Date:         Fri, 14 Oct 2005 18:23:41 -0400
Reply-To:     Jeff Miller <jeffmiller@ALPHAPOINT05.NET>
Sender:       "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From:         Jeff Miller <jeffmiller@ALPHAPOINT05.NET>
Subject:      Re: mixed models
Content-Type: text/plain; format=flowed; charset="iso-8859-1";
              reply-type=original

I would one to disagree strongly about S&W's HLM approach. But, what do you
mean by an "HLM perspective"?

Jeff

----- Original Message -----
From: <emaguin@acsu.buffalo.edu>
Newsgroups: bit.listserv.spssx-l
To: <SPSSX-L@LISTSERV.UGA.EDU>
Sent: Friday, October 14, 2005 5:40 PM
Subject: Re: mixed models


> Andrew,
>
> I've been looking for something like this too and i think the answer is
> No.
> The Singer and Willett book is very good and, while others may disagree
> strongly, it feels like it's written from an HLM perspective. That said,
> you should look at the UCLA website (www.ats.ucla.edu/stat/spss). The site
> has worked examples of using spss to do growth models. The level of
> detail,
> in conjunction with the S&W book, is like what spss manuals used to be
> like. Other than that, i think you might want to look at SAS
> documentation.
> In particular a book called something like Linear Mixed Models for SAS or
> somthing like that. I have the ref at home. It is written by the guy who
> wrote SAS's mixed model proc. Actually, i'm curious. A question for those
> of you who use both spss and sas for mixed models--if there are any such
> people. If you are familiar with this book, would it help somebody with
> the
> spss mixed model routine?
>
> Gene Maguin
>
>