problrm of Multiple testing

[morphmet]

-------- Original Message --------
Subject:        problrm of Multiple testing
Date:   Tue, 12 Jan 2010 12:32:09 -0800 (PST)
From:   Samor Gandhi <samorgan...@yahoo.com>
To:     morphmet_modera...@morphometrics.org



Hello,

I would like to run multiple testing and I have problem that I have
different number of samples in each group and my data is extreme value
distributed so if I apply the t-test or the Wilcoxon test and I adjust
my p-values no tests will be significant! If I would apply a confidence
interval instead of the t-test or the Wilcoxon test do I need to any
adjustment step?

Many thank in advance,
Amor



--
Replies will be sent to the list.
For more information visit http://www.morphometrics.org