Date:         Mon, 25 Oct 1999 10:26:01 -0500
Reply-To:     "Nichols, David" <nichols@SPSS.COM>
Sender:       "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From:         "Nichols, David" <nichols@SPSS.COM>
Subject:      Re: hierrachical model
Comments: To: Cuong Pham <c_pham_viet@HOTMAIL.COM>

SPSS does not currently have a procedure analogous to PROC MIXED in SAS. You
can fit a linear model with nested effects in GLM/UNIANOVA, but you get OLS
estimation, not ML/REML. You can get ML/REML estimates of variance
components in VARCOMP, but no tests of fixed effects. There is a macro
available from http://www.spss.com/tech/macros written by Don Hedeker of the
University of Illinois here in Chicago, that might give the kind of results
you want. Scroll down to the section "Macros not created by SPSS" and click
on the first offering.

David Nichols
Principal Support Statistician and
Manager of Statistical Support
SPSS Inc.

-----Original Message-----
From: Cuong Pham [mailto:c_pham_viet@HOTMAIL.COM]
Sent: Saturday, October 23, 1999 1:26 AM
To: SPSSX-L@LISTSERV.UGA.EDU
Subject: [SPSSX-L] hierrachical model


Dear folks

I want to use hierrachical model to analyse a data set with the subjects are
nested within schools.

Is there anyway I can do it in Spss 9.0 ? (I know that SAS can do this kind
of data with the Proc Mixed)

Thank in advance.

Cuong