Why not! Just imagine the faces of the reviewers that have to read something like "We tested group differences using a t-test. For (that) extra robustness, we bootstrapped the test 1000 times. The average p-value across simulations was less than 0.05 (p<0.05)." Cursed.
I am worried this might affect time in the opposite direction. We might emerge from the eighth level of significance, believing we had been working on our multidimensional bootstraps for an afternoon - only to find that centuries had passed
Comments