I have a design in which subjects were scanned under two conditions
(separate days): A and B. These conditions were counter-balanced among
subjects. I am running a paired t-test to determine differences between
the two conditions and was wondering how I would add a covariate (or if
it's even necessary) to distinguish which condition order each subject
received, i.e. condition A first or condition B first.
On a similar note, we have a subjective measure that was taken before each
run and are interested in determining whether the change in this measure
between condition A and B correlates with change in brain activation when
comparing condition A and B. Is a paired t-test still the best test to use,
and how do I put in a covariate that has one value per subject (change
score of A-B), when I have two contrast images per subject in my design?
I appreciate any insight.
Best regards,
Rachel
|