On 20 Sep 2001 11:05:08 -0700, [EMAIL PROTECTED] (Jon Cryer) wrote:
>(quoting Robert: "even when N=20, a uniform distribution can be treated as
> normal for most purposes.")
>
> I assume you meant to say that for N=20, the sample mean based on a random
> sample from a uniform distribution can be assumed to have a normal
> distribution for most purposes.
>
> Right?
I thought he was intending the stronger statement: a lot of
uniforms can be treated as normal, especially for small N
and for moderate effect size. Conover, et al., showed the
equivalency between doing (a) the old rank-order tests (like the
MWW), and (b) simple t-tests, etc., on the rank-transformations.
Robert waffles by saying 'most' purposes, so I have to
find it easy to agree. When might you *not* treat a uniform,
N=20 as normal? - perhaps when the R^2 is too high
(above .90)?
--
Rich Ulrich, [EMAIL PROTECTED]
http://www.pitt.edu/~wpilib/index.html
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================