Sex is a nominal variable but can be coded as 0,1. 

If it's coded as 0 (male), 1 (female); then the slope means, independent of Diagnosis, how much thicker the brain the female (vs. male) is.

But if I model Sex as a nominal variable, there is always an interaction term between Diagnosis and Sex in qdec, whether DODS or DOSS.

I would prefer not modeling the interaction term, but I don't know how to do so in qdec or mri_glmfit.

Daniel


-- 
Yung-Jui "Daniel" Yang, PhD
Postdoctoral Researcher
Yale Child Study Center
New Haven, CT
(203) 737-5454

On 9/5/13 11:27 AM, "Douglas N Greve" <greve@nmr.mgh.harvard.edu> wrote:


what does a slope of Sex mean? Is Sex a continuous variable?


On 09/05/2013 11:03 AM, Yang, Daniel wrote:
Hi Doug,

So I have two variables: Diagnosis (patient vs. control) and Sex
(female vs. male).

My model would be:

cortical thickness = intercept + (slope1 * Diagnosis) + (slope2 * Sex).

Thanks!
Daniel


--
Yung-Jui "Daniel" Yang, PhD
Postdoctoral Researcher
Yale Child Study Center
New Haven, CT
(203) 737-5454

On 9/4/13 10:10 PM, "Douglas Greve" <greve@nmr.mgh.harvard.edu


     I'm not sure you can do that with a linear model. What would your
     model look like?
     doug



     On 9/4/13 9:18 PM, Yang, Daniel wrote:
     Hi Freesurfer,

     Is it possible to specify only two main effects but no
     interaction effect by using mri_glmfit?

     If yes, could you provide an example?

     Details are much appreciated!

     Thanks!
     Daniel

     --
     Yung-Jui "Daniel" Yang, PhD
     Postdoctoral Researcher
     Yale Child Study Center
     New Haven, CT
     (203) 737-5454

     On 9/3/13 8:45 PM, "Yang, Daniel" <yung-jui.yang@yale.edu

         Dear FS Experts,

         My model has two categorical variable: Diagnosis and Sex. I
         am interested in the effect of Diagnosis but I want to
         control the Sex effect. Toward this goal, I want to model two
         main effects and want to avoid modeling their interaction
         term. Is this something feasible in FreeSurfer group
         analysis? Would using DOSS help with the solution?

         Thanks!
         Daniel
         _______________________________________________
         Freesurfer mailing list


         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.




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


--
Douglas N. Greve, Ph.D.
MGH-NMR Center
greve@nmr.mgh.harvard.edu
Phone Number: 617-724-2358
Fax: 617-726-7422

Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting
FileDrop: https://gate.nmr.mgh.harvard.edu/filedrop2
www.nmr.mgh.harvard.edu/facility/filedrop/index.html
Outgoing: ftp://surfer.nmr.mgh.harvard.edu/transfer/outgoing/flat/greve/