I'm trying to calculate the minimum detectable effect in an experiment after n samples.

I'm able to use the pwr package like this to compute the minimum detectable effect.


pwr.2p.test(n = 10000, sig.level = 0.05, power = 0.8, 
    alternative = "two.sided")

pwr.2p.test(n = 50000, sig.level = 0.05, power = 0.8, 
    alternative = "two.sided")

However, this returns the MDE in terms of h.

 Difference of proportion power calculation for binomial distribution (arcsine transformation) 

          h = 0.0177296
          n = 50000
  sig.level = 0.05
      power = 0.8
alternative = two.sided

The base rate in the control group of the experiment is 0.10 or 10%. How can I convert the h MDE to a percent difference in the rates between the groups?


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.