|
Hi - I read David Nichols' posting from way back in July 1996 'Re: GLM and MANOVA (SPSSWIN 7.0)
A bug?', as well as the relevant sections in Levine's Guide to SPSS for Analysis of Variance;
however, I wanted to make sure that I'm on the right track with a varying covaraite within a mixed
factor design. Any help would be greatly appreciated.
1) Here is my current syntax for SPSS 11 for Mac OSX:
MANOVA
sp_0 sp_1 sp_6 sp_12 ip_0 ip_1 ip_6 ip_12 BY conditio(1,4) ip_sp(1,2) WITH tf_0 tf_0dum tf_1
tf_1dum tf_6 tf_6dum tf_12 tf_12dum
/wsfactors = test(2) rp(4)
/omeans
/design.
I added in the dummy covariates so that the number of covariates matches with the number of
wsfactor levels. Is this the right thing to do if I want the covariates to vary with the within-subject
factor of rp?
2) I am also working on trying to get the adjusted means, with the tf covaried out. I realize that
this is not straightforward, given that I have multiple within subject factors, one with more than 2
levels (so PMEANS doesn't work properly). I was wondering if there is an easy way to get the
correct adjusted means table.
Thanks so much,
Justin Hulbert
|