!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
CHANCE News 2.10
(16 May to 1 June,1993)
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
Prepared by Laurie Snell
Please send suggestions for articles or format to
dart.chance@dartmouth.edu
Previous issues of this chance news and other materials for
a chance course are on the chance gopher. Just point your
gopher to chance.dartmouth.edu
===================================================
I shall never believe that God plays dice with the world.
Albert Einstein
===================================================
ARTICLES ABSTRACTED
>>>>>==========>> Here is a contribution from Peter Doyle. Formula projects limits seen on human existence. The New York Times, 1 June 1993, Sec. C Page 1. Macolm W. Browne J. Richard Gott, Princeton astrophysicist, has written an article in the current issue of Nature explaining how to obtain 95% confidence intervals for the remaining lifetime of about anything you wish. The idea is that we find ourselves at a randomly chosen point of the lifetime of whatever it is, and thus with probability .95 we are somewhere between 1/40 and 39/40 finished. Say, for example, we estimate that the human race has been around for about 200,000 years. If we're only 1/40 of the way through the lifetime of the human race, then we have 39 times 200,000 years or about 8 million years left to go. If we're at the other end of the 95% confidence interval, that is 39/40 of the way through, then we have only 1/39 times 200,000 years or about 5000 years to go. Having developed this beautiful rule of thumb, Gott goes on to demonstrate, by example, the dangers of taking this kind of thing too seriously. He concludes that the space program, now 32 years old, will (with 95% confidence) end before another 1200 years are up, surely too short a time for us to colonize the galaxy and thus escape the 8,000,000 year deadline derived above. Fortunately for the human race, Gott's whole theory is only about a year old, and thus can be expected to last somewhere between another 39 years and another 9 days. Thus it will most likely die long before it has a chance to doom the space program, and thereby the whole human race. <<<========<<
>>>>>==========>> Exam ruling derails foreign doctor's dream. Los Angeles Times, 1 June 1993, A3. Tony Perry Luis Garcia-Higgins is a naturalized U.S. citizen who is licensed as a doctor in Mexico and had thought that, after a number of unsucessful tries, he had finally passed the Foreign Medical Graduate Examination taken in San Diego last January. However, after taking the test he received a letter saying that a statistical analysis had "revealed a strong relation of joint wrong answers between you and one or more examinees, which suggests that the examination may not be a valid measure of your knowledge". Actually, the strong relations was with five people who took the test in New York, one in Philadelphia and one in New Orleans. In all 109 people got such a letter after the January test. The letter to Grarcia said that the "approximate probability" that he arrived at the answers innocently was "less than one in a billion". Garcia received the usual offer to take another test but has decided to fight the case with the help of Professor Peter Irons, a professor of political science at UC San Diego. In addition, an attorney in Rhode Island, Bernard Fergeson, is considering a class-action lawsuit on behalf of medical graduates who have had their test scores invalidated. Commission attorney Hubbard has said that he is confident that their statistical methods will withstand professional review. Of course, the Commission will be careful to say that they have not accused Garcia of cheating and so the statistical issue may never get challenged in court but if it does it will be interesting to follow this case. The article quotes (not very accurately) J. Laurie Snell "one of the nation's leading authorities in probability statistics" (whatever that is). We used this last article for discussion in our last CHANCE class. We decided this time to have the students come up with questions which they would then answer in groups. Here are the questions they proposed: Is Garcia being accused of cheating? Is Garcia a cheater? Could it be that some questions are harder than others, so that people will tend to get the same questions wrong? Could the language barrier have been a problem? Should Garcia volunteer to take the test again? Would you want this guy to be your doctor? Shouldn't we be more concerned about matching right answers than matching wrong answers? Should there be a limit on the number of times you can take the test? Why wasn't the test cancelled? The questions led to a lively discussion. <<<========<<
>>>>>==========>> Report calls simple test effective in reducing colon cancer deaths. The New York Times, 13 May, 1993, Section A, Page 1. Jane E. Brody After the last chance news, Nathaniel Beck commented: You missed the NY Times piece on using fecal-occult (didn't know medicine had a sense of humor!) blood test to detect colon cancer. Nice discussion on the continuation of the story inside on a simple cost-benefit analysis (though my own cost-benefit analysis on the numbers presented shows that cost per life saved is quite reasonable by American standards). <<<========<<
>>>>>==========>> Colon cancer screening; contradictory results puzzle researchers. Washington Post, 25 May, 1993, Page Z9. Sandra G. Boodman Researchers at the Mayo Clinic recently published in the Journal of American Medical Association a study that concluded that simple and widely used stool tests were ineffective in detecting colon cancer. The study reported in the previous article reached as very different conclusion. This article speculates on reasons for the difference and presents the views of experts on where we are with this problem (confused). <<<========<<
