Hi FSL group,
I'm starting to analyze DTI results (FA values) in SPSS from an exploratory analysis looking at tractography of all the pairwise connections in an eye-movement network that consists of 16 ROIs. There are 120 possible pairwise tracks to analyze. I have 37 subjects with data. However, there are some missing values for some of the subjects for some of the tracks of the ROI pairs.
My question is, what is the usual protocol in dealing with missing data? Is a method such as imputation (replacing the missing data with the average of the existing values or replacing it with the expected value) a valid approach to fill in the missing data or would you only recommend looking at tracks in which everyone has a value? I would be greatly appreciative of your advice!!!
Best,
Dana
|