Hi Shady,

Try assembling the design as this:

EV1: Group 1 (coded as 0/1)
EV2: Group 2 (coded as 0/1)
EV3: Group 3 (coded as 0/1)
EV4: Sex (coded as +1/-1)
EV5: Site (coded as +1/-1)

The contrasts to compare the three groups are then:
C1: [1 -1 0 0 0]
C2: [1 0 -1 0 0]
C3: [0 1 -1 0 0]
C4: [-1 1 0 0 0]
C5: [-1 0 1 0 0]
C6: [0 -1 1 0 0]

From the description it seems site and group are confounded, although not completely, such that it should be possible to still draw some inferences, although without much power.

The above does not include terms for interactions. If it's something you could be interested in, then extra EVs can be added.

All the best,

Anderson


On 30 August 2016 at 11:40, Shady Rahayel <shady.rahayel@gmail.com> wrote:
Greetings,

I want to design a matrix that contains 3 fixed factors. The first is diagnosis (3 levels), the second is gender (2 levels), and the last is imaging site (2 levels). However, the thing is that the imaging site factor only applies to only one level of the diagnosis factor (i.e. patients from the first level of the diagnosis factor were seen in 2 different imaging sites, whereas for the two other groups of patients the same imaging site was used).

I have followed the tutorials but I still be am confused concerning how I should go about creating the design matrix with 3 fixed factors. Would it be:

GroupDescriptorFile 1
Title xx
Class Diagnosis1-Male-Site1
Class Diagnosis1-Female-Site1
Class Diagnosis1-Male-Site2
Class Diagnosis1-Female-Site2
Class Diagnosis2-Male-Site1
Class Diagnosis2-Female-Site1
Class Diagnosis3-Male-Site1
Class Diagnosis3-Female-Site1
Input    etc etc...

Afterwards, if I'm looking for a difference between Diagnosis1 and Diagnosis2, what would be the contrast?
0.25 0.25 0.25 0.25 -0.5 0.5 0 0 ?

Thank you very much for your help,

Shady

--
Shady Rahayel
Étudiant au Ph.D. en neuropsychologie R/I
Université du Québec à Montréal

Centre d'Études Avancées en Médecine du Sommeil
Hôpital du Sacré-C
oeur de Montréal

_______________________________________________
Freesurfer mailing list
Freesurfer@nmr.mgh.harvard.edu
https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer


The information in this e-mail is intended only for the person to whom it is
addressed. If you believe this e-mail was sent to you in error and the e-mail
contains patient information, please contact the Partners Compliance HelpLine at
http://www.partners.org/complianceline . If the e-mail was sent to you in error
but does not contain patient information, please contact the sender and properly
dispose of the e-mail.