Dear fsl experts,
I have a question regarding TBSS analysis. I would like to compute the correlation (or regression) between FA and behavioral scores.
I have one group of subjects (15 subjects) and behavioral scores (BS) taken from 3 tasks plus 1 regressors of no interest, age that I have demeaned (is it correct?)
I assume that the design matrix should look like this:
/NumWaves 5
/NumPoints 15
/PPheights 1 1 1 1 1
/Matrix
1 1,0434 1,1634 1,9836 -4,7333
1 0,3511 0,6998 0,6660 -6,7333
1 0,3733 0,1620 1,3627 4,2667
1 -0,7298 1,1634 0,9047 -3,7333
1 -0,3459 -0,1429 0,6176 -1,7333
1 -0,0511 0,1898 -0,1555 -3,7333
1 -0,0954 -2,1662 0,0961 -4,7333
1 -0,7031 -1,0768 -0,6896 -5,7333
1 0,3342 0,4163 -0,7304 9,2667
1 0,8755 -0,1218 -0,4432 -2,7333
1 -1,3132 -1,0332 0,5826 6,2667
1 0,4781 -0,9344 0,2008 -0,7333
1 -0,0692 0,1785 -1,1205 18,2667
1 0,6799 1,1634 -1,0984 1,2667
1 -0,8272 0,3381 -2,1765 -4,7333
Is the following contrast matrix correct?
/NumWaves 5
/NumContrasts 6
/PPheights 1 1 1 1 1
/Matrix
0 1 0 0 0
0 -1 0 0 0
0 0 1 0 0
0 0 -1 0 0
0 0 0 1 0
0 0 0 -1 0
When I run the randomize command I receive an error message.....are the NumWaves/NumConntrast an PPWheights correct?
Any help will be greatly appreciated
Best, Sandra