Neo Sunrider <[EMAIL PROTECTED]> wrote in message
0gmh6.222677$[EMAIL PROTECTED]">news:0gmh6.222677$[EMAIL PROTECTED]...
> I am just taking an undergraduate introductory stats course but now I
> am faced with a somewhat difficult problem (at least for me).
>
> If I want to test a hypothesis (t-test, z-score etc.) and the underlying
> distribution will under no circumstances aproach normal... (i.e. the results
> of the experiement will always be something like 100*10.5, 40*-5 etc.) The
> Central Limit Theorem doesn't help here, or does it?
That depends. Depending on what you're testing, and on sample
size, it may or may not. But it doesn't matter.
It sounds like you probably have a simple enough situation that
you can make some appropriate assumptions and derive the distribution
of a reasonable statistic.
> Can anyone explain, or point me in the right direction - how can I test in
> these cases?
Nonparametric methods - especially permutation techniques - would be
a good place to start looking. Even if you don't go this way it might
clarify some of your ideas in the specific situation you're dealing with.
Glen
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================
