Open derekbeaton opened 5 years ago
(1) I think I know the solution, and (2) the correlations should change over to abs() or cor()^2 regardless... the direction is irrelevant (and actually problematic).
because I'm terrible at comments, this is w.r.t. split-half/two-fold repeated approaches
I'm not so sure those are correct anymore... because of arbitrary flips.