all of this discussion re: t test and normality and when it makes a
difference and when it does not ... plus all the other assumptions made in
all kinds of tests .... with similar sometimes/sometimes not problems ...
just reinforces the fact that we need to be real skeptical about p values
... in fact, this entire process of letting "results" and "interpretations"
from studies hang in the balance DUE to our "significance" tests ... is a
huge myth we have been perpetrating for ever so long ...
when we will wake up?
our focus should be on the design and implementation of investigations ...
FAR MORE and, on the statistical analysis .... FAR LESS
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================
