Posted on 05/23/2007 12:43:06 PM PDT by LibWhacker
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 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. Advertisement
Two papers published this spring in the open-access journal PLoS Medicine by Benjamin Djulbegovic from the University of South Florida and Ramal Moonesinghe from the CDC have delved into the issues raised by Ioannidis and suggested possible ways to mitigate this apparent failure of scientific enterprise. One of the suggestions is to ensure that experimental results are independently replicable. "More often than not, genuine replication is not done, and what we end up with in the literature is corroboration or indirect supporting evidence," says Moonesinghe.
The culprits appear to be the proverbial suspects: lies, damn lies, and statistics. Jonathan Sterne and George Smith, a statistician and an epidemiologist from the university of Bristol in the UK, point out in a study in British Medical Journal that "the widespread misunderstanding of statistical significance is a fundamental problem" in medical research. What's more, the scientist's bias may distort statistics. Pressure to publish can lead to "selective reporting;" the implication is that attention-seeking scientists are exaggerating their results far more often than the occasional, spectacular science fraud would suggest.
Cash-for-science practices between the nutrition and drug companies and the academics that conduct their research may also be playing a role. A survey of published results on beverages earlier this year found that research sponsored by industry is much more likely to report favorable findings than papers with other sources of funding. Although not a direct indication of bias, findings like these feed suspicion that the cherry-picking of data, hindrance of negative results, or adjustment of research is surreptitiously corrupting accuracy. In his essay, Ioannidis wrote, "The greater the financial and other interest and prejudices in a scientific field, the less likely the research findings are to be true."
Academic bias could also be to blame. As Ioannidis puts it, "Prestigious investigators may suppress via the peer-review process the appearance and dissemination of findings that refute their findings, thus condemning their field to perpetuate false dogma." Advocates of prevailing paradigms have been observed to band together in opposition against alternative ideas with perhaps more antagonism than one might expect from objective scientific debate. And the opposition isn't limited to publication of new science; jobs and grants are also more easily allocated to those affiliated with the scientific party in power.
Ioannidis is adamant that the problem is widespread. "I have heard from scientists from many different fields who think that the problems are the same in their fields as well," he says. "This is a potentially severe crisis, unless we realize the issue and try to address it."
With the debate over the causes and solutions of high rates of falsifiable research findings ongoing, how the problem is seen in the eyes of a skeptical public may be another issue altogether. Virginia Barbour, managing editor of PLoS Medicine, puts it simply: "In terms of perception, the point is that science doesn't emerge from single new findings that become 'breakthrough' stories in the media, but rather from developments that mature over months or years, with different sources of experimental validation."
The ones about global warming being caused by human activity are.
Agree.
About two years ago some kids at MIT submitted an item to this board. They fabricated the whole thing. They even made up words. It was a prank to expose this board as a sham.
Interesting read.
This can’t possibly be. Finacial gain over scientific integrity. I'm totally blown away!
” Cash-for-science practices between the nutrition and drug companies and the academics that conduct their research may also be playing a role. A survey of published results on beverages earlier this year found that research sponsored by industry is much more likely to report favorable findings than papers with other sources of funding. Although not a direct indication of bias, findings like these feed suspicion that the cherry-picking of data, hindrance of negative results, or adjustment of research is surreptitiously corrupting accuracy. In his essay, Ioannidis wrote, “The greater the financial and other interest and prejudices in a scientific field, the less likely the research findings are to be true.” “
Note that the author completely ignores the most egregious sinner here — the “environmentalists” and the “nanny-staters” (ie ‘second-hand smoke’) unabashedly use junk-science to ‘prove’ their positions...
It’s all just the nasty ol’ corporations, don’tcha know......
In 1989, a CDC ‘White Paper’ stated that by the year 2000, one in six male college students would be infected by the HIV virus.
Obviously, it never happened. Just as obviously, I don’t believe anything that comes from the CDC, which isn’t science based, its political correct based.
marked
bump
BTW, this can happen without any concious intent on the part of the researchers.
It has been conclusively proven over and over, in (truly scientific) double-blind studies, that scientists tend to find what they expect to find, even when they have no discernable financial or career incentive to do so. Such incentives obviously make it even easier to overlook countervailing evidence.
ping
ping
Before my mother died she took my wife aside and told her she would get the silver upon her death with the admonishment to pass it on.
