Spinning is also not 50/50, on US pennies at least. If you give them
very vigorous spins on a stable flat surface, tails comes up more often
than heads. I have done both of these in class in the context of null
hypothesis tests (setting null at .5); The last time I spun pennies I
got 73 tails of 120 spins (61%). I asked a physicist here to explain the
spinning bias years ago, but forgot the details. I vaguely remember his
answer tied in with the bevel issue, but forgot the details. 


============================================
John W. Kulig
Professor of Psychology
Plymouth State College
Plymouth NH 03264
============================================
Za dvumia zaitsami pogonish'sia - ni odnogo ne poimaesh' [If you chase
after two rabbits, you won't catch even one]Russian proverb.

> -----Original Message-----
> From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]
On
> Behalf Of Duncan Murdoch
> Sent: Thursday, February 26, 2004 9:22 PM
> To: [EMAIL PROTECTED]
> Subject: Re: [edstat] tossing a coin is not a random process
> 
> On 26 Feb 2004 15:56:41 -0800, in sci.stat.edu [EMAIL PROTECTED] (Dennis
> Roberts) wrote:
> 
> >A key "may be" in the minting process ... I heard once at a
conference
> >where a demo was done .. lining up coins on edge and pounding the
table
> ...
> >that the p is not 50/50 for H and T to land face up ... that there is
a
> >slight bevel to the edge (not noticeable by eye) that makes ejecting
from
> a
> >mold easier
> 
> The bias for "tipping" coins like that is very noticeable.  I do an
> experiment on that in my intro stats course most years.  As long as
> the table is flat, and you don't hit it too hard, the probability of H
> is around 0.4 with Canadian pennies.  (This year's data was 130 H in
> 345 tips.)
> 
> Duncan Murdoch
> .
> .
> =================================================================
> Instructions for joining and leaving this list, remarks about the
> problem of INAPPROPRIATE MESSAGES, and archives are available at:
> .                  http://jse.stat.ncsu.edu/                    .
> =================================================================

.
.
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
.                  http://jse.stat.ncsu.edu/                    .
=================================================================

Reply via email to