Makes sense, thanks. So would the following designs below get what I
expect?
A group test, and take 2 scores into account my design would look like this:
Group EV1 score1 score2
1 1 10 50
1 1 20 90
1 -1 10 200
1 -1 2 5
1 1 4 5
1 -1 30 8
Contrast = 1 0 0
Or a test between 3 groups may be:
Group EV
1 1
1 1
1 2
1 2
1 3
1 3
Contrast = 1
-----Original Message-----
From: FSL - FMRIB's Software Library [mailto:[log in to unmask]] On Behalf
Of Brian Patenaude
Sent: Wednesday, October 10, 2007 4:15 AM
To: [log in to unmask]
Subject: Re: [FSL] first_utils
Hi,
The restriction is that no linear combination of the columns sum to a column
of ones. first_utils demeans the design matrix such that the first column of
the design matrix is all ones. This is why splitting the groups into into
multiple columns causes an error. first_utils will demean all regressors
internally.
Cheers,
Brian
> ok, the preliminary problem .. i needed to use 1 EV with 1 or -1 for
> the group difference.
>
> apparently the group difference needed to be in 1 ev and not multiple
> evs, but how would you add an additional regressor if you can only
> use
> 1 EV?
|