When our daughter married, my wife presented her with the blue velvet lined solid wood box.
I don’t remember ever using that silver set but I did polish it when I was ten.
One day when I was about 13 my grandmother confided in me that my mom’s silver was merely plated.
I’ve never checked the authenticity of that box of treasure.
bookmark
Pure bunk and a federal appeals judge said so.
But the state of California, for example, still pushes the false conclusion of invalid "study" in an aggressive anti-smoking all-media campaign -- also at the taxpayer's expense.
In this state, we get to pay to be lied to, to have our intelligence insulted, to lose our freedoms.
I don’t know what percent, but many surely are. My professors were always in a huff to publish something, anything, because that was required as part of the job. On professor admitted that publication, not content, was important. Some of their writings sounded contrived.
regardless of the falsifications over 90% of technical papers are useless anyway.
the implication is that attention-seeking scientists are exaggerating their results far more often than the occasional, spectacular science fraud would suggest.
***It seems like conventional wisdom had been looking at many of these scientific findings with a skeptical eye for a long time, and it turns out to be for good reason.
The rule is ‘publish or perish’ for them to keep their place in the school
So there is an 80% chance that his own research is wrong.
ping
LOL
Science now consists of teasing meaningful correlations out of mountains of random statistical noise by dodgy methods of multivariate analysis.
This is true in medicine, climatology, and many other fields.
-ccm
It'd be like me building a 10-story building, then when its safety was called into question, going out and hiring an engineer to justify what I had done. There's no way in God's green Earth he'd be able to certify that building as safe.
Until a better picker comes along. In medicine, Polymerase Chain Reaction (PCR) has opened up a whole new forest of low-hanging fruit.
You remind me of the Director of the Patent Office in 1893 who recomended closing the office because everything worth inventing had already been invented.
Not only are you wrong here, but you're spectacularly wrong. In field after field (astronomy, genetics, information theory, neuroscience, geology, quantum mechanics, and I could literally go on for hours), more has been discovered in the last fifty years (and in some cases, the last twenty-five) than was discovered in all of human history to that date.
Not only is science progressing, but science is progressing faster than ever before.
Maybeso, ccmay. But all we need is taller pickers.
Most scientists today are runts standing on the shoulders of giants. Dwarfism is an innate aspect of a government controlled education..
It’s amazing that kind of thing is going on. It’s happening today with global warming; some of the world’s top statisticians have explained how envirowhackos have totally misrepresented and screwed up the analysis. Yet they persist.
How can they be bunk if they are “peer reviewed?”
Some of the articles say little of importance with a series names tacked onto them, each one getting credit as co-author and then the co-authors return the favor and tack the first name on their articles.
My secret FReeper name is Sir Chasm.
Don’t tell nobody ;)
You seem to be of the apparently mistaken belief that "Peer review" means someone checking the work. McIntyre of Mann-debunking fame has presented several examples of editors of prominent peer-reviewed journals claiming that in their decade or longer terms as editor that no one has ever actually previously asked for any of the data.
LOL, okay. :-)
BTW, 42 isn’t intended to be disparaging to you. I wasn’t aware of how little there often was involved in “peer review” for some fields.
Its a well known fact that 67.3% of all statistics are made up.
It’d be nice if they could get it, but no researcher I ever heard of would part with his data. They’re his proprietary intellectual property and no one can demand that he hand them over.
If the chief investigator can answer all those questions to the peer reviewers' satisfaction, the journal would go ahead with publication. I would often scrounge around in the basement for a bunch of key phrases that would let the reviewers know they were dealing with a real statistician, and in every case they would then back off, even if what I was saying was... let's say, perhaps, statistically questionable. In other words, a real statistician would've called me on it, but I never was.
Even if it is taxpayer funded?
In the studies in question when push came to shove, it was discovered that in addition to statistically unproven methods, the methods they used were not accurately described in the paper, and the data series used were not accurately cited. The NAS review also found that a lot of the data was not archived (so it no longer exists).
Associated with this were other studies which the authors couldn't say where (geographically) they got the data from.
How can you possibly have replicable studies without knowing what the heck people did, and to what they did it?
I never told an outright lie or falsehood in that work, but I was definitely a professional spinmeister for people who were under tremendous, I think even unreasonable, pressure to publish or perish. That's how I got published as co-author. Not proud of it, either. The system ain't what I thought it was going in.
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.