Dear Paul,
This looks correct to me - the design matrix, the contrasts and the interpretation of the contrasts.
All the best,
Mark
On 2 May 2014, at 03:27, Paul Chou <[log in to unmask]> wrote:
> Dear all FSL experts
>
> I have a question about how to construct suitable GLM design matrix and contrast for testing the relationship between regional gray matter volume and categorical variables? My experimental design as follows:
>
> Research goal: To identify which brain region is correlated to specific categorical variables uniquely ? I have 5 clinical factors of each participants. Each clinical factor is "1" (indicated presence) and "0" (indicated absence).
>
> Could I construct the design matrix as follows : (Using 6 participants as an example)
>
> (Clinical factor 1) (Clinical factor 2) (Clinical factor 3) (Clinical factor 4) (Clinical factor 5) (Constant)
> 1 1 0 0 0 1
> 1 0 0 0 0 1
> 1 0 1 0 0 1
> 0 1 1 1 0 1
> 0 0 0 0 1 1
> 0 0 0 0 0 1
>
> and using the following contrast to testing which brain region is correlated with each clinical factors ?
>
> contrast 1: 1 0 0 0 0
> contrast 2: -1 0 0 0 0
> contrast 3: 0 1 0 0 0
> contrast 4: 0 0 1 0 0
> contrast 5: 0 0 0 1 0
> contrast 6: 0 0 0 0 1
>
> could I interpret the result as follows:
>
> Contrast 1: 1 0 0 0 0 (The gray matter volume of participants with presence clinical factor 1 is larger than the participants with absence clinical factor 1 after controlling clinical factor 2 3 4 5)
>
> Contrast 2: -1 0 0 0 0 (The gray matter volume of participants with presence clinical factor 1 is smaller than the participants with absence clinical factor 1 after controlling clinical factor 2 3 4 5)
>
> Best
>
> Paul
|