Hi SAS Community! The aim id to calculate the sample size for average bioequivalence trial. I would like to replicate the below example from "Sample Size Calculations in Clinical Research" by Chow and Shao (3rd edition).  Below you can see the formula that is used as well.     So, with the SD=0.4, delta = 0.05, limit = 0.223, alpha = 0.05 and 80% power I am using the below code:   proc power; twosamplemeans test=equiv_diff DIST=NORMAL lower = -0.223 upper = 0.223 meandiff = 0.05 stddev = 0.4 npergroup = . power = 0.8; run;   But I am getting 69 subjects required, not 21 (as in the book, or 24 from the table approximation)   Is there a different formula SAS is using? Or should I use a different procedure? Sorry if I am missing something, just trying to get my head around that.   Thank you Guys, Agnieszka     ... View more