Because sampling distributions of indirect effects are seldom normally distributed, the Macro Process uses non-parametric bootstrapping. (bmcpsychology.biomedcentral.com)
Worksheets on: a. the null and alternative hypothesis b. type I and type II errors c. checking assumptions d. sampling distributions e. the normal distribution f. the t - distribution g. interpreting p - values Answers for worksheets provided in separate document. (tes.com)
A proper sampling distribution could, in principle, be constructed which takes into account cherrypicking bias. (realclimate.org)