I have produced two P-value histograms for μ=-0.5, one for a two-sided test and another with a one-sided test. I notice that for the first case (two-sided test) I get a righ
