At 09:42 PM 4/3/02 +0200, Virgil Stokes wrote:
>Is there an exact algorithm for calculating the sample size in a paired
>t-test for
>the mean difference when:
>
>  1) the standard deviation of the population is unknown

if you can't assume that the two sigmas are some values .... different or 
the same ... then i think you will NOT have one of the pieces of 
information needed to solve for n in this scenario



>  2) there is a finite population
>  3) continuous measurements
>  4) both alpha and beta (or the power) for the test are specified
>
>Any references would be appreciated.


and, if you know beta ... then you know power but, power for WHAT? 
detecting a difference of how large between the null and the assumed to be 
true effect?

without other information ... i think you are in a pickel


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

Dennis Roberts, 208 Cedar Bldg., University Park PA 16802
<Emailto: [EMAIL PROTECTED]>
WWW: http://roberts.ed.psu.edu/users/droberts/drober~1.htm
AC 8148632401

.
.
=================================================================
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