I seem to recall it was R.A. Fisher who suggested 0.05 as a measure
of disbelief in the null hypothesis.
Both in lectures to first-year students and to seminars to
professionals I often conduct the following class experiment to
explain why 0.05 is a plausible reflection of how we judge things.
Ask each individual in the room to state how many SUCCESSIVE heads
or SUCCESSIVE tails he/she would need to observe when an "arbitrary"
coin is repeatedly tossed before concluding that the coin is not
fair. This is a two-sided significance test based on a runs
statistics for which it is easy to elicit responses. For each
person compute the corresponding level of significance -- for
instance, if I said 7 heads or tails in a row then the probability of
observing 7 or more heads or tails in succession would be 2 x
(0.5)^6) -- and take the average level of significance. The results
have always been surprisingly in the vicinity of 0.05 (though usually
less) and the effect on a group of engineers or managers is always
dramatic. Students without experience with real data don't really
get it but they like to interact anyway.
>Date: Fri, 17 Mar 2000 12:16:35 -0600 (CST)
>From: David Cross/Psych Dept/TCU <[EMAIL PROTECTED]>
>Subject: Re: Power for Pilot Studies
>
>I do believe that to the extent there is a "conventional" alpha level,
>that level is .05. In the behavioral sciences, this convention is adhered
>to rather strictly, especially by journal editors and review panels!
...........................................................................
@@@@@@@@@@@
@@@@@@@@@@@@@ Fred M. Hoppe, Ph.D.
@@@@@@@@@@@@@@@ Professor, Mathematics and Statistics
@/ __ __ \@ McMaster University
@| (q) (p) |@ 1280 Main St. West
{ / \ } Hamilton, Ontario, L8S 4K1, Canada
| ~~~~~ | Phone: (905) 525-9140 Ext. 24688
| \_____/ | Fax: (905) 522-0935
\ / E-mail: [EMAIL PROTECTED]
-<>-<>-
...........................................................................
===========================================================================
This list is open to everyone. Occasionally, less thoughtful
people send inappropriate messages. Please DO NOT COMPLAIN TO
THE POSTMASTER about these messages because the postmaster has no
way of controlling them, and excessive complaints will result in
termination of the list.
For information about this list, including information about the
problem of inappropriate messages and information about how to
unsubscribe, please see the web page at
http://jse.stat.ncsu.edu/
===========================================================================
