Not sure that I understand your question.
UniAnova can be used to generate contrasts that compare the
overall mean of ONE variable to the means of that same ONE
variable by groups. I just ran UniAnova with Deviation contrasts
and it works for me. One "Gotcha" - When you use the SPSS GUI
to specify a contrast, you must press the CHANGE button to
make the Deviation contrast take effect.
Here is sample syntax
UNIANOVA
y BY g
/CONTRAST (g)=Deviation
/METHOD = SSTYPE(3)
/INTERCEPT = INCLUDE
/CRITERIA = ALPHA(.05)
/DESIGN = g .
-----Original Message-----
From: SPSSX(r) Discussion [mailto:
[hidden email]] On Behalf Of
[hidden email]
Sent: Tuesday, September 11, 2007 2:59 PM
To:
[hidden email]
Subject: Deviation Contrasts
I feel like I must be missing something, but here goes.
If I wanted to compare the overall mean of one variable vs the mean of each
group (except one) of another variable, it seems I should be able to use GLM
Univariate and ask for a deviation contrast. If the overall mean is 10, and
the mean of one group is 15, I would expect the value of that particular
contrast to = 5. What I'm finding is that the value of the contrast is not
equal to the difference between the group mean and the overall mean - and so
I must be misinterpreting this.
Any suggestions.
Locked
