A partial posterior <i>p</i> value test for multilevel mediation

A variety of inferential tests are available for single and multilevel mediation but most come with notable limitations that balance tradeoffs between power Type I error. We extend the partial posterior p value method (p3 method) to test mediation. This contemporary resampling-based composite approach is specifically suited complex null hypotheses. develop p3 investigate its performance within context two-level cluster-randomized studies. Similar in single-level studies, we found performed well relative other suggesting it provides a judicious error rate power. While bias-corrected bootstrapping achieved best overall performance, serves as an alternative tool researchers investigating especially useful when conducting priori analyses. To encourage utilization, provide R code implementing method.