Dear Shary,
except for 1st-level designs with high-pass filtering, only what is shown
in your design matrix is regressed out of your data. SPM does not make any
difference between covariates of interest or covariates of no interest,
and especially the entire design is fitted simultaneously. You have to
decide what is of interest by defining contrasts and viewing their
results. If a variable is a nuisance variable, then just do not test for
it.
Hope this helps,
Volkmar
On Thu, 7 Dec 2006, [log in to unmask] wrote:
> Dear all,
This is related to my previous e-mail. I have a 2 sample t-test with 3 covariates of no interest
(Nuisances). I have entered as covariates Age, Years of Education and
Sex and centered them as implied by ANCOVA. I repeated this and centered
the covariates by GM (I think this means Grand Mean) with NO
interaction with factors.
In both cases, the matrix (see the attached
screen shots) shows Age, Years of Education and Sex as a covariates. Why is this happening?
Thanks in advance
Shary
--
Volkmar Glauche
-
Department of Neurology [log in to unmask]
Universitaetsklinikum Freiburg Phone 49(0)761-270-5331
Breisacher Str. 64 Fax 49(0)761-270-5416
79106 Freiburg http://fbi.uniklinik-freiburg.de/
|