Shareef Siddeek wrote:
> Anyway, what is the general, easy to understand, definition of df that could be
> applied in any situation: chi-square test, ANOVA, linear and non-linear
> multiparameter model fitting by least square, MLH method. etc.?
This question is raised every few years. It's about time for it to
cycle around again. There will undoubtedly be some attempts at an
answer, but if "easy to understand" means nonmathematical, the
answer is that there isn't any. Good's article is about as good as
it gets (pun is probably intended at some unconscious level).
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================
