mercredi 24 novembre 2010

Science fails to face the shortcomings of statistics

November 19, 2010

Science fails to face the shortcomings of statistics

Tom Siegfried in Science News:

ScreenHunter_01 Nov. 19 12.04

Supposedly, the proper use of statistics makes relying on scientific results a safe bet. But in practice, widespread misuse of statistical methods makes science more like a crapshoot.

It’s science’s dirtiest secret: The “scientific method” of testing hypotheses by statistical analysis stands on a flimsy foundation. Statistical tests are supposed to guide scientists in judging whether an experimental result reflects some real effect or is merely a random fluke, but the standard methods mix mutually inconsistent philosophies and offer no meaningful basis for making such decisions. Even when performed correctly, statistical tests are widely misunderstood and frequently misinterpreted. As a result, countless conclusions in the scientific literature are erroneous, and tests of medical dangers or treatments are often contradictory and confusing.

Replicating a result helps establish its validity more securely, but the common tactic of combining numerous studies into one analysis, while sound in principle, is seldom conducted properly in practice.

Experts in the math of probability and statistics are well aware of these problems and have for decades expressed concern about them in major journals. Over the years, hundreds of published papers have warned that science’s love affair with statistics has spawned countless illegitimate findings. In fact, if you believe what you read in the scientific literature, you shouldn’t believe what you read in the scientific literature.

More here.

Posted by Abbas Raza at 06:04 AM | Permalink

Comments

As Mark Twain observed, "There are lies, .....lies and statistics". Parenthetically, his autobiography is flying off the bookstore shelves http://www.nytimes.com/2010/11/20/books/20twain.html?_r=1&hp.
The cancer generals of the war on cancer are probably the biggest purveyors of lies with statistics. After all, what else could have kept these dismal failures in business for so many decades?
The late Professor Dr. Hardin Jones, Ph.D., was a professor of medical physics, physiology and expert in medical statistics at UC Berkeley. He also published with the great Linus Pauling Ph.D.(two time solo Nobelaureate, chemistry and peace) and Abram Hoffer, M.D., Ph.D. Many years ago professor Jones proved that those cancer patients who refused the orthodox treatments of surgery, chemotherapy and radiation, lived up to four times longer than those who accepted them.
The cancer generals never came up with such conclusions obviously, since if they did, they would be out of business.
But professor Jones was an independent scientist not corrupted by drug company money. Jones was right and the cancer generals were wrong, and little has changed today. Basically professor Hardin Jones proved the entire war on cancer to be a dismal, corrupt, fraud, a massive medical quackery on a massive scale. It is no wonder the low level non scientists of the medical orthodoxy sought to lie and obstruct his work, no? The U.S. FDA is a corrupt cesspool bought and paid for by drug and food companies. Americans are the biggest fools on the planet from watching too many fixed football "sports" shams on the television.
Most medical doctors in the past 50 years have not even had a physics course much higher than high school physics without calculus. The corrupt physics "profession" if you can call it that, have caved and allowed themselves to be corrupted by demands of the medical accrediting agencies to keep the standards low for doctors, lest too many flunk out. When I started out, physics and engineering majors had to take four full long (17 or18) week semesters of a rigorous engineering physics course sequence in parallel with four rigorous calculus courses, two in differential calculus and two in integral calculus followed by differential equations, linear algebra, numerical analysis, complex variables and other advanced topics all before the junior year. On the other hand pre med majors took two semesters of watered down physics without calculus.
In addition to numerous discussions of statistics in nearly all physics classes, I took advanced statistics and probability theory at Berkeley.
How much statistics do you think most medical doctors take? They are kept away from these "hard" classes because otherwise there wouln't be enough of them around, get it?
Even the physicists are corrupt for their dismal failure to speak up about this academic corruption for over 50 years. And even today, the introductory physics courses at the "best" schools like MIT have been watered down to a lousy two semesters. Physics is irrelevant today, made so by specialization. All they do today is mechanics and electricity and magnetism.
We did mechanics, heat, sound, electricity and magnetism, optics and modern physics. All had labs too. We even performed the Millikan Oil Drop Experiment in the introductory lab! Shame, shame, shame on these low level schools for lowering the standards.

Posted by: Winfield J. Abbe | Nov 19, 2010 7:52:38 PM

It is noteworthy that the illustration is in error. The graph is of a probability DENSITY function (pdf). The vertical axis is labelled incorrectly as 'probability' - it is only the area shaded orange that can possibly be so designated. The shaded area is the probability that the variate is larger than the value indicated on the horizontal scale by the dot.
The range of values on the vertical axis of a pdf can be as large or as small as is appropriate to the variate it characterises. On the other hand, probability can only range from zero to one. One, ipso facto, is the total area under any pdf.

Posted by: narayan | Nov 20, 2010 3:29:01 PM

Posted via email from hypha's posterous

Aucun commentaire:

Enregistrer un commentaire