Date:         Wed, 8 Oct 1997 12:18:10 -0700
Reply-To:     Hiroyuki Yamada <hryamada@SOCRATES.BERKELEY.EDU>
Sender:       "SAS(r) Discussion" <SAS-L@UGA.CC.UGA.EDU>
From:         Hiroyuki Yamada <hryamada@SOCRATES.BERKELEY.EDU>
Subject:      Which method to use in Proc CALIS using a polychoric corr matrix
Content-Type: TEXT/PLAIN; charset=US-ASCII

Hi, All,
 
        I am trying to use a polychoric correlation matrix and put this as
input to Proc CALIS. Then, my question is: Which METHOD is the most
appropriate for this analysis?
        The on-line manual suggests to use METHOD=ULS. But, one professor
suggested to me that I use METHOD=GLS. I am now a bit confused. There may
be no strict rule on this. But I would really appreciate it if you give me
any thoughts and suggestions regarding what METHOD to be used and
your reasoning and justification for it. Thanks.
 
Hiro