Dear Gelareh,
> I am trying to desing a GLM to show areas in the white matter of the
> the barin which have progressive decrease in FA during one year. To
> make it simpler I explain this for 2 subjects. Data are for 2
> subjects for 5 timepoints( 0,1,3,6,12 month) during one year. I
> designed below GLM:
>
> Group EV1 EV2 EV2
> Inteceps Subject1
> Subject2 Contrast
> EV1 EV2 Ev3
> 1 1 1
> 0 Subject
> 1 0 1 0
> 1 1 1
> 0
> Subject2 0 1 0
> 1 1 1 0
> 1 1 1 0
> 1 1 1 0
> 1 1 0 1
> 1 1 0 1
> 1 1 0 1
> 1 1 0 1
> 1 1 0 1
if I understand you correctly you are trying to fit a straight line to
the longitudinal data of each of your two subjects to see if these
lines are different. These can be different in two ways, they can have
different intercepts (i.e., they were different already at the start
of your scanning) or they can have different slopes, i.e. they evolve
differently over the period of scanning.
I am guessing from your naming that EV1 is the intercept, but since
you have modeled a single intercept you are forcing it to be identical
for the two subjects. I am further guessing EV2 and EV3 are meant to
model the slope for subjects 1 and 2 respectively, but since they are
all ones you are just modeling the mean (over time) of those scans
(which by the way is the same as individual intercepts).
I think the model you want is
1 1 0 -4.4 0
1 1 0 -3.4 0
1 1 0 -1.4 0
1 1 0 1.6 0
1 1 0 7.6 0
1 0 1 0 -4.4
1 0 1 0 -3.4
1 0 1 0 -1.4
1 0 1 0 1.6
1 0 1 0 7.6
where columns 2 and 3 are the intercepts and colums 4 and 5 are the
slopes where it has been assumed that changes are linear over
"calender time".
Good luck Jesper
|