Dear FSL users,
I have 2 groups (patients and HCs) and I'd like to run a simple 2 sample t-test with more than one (let's say 3, age, sex and QI) nuisance variable.
After running all the tbss scripts, I've created the design .mat and .con files with the design_ttest2 command.
With a text editor, I've added THREE mean-centred columns in the .mat, under Matrix, one for each nuisance variable.
I've also changed NumWaves to 5, and added three values to PPheights.
In the .con file, I've added THREE 0 to the end of each contrast, changed NumWaves to 5, and added three values to PPheights.
Now my question is: if I want to look for differences between patients and HCs without being misled by age (or sex or QI), should I also change the NumContrasts field in the .con file?
Thanks.
Regards,
Elisa