Date: Fri, 22 Mar 2002 10:54:24 -0800
Sender: "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From: "David L. Cassell" <Cassell.David@EPAMAIL.EPA.GOV>
Subject: Re: Normality test
Content-type: text/plain; charset=us-ascii
"Elmaache, Hamani" <Hamani.Elmaache@CCRA-ADRC.GC.CA> wrote:
> Which Procedure can I use to make Normality test?
Without more information, I'll venture to say PROC UNIVARIATE.
Just add the NORMAL option into the PROC statement. I recommend
that you go with the Shapiro-Wilk statistic, rather than one of
the stats based on the empirical distribution function. I wrote
rather a lot on this subject a few days ago, so you can search
the archive for my ramblings if you want to.
But why do you need a test of normality, and how big is your
dataset, and what other strictures are there on your analysis?
Sometimes people want a test of normality when it is (1) not
needed, or (2) not appropriate. So I'm just wondering. [Reply
to the list on this, please.]
David Cassell, CSC
Senior computing specialist