[ExI] Randi again
Damien Broderick
thespike at satx.rr.com
Tue Mar 4 17:19:02 UTC 2008
At 01:08 PM 3/4/2008 -0300, HeMm wrote:
>Randi responded to
>the article
>(http://www.randi.org/joom/content/view/169/1/#i9) ... This can go on and
>on forever. But I still trust Randi more than the grail and object to the
>term 'bogus'.
Uh huh. You trust this sort of thing, eh?
<Randi quote starts:>
My abysmal ignorance of statistics requires that
I frequently appeal to statistician Chip Denman
of the University of Maryland for frequently
sobering advice and counsel. Having just received
some of that wisdom, Im announcing a further
refinement and generous it is! to the JREF
million-dollar challenge. These changes will go
up on the rules page as soon as we can get around to it
Says Mr. Denman:
...Setting the bar for significance is
"merely" a matter of deciding how risk-tolerant
you're willing to be. I believe that it is
entirely sensible to set a high bar for the $1M
prize. Maybe one out of a million is a bit
extreme, but it's your money and your risk.
On the other hand, you might consider a
lower bar for the preliminary test and still
protect yourself overall. For instance, you could
use .01 (which is frequently seen in the
scientific and statistical literature) for the
preliminary, and a 1 out of 100,000 rule for the
final test and taken together, you'd know there
was only a one-in-a-million shot that someone could get lucky on both.
Thats what well do. Well choose the other
hand. So, as of now, we will require that
applicants beat a one-in-one-hundred chance of
success by dumb luck or co-incidence for the
preliminary test, and then a
one-in-one-hundred-thousand chance in the formal
test a point that has not yet been reached in
the past ten years of our trying
<end Randi quote>
So both the abysmally ignorant Randi and his
stats expert tell us that 1 in 100 multiplied by
1 in 100,000 equals 1 in a million. Leaving aside
the impropriety of multiplying these
probabilities, that's just... amazing, Randi!
Damien Broderick
More information about the extropy-chat
mailing list