dear volkmar.
i read your multiple answers about the opportunity in spm5s multiple
regression design specification to enter covariates. i am facing an
similiar problem and would like to ask you (or someone else in the list)
whether you could help me.
on the second level i want to answer the question, which brain regions
correlate with the amount of pain (main regressor) in my two groups of
participants (patients and healthy controls, i don't want to answer
possible group differences right now, maybe afterwards. any help for this
would be appreciated).
in spm2 i would perform a simple regression and enter the pain score as
covariate. the same is valid for spm5, i guess. i did understand that the
right place to enter would be in the regression section.
now i want to integrate the amount of anxiety prior the study as an
influential covariate, to control for the influence of anxiety on the brain
activation. my first thought was, to enter this in the covariate part of
the design specification. what i got, was a model with two covariates. this
doesn't look right to me, because both variables (pain and anxiety) seems
to be regarded as equal and not one as the one which i want to ask the
correlation and the other as the variable i want to control for. am i
correct? what would be my mistake? is there a better way to answer my
question?
thanks a lot in advance
markus
|