Thanks Joe.
One last confusion. How would I model multiple subjects? Do I include them all
in a the design matrix and contrast all conditions and all subjects? This
would seem to allow large activations in one individual to scew the data.
Analyses I have done in the past used either random effects or conjunctions
for the group data.
Joe Devlin wrote:
> Hi Paul,
>
> >My study was designed using a common baseline for all conditions. but was
> designed
> >as follows:
> >ARBRCRBRCRAR etc.
> >Each on was followed by an off. The active conditions were randomly
> chosen.Since
> >this is slightly different from what you described can you comment.
>
> This is essentially the same as the second case where you'd need to use
> inclusive
> masking on a main effect. If you model the baseline implicitly so the
> design matrix
> has 3 condition (A, B, and C), then you can use a contrast of [1 1 1] and
> mask it with
> [1 0 0], [0 1 0], and [0 0 1] inclusively at p<0.05. If you want a more
> conservative
> estimate, decrease the masking level (ie p<0.001).
>
> Hope this makes sense.
>
> BTW, I see I didn't cc: my reply to the list. I think it'd be a good idea
> to get feedback
> from others on this too in case they find any flaws in this approach or
> have more clever
> ideas.
>
> Best of luck,
> Joe
>
> - Joe
|