Date:         Thu, 15 Apr 1999 15:59:12 -0400
Reply-To:     frank.mwaniki@PHARMA.NOVARTIS.COM
Sender:       "SAS(r) Discussion" <SAS-L@UGA.CC.UGA.EDU>
From:         Frank Mwaniki <frank.mwaniki@PHARMA.NOVARTIS.COM>
Subject:      Re: repost of question
Comments: To: ns33@cornell.edu
Content-type: text/plain; charset=us-ascii

Niren,
I would start with proc calis. there is reference a paper on the sas web site
that you may want start with which should be in sugi21 proceedings. Below is an
abstract.
G'Luck,
Frank
 
it's at http://www.sas.com/usergroups/sugi/sugi21/abstracts/abs348.html
<<Paper 256
 
At SUGI 20, the author presented a paper on performing path analysis with
manifest variables using PROC CALIS. This
paper goes one step further by showing how PROC CALIS can be used to perform
structural equation modeling with latent
variables. It presents a simple, step-by-step approach for testing structural
equation models using the SAS System. The paper
introduces basic issues in structural equation modeling and shows how to perform
 confirmatory factor analysis and
latent-variable path analysis with the CALIS procedure. It illustrates a
two-phase approach to latent-variable path analysis
recommended by Anderson & Gerbing (1988). The paper also shows how to use PROC
CALIS to test the investment model,
a theory of commitment in romantic relations drawn from the social psychology
research literature (Rusbult, 1980). >>
 
 
 
 
 
Niren Sirohi <sirohi@RESEARCH.JOHNSON.CORNELL.EDU> on 04/15/99 03:01:28 PM
 
Please respond to ns33@cornell.edu
 
To:   SAS-L@UGA.CC.UGA.EDU
cc:    (bcc: Frank Mwaniki/PH/Novartis)
Subject:  repost of question
 
 
 
 
anyone out there who has don latent class modeling using SAS? Which proc
does one use?
 
niren