Dear FSL experts,
I have processed DTI data using TBSS, and am now up to creating my design in Glm. I would like to look at associations between FA/MD and education, controlling for sex and age in a SINGLE group of participants. I wonder what design is the preferable one, A or B (below), given I am not interested in the relationship between group mean and FA/MD?
A)
EV1:Education (demeaned) EV2:Age(demeaned) EV3:Sex(demeaned)
Input1 EV1data EV2data EV3data
B)
EV1 EV2:Education(demeaned) EV3:Age(demeaned) EV4:Sex(demeaned)
Input1 1 EV2data EV3data EV4data
My contrasts of interest are:
Pos Educ 1 0 0
Neg Educ -1 0 0
Pos Age 0 1 0
Neg Age 0 -1 0
Pos Sex 0 0 1
Neg Sex 0 0 -1
These values will come under EV1, EV2, and EV3 respectively in scenario A; and EV2, EV3, and EV4 respectively in scenario B.
Hopefully this makes sense! Any help would be greatly appreciated.
Kind regards,
Louise Lavrencic
|