When describing the balance on covariates should we look at unidimensional
imbalance or at the difference in means? Can we just look at one of
these values to check balance or should we compare them to something (ie the
range)?
Thanks
Show replies by date