Guest guest Posted February 15, 2005 Report Share Posted February 15, 2005 Never believe a statistic unless you understand the exact parameters it seeks to describe. Also, study abstracts can be very misleading. The full study should 'always' be read, if you wish to critically understand the protocol, IMHO. -pz > Re: statistics and cancer > > A few scientists I know have told me that statistics are very easy > to manipulate in order to make a point. I like to have a healthy > scepticism of them and of anyone that says they have 'the answer' > for...(fill in the blank). > Unless of course I agree with them. > Cheers,\>>> > >>>>>>> > That is why you need to study statistics as well as the studies you > review. > Quote Link to comment Share on other sites More sharing options...
Guest guest Posted February 16, 2005 Report Share Posted February 16, 2005 pz, " them " in my post referred to statistics not necessarily the scientists. I agree with you and Alon about the need to read and understand studies along with statistics. IMHO,we all come to look at studies with our own personal bias and that will color our view no matter how 'scientific' we are. almost always never say never Jill Likkel > Never believe a statistic unless you understand the exact parameters it > seeks to describe. Also, study abstracts can be very misleading. The full > study should 'always' be read, if you wish to critically understand the > protocol, IMHO. > -pz > > > Re: statistics and cancer > > > > A few scientists I know have told me that statistics are very easy > > to manipulate in order to make a point. I like to have a healthy > > scepticism of them and of anyone that says they have 'the answer' > > for...(fill in the blank). > > Unless of course I agree with them. > > Cheers,\>>> > > >>>>>>> > > That is why you need to study statistics as well as the studies you > > review. > > Quote Link to comment Share on other sites More sharing options...
Guest guest Posted February 17, 2005 Report Share Posted February 17, 2005 Statistically True: placebo pills are used in clinical trials to measure the true effect of a drug or supplement. Statistically True: Drug companies make their own placebo pills for research purposes. And that THEY choose the placebos that match those in the drug and will affect the outcome of the trial. And they are not required to disclose the ingredients they use. Statistically True:; St John's Wort, a Placebo, and Prozac are Statistically as effective with the Placebo having the least side affects. True: The number of physicians in the U.S. today is +/- 700,000 True: Accidental deaths caused by physicians per year is 120,000 True: Accidental deaths per physician is 0.171 (source: U.S. Dept. of Health & Human Services). Statistically UNTrue: the physician does not actually kill anybody. Drugs kill. THINK ABOUT THIS: A. The number of gun owners in the U.S. is 80,000,000 (yes, eighty million). B. The number of accidental gun deaths per year (all age groups) is 1,500 C. The number of accidental deaths per gun owner is .0000188 Statistically, doctors are about 9,000 times more dangerous than gun owners. Yet. the public believes ma huang, st john's wort, ban Xia and soon ginseng are dangerous and either should be banned or only dispensed by a physician ! Ed Kasper LAc. Licensed Acupuncturist & Herbalist Acupuncture is a jab well done www.HappyHerbalist.com Santa Cruz, CA. Quote Link to comment Share on other sites More sharing options...
Guest guest Posted February 17, 2005 Report Share Posted February 17, 2005 Statistically True: Drug companies make their own placebo pills for research purposes. And that THEY choose the placebos that match those in the drug and will affect the outcome of the trial. And they are not required to disclose the ingredients they use. >>>>Ed what is your point with this one? Quote Link to comment Share on other sites More sharing options...
Guest guest Posted February 18, 2005 Report Share Posted February 18, 2005 RE: Ed Kasper's " Statistically True " The first two paragraphs refer to methodological issues, not statistical ones. The third paragraph should cite a source, as it is a fact with which some " might " disagree, or it is something others might like to cite in their own writings. Having read the thread about statistics, it strikes me that if herbalists or acupuncturists want to be considered mainstream, then they should understand the tools of scientific rationalism so that when they criticize it, they do so intelligently. Certainly there is plenty to criticize. Misusing terms, bashing statistical inference as a way to lie, and alternately retreating from scientific thought to Chinese dogma and back again only makes the profession look like the chiropractic profession of the 1970's. You certainly don't want that to continue. Quote Link to comment Share on other sites More sharing options...
Guest guest Posted February 18, 2005 Report Share Posted February 18, 2005 Here are some statistics that could convince any layperson that our herbs pose a great danger: http://www.medceu.com/tests/herbal.htm Ed Kasper LAc <eddy wrote:Statistically True: placebo pills are used in clinical trials to measure the true effect of a drug or supplement. Statistically True: Drug companies make their own placebo pills for research purposes. And that THEY choose the placebos that match those in the drug and will affect the outcome of the trial. And they are not required to disclose the ingredients they use. Statistically True:; St John's Wort, a Placebo, and Prozac are Statistically as effective with the Placebo having the least side affects. True: The number of physicians in the U.S. today is +/- 700,000 True: Accidental deaths caused by physicians per year is 120,000 True: Accidental deaths per physician is 0.171 (source: U.S. Dept. of Health & Human Services). Statistically UNTrue: the physician does not actually kill anybody. Drugs kill. THINK ABOUT THIS: A. The number of gun owners in the U.S. is 80,000,000 (yes, eighty million). B. The number of accidental gun deaths per year (all age groups) is 1,500 C. The number of accidental deaths per gun owner is .0000188 Statistically, doctors are about 9,000 times more dangerous than gun owners. Yet. the public believes ma huang, st john's wort, ban Xia and soon ginseng are dangerous and either should be banned or only dispensed by a physician ! Ed Kasper LAc. Licensed Acupuncturist & Herbalist Acupuncture is a jab well done www.HappyHerbalist.com Santa Cruz, CA. Chinese Herbal Medicine offers various professional services, including board approved continuing education classes, an annual conference and a free discussion forum in Chinese Herbal Medicine. Quote Link to comment Share on other sites More sharing options...
Guest guest Posted February 19, 2005 Report Share Posted February 19, 2005 , DrGRPorter@a... wrote: > > Misusing terms, bashing statistical inference as a way to lie, and > alternately retreating from scientific thought to Chinese dogma and back again only > makes the profession look [bad] We seem to like statistics that show most disease is related to diet and lifestyle, which thus validates our long held belief. But we categorically reject research that supports the use of therapies that don't mesh with our philosophies, like that on gastric bypass and mortality. 4 times as many people will die from untreatable diseases associated with obesity than will die if they get the surgery (including a certain number of perioperative deaths). This should obviously be a last resort, but would you sentence someone to death because they couldn't stay on a diet. That doesn't seem very humane, either. Statistics need to be considered critically. They can only be manipulated if you don't read and understand the details. Quote Link to comment Share on other sites More sharing options...
Guest guest Posted February 19, 2005 Report Share Posted February 19, 2005 They can only be manipulated if you don't read and understand the details. >>>Very true but often seen Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You are posting as a guest. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.