home *** CD-ROM | disk | FTP | other *** search
- Comments: Gated by NETNEWS@AUVM.AMERICAN.EDU
- Path: sparky!uunet!paladin.american.edu!auvm!PO.CWRU.EDU!PAT
- Message-ID: <9208241239.AA11210@slc10.INS.CWRU.Edu>
- Newsgroups: bit.listserv.sas-l
- Date: Mon, 24 Aug 1992 08:39:29 -0400
- Reply-To: "Paul A. Thompson" <pat@po.CWRU.Edu>
- Sender: "SAS(r) Discussion" <SAS-L@UGA.BITNET>
- From: "Paul A. Thompson" <pat@PO.CWRU.EDU>
- Subject: Re: discussion group for covariance structure modeling?
- Comments: To: KHILL%LUCY.WELLESLEY.EDU@ncsuvm.cc.ncsu.edu
- Comments: cc: sas-l@vtvm2.cc.vt.edu
- Lines: 37
-
- >
- >To: EDSTAT-L@NCSUVM.CC.NCSU.EDU
- >X-VMS-To: NET::"EDSTAT-L@NCSUVM.CC.NCSU.EDU"
- >
- >Hi!
- >
- >I am currently in a group planning a fairly large study for this academic
- >year in which we intend to use Covariance Structure Modeling (probably
- >LISREL VI) to analyze the data, however, the PI is suggesting a design
- >which I just don't know if LISREL can handle. [She wants to embed a
- >within-subjects experimental manipulation into the design....were it
- >between-subjects I could see where we could use stacked models...all of the
- >literature that I can find on analyzing experimental designs using CSM
- >talks about _between_ subjects!].
- >
- >I realize this is not the forum for this question, so I am asking if anyone
- >knows if there is a LISREL or CSM discussion group that I might
- >pass this by, and how I might subscribe to it.
- > Thanks!
- >
- I really think that this is straight-forward. Remember, the multivariate
- approach to repeated measures (MARM) places the different levels of the
- repeated factor on the dependent side, and then uses contrasts among them
- to test hypotheses. This is what SEM does, except not as explicitely.
- So, you must find a covariance structure which tests the MARM hypothesis,
- which could be done by setting up the appropriate dummy variables, etc.
- I am not really a SEM pro, so this is just a guess.
-
- You may also wish to consult someone who is a pro. Numerous choices
- abound. It may be the case that the SAS tech report re SEM (CALIS) has
- an example relevant.
-
- --
- Paul Thompson, Ph.D. | Department of Psychiatry | (216) 844-7463
- Case Western Reserve University | Cleveland, OH 44106
- ------------------------------------------------------------------------------
- When I hear "family values," I reach for my revolver. - paraphrasing Trotsky
-