Date:         Wed, 29 Jan 1997 14:54:50 GMT
Reply-To:     David Nichols <nichols@SPSS.COM>
Sender:       "SPSSX(r) Discussion" <SPSSX-L@UGA.CC.UGA.EDU>
From:         David Nichols <nichols@SPSS.COM>
Organization: SPSS, Inc.
Subject:      Re: ANOVA

In article <970128145146_981679469@emout03.mail.aol.com>,
Allison McCluer  <AMcCluer@AOL.COM> wrote:
>I recently got SPSS 6.1 for Mac.  I need to do a 2x2x2 Analysis of Variance.
> All I seem to be able to do is a one way ANOVA.  I'm not sure what I'm doing
>wrong.  Well, to be honest I'm not sure what I'm doing.  I would appreciate
>any help anyone can offer.  Thanks,
>                                             Allison
 
Click on Statistics->ANOVA Models and then see what the options are. If you
see only Simple Factorial, you can use that for some models you might fit
with a 2^3, including the full factorial model. If you also have General
Factorial, Multivariate and Repeated Measures, you might want to use the
General Factorial option, as it offers more things (parameter estimates,
ability to fit any model, etc.).
 
If you only have the Base module (just Simple Factorial on that menu) and
want to fit a model that involves some but not all of the two way
interactions, you'd have to use REGRESSION to do that. In that case, it's
important to understand how choices about parameterization of the model
affect interpretation of the parameters and generation of sums of squares.
There is a series of articles you might want to look at on the SPSS ftp
site. You can reach them via the Web page, by clicking on Tech Support,
then Statistical Articles under additional resources. The paramest.txt
and the catreg1-5.txt articles are of particular relevance. Assuming you
find some interactions, the simple.txt article is also relevant if you
have access to the other options (which means you have the Advanced
Statistics module with the MANOVA procedure). Simple effects tests can
also be done in REGRESSION, as alluded to in the catreg articles.
 
--
-----------------------------------------------------------------------------
David Nichols             Senior Support Statistician              SPSS, Inc.
Phone: (312) 329-3684     Internet:  nichols@spss.com     Fax: (312) 329-3668
-----------------------------------------------------------------------------