>>>>>==========>> The reader response: Oyvey. Science, 21 May, 1993. Editorial by Daniel E. Koshland, Jr Science has recently published issues on special topics, such as women in science, and have invited reader response and published a summary of responses saying, for example, that 76% of the respondents felt one way about a subject. Science also received an avalanche of letters from statisticians who were concerned with the use of the word poll or survey for such a non scientific process. When Science pointed out that they were in fact interested in hearing from those who felt most strongly enough about the issues to write and had made no claim to be doing a accurate poll. The statisticians responded by saying that Science was such a prestigious journal that they were still worried that people might think that this was an o.k. way to do a serious poll. Apparently, the statisticians involved accepted the editor's proposal that this kind of study be called an "Oyvey" survey. It is remarked that this change resulted from on Oyvey of statisticians. <<<========<<
>>>>>==========>> The art of turning the tables. The Times, 20 May 1993, features. Tom Rhodes A review of the book THE MONEY SPINNERS, How Professional Gamblers beat the Casinos at their own Game, By Jacque Black, published by Faber. It is presumed that Black is not the author's name but that the author has had a great deal of experience gambling in London casinos. The book provides accounts of some of the famous characters in gambling such as Edward Thorpe and a complete account of the history of the casino from the Palais Royal in Paris to Las Vegas and Atlantic City. The appendices which constitute about a third of the book offer guides to games and how to beat the system. More later after I have read the book if I find it interesting. <<<========<<
>>>>>==========>> Studies show Vitamin E may reduce heart disease. Los Angeles Times, 20 May, 1993 Sheryl Stolberg The current issue of the New England Journal of Medicine reports two studies, one with men and one with women that show that men and women who took daily Vitamin E supplements of at least 100 international units had a 40% lower risk of heart disease than those that did not. A possible explanation is that cholesterol must undergo oxidation to get inside the cells that line the walls of blood vessels and Vitamin E helps prevent this oxidation. The studies were based on the Nurses's Health Study and the Health Professionals Follow-Up Study that rely on the subjects' answers to detailed questionnaires about their eating habits. Authors of the studies recommend waiting until more definitive clinical trials are carried out and until there is assurance that their are no long term risks in taking such large doses of Vitamin E. Others point out that this is more caution that is often advised for a drug which has much more potential for harmful side effects. <<<========<<
>>>>>==========>> Gender bias charged in National Merit Scholarship Test. Los Angeles Times, 26 May, 1993, A1 Elizabeth Shogren The organization FairTest reports that only about 35% of the National Merit Scholarship winners are girls despite the fact that studies suggest that girls do better than boys in grades in high school and college. The choice of semifinalists for the National Merit Scholarship is based entirely on the results of the Preliminary Scholastic Aptitude Test. The article refers to the article below that documents the fact that women do less well than men on SAT exams. <<<========<<
>>>>>==========>> Sex differences in Performance on the mathematics section of the Scholastic Aptitude Test: A bi-directional validity study. Harvard Educational Review, Fall 1992. Howard Wainer and Linda S. Steinberg. This study by Educational Testing Service researchers reviews the literature concerning the difference between men and women on the mathematics SAT tests and reports on their own study. They compare SAT-M scores for men and women with the same grades and also grades for men and women with the same SAT-M scores. They show that women consistently do about thirty points lower on these tests. Since women do as well as men in college this raises questions about the proper use of the SAT-M exams in college admission and competitive scholarship programs. They discuss some possible solutions ranging from giving women extra points to doing nothing. The authors favor continuing to try to understand what is going on. <<<========<<
>>>>>==========>> Little benefit seen in prostate surgery Study finds six fold increase in rate of cancer operation since 1984 . Washington Post, 26 May, 1993, A1. David Brown An account of an two articles in the current Journal of the American Medical Association related to treatments for prostate cancer. One study considers possible treatments and suggests that watchful waiting is often preferred to removing the prostate through surgery, or radiation treatment when prostate cancer is identified in its early stages. The authors use utility theory methods (indeed even Markov chain models) to compare the benefits of various possible treatments. The second study finds that the rate of the performing prostate surgery has increased 600 percent from 1984 to 1990. Wide geographical variations are found for performing the surgery. West coast men have the operation almost four times as often as those in New England. A member of the prostate research group is quoted in the New York Times as saying "The take- home message is that we don't know what w'ere doing, but we're doing a lot of it'. The authors suggest that there is very little real evidence that surgery is effective and suggest that if surgical procedures had to be tested and approved by the standards required for drugs that much of the prostate surgery being done would not be permitted with the present information on the success of the surgery. <<<========<<
>>>>>==========>> NRC report on DNA typing. Science, 21 May, 1993 Letters from B. Devlin, Neil Risch and Ranajit Chakraborty Daniel L.Hartl and Richard C.Lewontin wrote a letter to Science 23 April (p 473) concerning the Devlin and Risch critique (Policy Form, 5 Feb.,p 748) of the National Research report on DNA typing. These letters are in reply to this letter. The whole collections gives a good view of the current controversy over the "multiplication rule" and other questions relating to the determination of the probability for a match in DNA fingerprinting. <<<========<<
>>>>>==========>> !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! CHANCE News 2.10 (16 May to 1 June,1993) !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! Please send suggestions to: jlsnell@dartmouth.edu >>>==========>>|<<==========<<<
>>>==========>>|<<==========<<<