Dear FSL community,
I have a statistical question again, regarding inclusion of one and the same subject at different time points in an otherwise cross-sectional analysis.
I have measures from subjects at 4 different timepoints. Due to quality reasons and some missing data points for some subjects, we decided to take only 1 scan per subject but chose the set of scans to be roughly equally distributed over all 4 timepoints.
We then calculated simple two-sample unpaired t-tests with nuisance variables (age and gender) and simple regression analysis because the subjects can be devided into 2 groups and can be correlated to a behavioural measure.
I now wondered, if I can improve my statistical power by including additional scans from subjects I already included in the analysis but at different timepoints? Or is this not good practice?
Would I need an exchangeability block-matrix in this case?
And would I have to indicate the same subject with the same number in this matrix, or rather all the subjects from the same timepoint?
Thanks for your oppinions and help
Franziska
########################################################################
To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1
|