chord with the scientific community at large: A recent essay by Ioannidis simply entitled "Why most published research findings are false" has been downloaded more than 100,000 times; the Boston Globe called it "an instant cult classic." Now in a Möbius-strip-like twist, there is a growing body of research that is investigating, analyzing, and suggesting causes and solutions for faulty research.
- João Medeiros, "Dirty Little Secret Are most published research findings actually false? The case for reform." (May 21, 2007)
Speaking as a longtime book editor - and mindful of the history of bridge engineering - I would say that self-correction is a sport best enjoyed in the privacy of one's own office before the big launch.
And you do not need any science training to know that fact.
My own sense is that too many people today are invested in proving stuff they are sure is true, and not enough in finding out what is really going on.
Find out why there is an intelligent design controversy: