Hello,
We would like to run correlational analysis on TBSS data within a single group (healthy controls), assessing the correlation between FA and age, and FA and a behavioral measure, using the randomize tool. I would like to check whether we are setting up our design matrices correctly (the .mat and .con files).
Reading through some of the previous posts on the FSL Support Forum, it seems that for a correlational analysis with age, we could include a single column in the .mat file, listing individual participants’ “demeaned“ ages (under the “/Matrix” section of the .mat file initially created through FEAT GUI for Single-Group Average), and in the con. file, we would include:
“/Matrix
1
-1”
[to assess for positive or negative correlations respectively].
To “demean” the age, we would subtract the the average age (of all participants) from the age of each participant.
Next, to assess for correlations between FA and our behavioral measure (while controlling for age), we could add the “demeaned” behavioral data in a second column in the .mat file, and use the following contrasts within the .con file:
“/Matrix
0 1
0 -1”
[with the first contrast assessing for positive and the second—for negative correlations between FA and the behavioral measure, while controlling for age].
Please let me know whether this would be the correct way to set up the design matrices, and if you have any suggestions.
Thank you very much for your help!
Petya
########################################################################
To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1
|