Hi,

My main analysis is on an intention to treat dataset, although I am looking at the per-protocol dataset to confirm if there were any differences. When running Cronbach's Alpha on my many scales, should I run it on the ITT dataset, the PP data, or both?

Thanks,

Max