Dear Alfonso and CONN Community
In advance excuse me if my questions are a little bit blunt I m new to the field.
I am currently using weighted GLM correlation measures within task condition to map the ROI to ROI effects of task performance. The HCP dataset that I am using consists around 200 MZ and 200 DZ twins out of 1200 subjects and I am wondering how to control for this statistically independent subjects since they are sharing exactly the same genes.
As an approach to this problem I was wondering whether if it would be appropriate to assign a number for each twin pair and use it as a covariate:
TwinPair1 sub1 1
TwinPair1 sub2 1
TwinPair2 sub1 2
TwinPair2 sub2 2
...
whereas second level modeling would be like:
All Subjects - Age - Gender - Education - Handedness - [b]TwinPairs[/b] - Relational Task [0 0 0 0 0 [b]0[/b] 1]
Regards
Erkam
In advance excuse me if my questions are a little bit blunt I m new to the field.
I am currently using weighted GLM correlation measures within task condition to map the ROI to ROI effects of task performance. The HCP dataset that I am using consists around 200 MZ and 200 DZ twins out of 1200 subjects and I am wondering how to control for this statistically independent subjects since they are sharing exactly the same genes.
As an approach to this problem I was wondering whether if it would be appropriate to assign a number for each twin pair and use it as a covariate:
TwinPair1 sub1 1
TwinPair1 sub2 1
TwinPair2 sub1 2
TwinPair2 sub2 2
...
whereas second level modeling would be like:
All Subjects - Age - Gender - Education - Handedness - [b]TwinPairs[/b] - Relational Task [0 0 0 0 0 [b]0[/b] 1]
Regards
Erkam