Dear Sarah,
if you carry out a regression on the psychopathy scores at the second
level, the intercept of the regression may be used as a one-sample t
test. More precisely, it is the one-sample t test with an added
covariate.
Best wishes,
Roberto Viviani
Dept. of Psychiatry,
University of Ulm, Germany
Quoting "<Sarah> <Gregory>" <[log in to unmask]>:
> Dear All
>
> I am looking at structural MR data for violent offenders and healthy
> controls. All have been
> scored on a dimensional scale, measuring psychopathy levels. I want
> to include the scores
> as a co-variate of interest, to see how volumes correlate with
> psychopathy scores.
>
> Would this be done using a one-sample t-test with all scans in one
> group but with a co-
> variate including the scores as vectors? Or would you still need to
> do a group comparison?
> What would the contrast look like!!!???
>
> Many Thanks
>
> Sarah
>
|