Date: Fri, 21 Oct 2005 23:06:57 -0700
Reply-To: David L Cassell <davidlcassell@MSN.COM>
Sender: "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From: David L Cassell <davidlcassell@MSN.COM>
Subject: Re: Skewed variables & surveys
Content-Type: text/plain; format=flowed
Peter Flom replied in part:
>PROC LOGISTIC is not that hard to use for an ordinal variable, in my
>opinion. Of course,
>if the proportional odds assumption is violated, it's a little trickier to
>the results of a nominal logistic regression, but not THAT hard.
>Also, as I believe David Cassell pointed out, if it's a survey the OP
>should be using SURVEYLOGISTIC.
>I've not used this, so can't attest to ease or difficulty, but I am sure
>that help is available. :-)
I would just like ot point out that PROC SURVEYLOGISTIC is really designed
have the 'look and feel' of PROC LOGISTIC. The statements in the proc are
really very similar (they don't do everything proc logistic can do, and they
a couple extra statements for stratum and cluster information if any). The
output is also designed to look a lot like proc logistic's output. If you
PROC LOGISTIC, you can use the other one too.
David L. Cassell
3115 NW Norwood Pl.
Corvallis OR 97330
Express yourself instantly with MSN Messenger! Download today - it's FREE!