I'd like to use Proc Power to estimate a required sample size as follows: Outcome is binary (Bernoulli trial). Expected population success rate is 0.025. What sample size will produce a 95% confidence interval for success rate with a width of 0.036 (half width = 0.018)?
Question: Can this be done with Proc Power and if so how? The CI method (exact, Wald, etc.) isn't important at this point.
ANOVA, or Analysis Of Variance, is used to compare the averages or means of two or more populations to better understand how they differ. Watch this tutorial for more.