Half a croissant, on a plate, with a sign in front of it saying '50c'
h a l f b a k e r y
Futility is persistent.

idea: add, search, annotate, link, view, overview, recent, by name, random

meta: news, help, about, links, report a problem

account: browse anonymously, or get an account and write.



Homeopathic statistics

Two wrongs, one right.
  (+8, -1)
(+8, -1)
  [vote for,

Homeopathy and bad statistics - they were made for eachother.

There are probably at least 1000 recognised, serious medical complaints, ranging from pancreatic cancer to piles.

It follows that, if a large collection of people taking any given homeopathic remedy are surveyed for these 1000 serious illnesses, at least one such illness will be found to be significantly more frequent in the remedy-takers, as compared to the normal population, at a confidence level of p=0.001.

Once this statistic has been unearthed, the full weight of regulatory power (or the tabloid press) can be brought to bear.

MaxwellBuchanan, Aug 30 2011

Texas sharpshooter fallacy http://en.wikipedia...harpshooter_fallacy
[xaviergisz, Aug 30 2011]

XKCD http://xkcd.com/882/
As ususal, there is an XKCD cartoon which describes this perfectly [hippo, Aug 31 2011]


       // the full weight of regulatory power (or the tabloid press) //   

       But applying a tiny fraction of said power would be even more effective, shirley ?
8th of 7, Aug 30 2011

       yeah but is "gullibility" a recognized disease ?   

       Didn't we work out that there might be something to it; the body's defenses act in the same way that birds react to a silhouette of a hawk cut out of cardboard... (or mumblings to that effect)
FlyingToaster, Aug 30 2011

       It's. Water.
MaxwellBuchanan, Aug 30 2011

       Didn't understand that last annotation, didn't understand main post. Is there a fallacious correlation is causation implication? Are you a homeophobe? Are you an adherent to the spiritual system of an obscure North American Indian tribe that worships the bear?
rcarty, Aug 30 2011

       //worships the bear// Does Ursula Andress count ?
FlyingToaster, Aug 30 2011

       Yes, bare.
rcarty, Aug 30 2011

       This may surprise you but i think this is an excellent idea. However, sometimes It's Lactose.
nineteenthly, Aug 31 2011

       Hear, hear! Take it to'em MB.
AusCan531, Aug 31 2011

       MB, have you lost most of your medical encyclopedic collection with the exception of Pa - Pi?
po, Aug 31 2011

       // Is there a fallacious correlation is causation implication?// Yes, and that should be good enough for a few headlines.
MaxwellBuchanan, Aug 31 2011

       oh how weird, [hippo] I was just looking at a bear cartoon from there.
po, Aug 31 2011

       [hippo]'s link depicts a sort of inverse situation, testing many compounds against one illness, rather than one compound against many illnesses. However, both approaches have their merits.
MaxwellBuchanan, Aug 31 2011

       Interesting that everyone is concentrating on the homeopathy angle. I've yet to see any statistics that weren't badly used. Statistics themselves are neither bad nor good; unless they are just made up of course, in which case they are lies rather than statistics.
DrBob, Aug 31 2011

       //I've yet to see any statistics that weren't badly used// - 100% of statistics seen by [DrBob] he judges to have been "badly used".
hippo, Aug 31 2011

       See, you just wasted that statistic on me when you could have used it on somebody who cares. Somebody who would probably, even now, be waving their fist ineffectually at their monitor and muttering "Curse that damned hippo!" under their breath. But that person wouldn't be me of course. I'm too cool. And calm and composed ... composed ... yeah, that's what I am. And calm. So yet another badly used statistic.
DrBob, Aug 31 2011

       100% of Borg Collectives reading [DrBob]'s last anno consider that he is neither calm nor collected, but is instead operating in a mode best described as "low-level panic", and not hiding it very well.
8th of 7, Aug 31 2011

       How does one avoid sampling bias in polling Borg collectives?
pertinax, Aug 31 2011

       Easy, just ignore them.
pocmloc, Aug 31 2011

       [pertinax] Simplicity itself. You only need an N of 1, and it doesn't matter who you pick.
mouseposture, Aug 31 2011

       isn't at least the partial weight of regulatory power already brought to bear, e.g. admonitions such as "this product is not intended to cure any disease" which routinely accompany advertising for homeopathic products?   

       I don't think there's much new here.
theircompetitor, Aug 31 2011

       The majority of the "regulatory power" is based on bad statistics. Therefore, much though you might like to, you'll hardly be able to make the situation worse.
lurch, Sep 01 2011

       Saying homeopathy should be subject to regulation is tantamount to saying people shouldn't be penalized for their gullibility. [-]
ytk, Sep 01 2011

       //"this product is not intended to cure any disease"//   

       Would cure dehydration, I'd have thought.
Loris, Sep 01 2011

       // I don't think there's much new here //   

       If you look very closely, we think you'll find that there's just a very tiny amount of New, diluted in a large volume of Old.
8th of 7, Sep 01 2011


back: main index

business  computer  culture  fashion  food  halfbakery  home  other  product  public  science  sport  vehicle