Print

Print
Hi Mark,

You need one extra EV for each subject that is to be excluded. In this example, there would be 2 EVs coding group membership, and 3 EVs coding the excluded subjects, one per subject. The overall number of EVs would be 5. Since here you have 9 subjects total, this design would have 4 degrees of freedom.

Suppose you had excluded by hand the volumes from the 4D file. Then there would be 6 subjects. The 2 EVs coding group membership would stay, and this other design would likewise have 4 degrees of freedom.

Using a single EV for all subjects isn't correct in that the regression coefficient for that 3rd EV would yield an average value for the "excluded" subjects (that wouldn't actually have been excluded, but rather constituted a 3rd group), and the residuals for these subjects would not be equal to zero, thus affecting the estimate of the variance. Plus, the degrees of freedom would be 9-3 = 6, which is incorrect as can be seen from the case in which you had excluded the subjects by hand (that has df = 4).

All the best,

Anderson


On 5 February 2018 at 13:35, Mark Wagshul <[log in to unmask]> wrote:
Dear FSL experts,

I found a post from Stephen from a number of years ago in which he suggested that you can exclude a subject from analysis, while still leaving the subject in the data file.  This would be done by adding an EV in which all other subjects are labeled 0 and the subject to be excluded is labeled 1.  In the group label EV's, this subject would labeled 0 for both groups (my analysis is a group comparison, with two groups, controls and patients).

I just wanted to clarify something from the post.  He said in the post that you would add an EV for EACH subject to be excluded.  Is this in fact the only way to do this, or can I simply add a single EV for all of the subjects to be excluded, and they would be labeled as 1 in this extra EV?

So, for example, if I wanted to exclude subjects #3, 5 and 7 (with subjects 1,2,4 in group 1 and subjects 6,8 and 9 in group 2), the design and contrast matrices would look like this:

Design matrix
1 0 0  (Group 1)
1 0 0  (Group 1)
0 0 1  (Exclude)
1 0 0  (Group 1)
0 0 1  (Exclude)
0 1 0  (Group 2)
0 0 1  (Exclude)
0 1 0  (Group 2)
0 1 0  (Group 2)

Contrast:
1 -1 0
-1 1 0

Thanks for the clarification.  Best,

Mark Wagshul