Ran a within-subjects pilot in today’s 10 a.m. lab meeting comparing laughs for slide jokes labeled “p <.05” versus unlabeled controls (counterbalanced; N=32), and the mixed-effects model yielded a non-significant fixed effect (β=0.08, SE=0.11, p=.47; observed power ≈.14). Has anyone elicited a significant chuckle without post hoc fishing, or do we accept the null and preregister silence?
.