I'm trying to estimate the sample size required to achieve 80% power at a 5% significance level for a superiority study comparing AUC from bioavailability data.

My challenge is the only previous pilot study I found reported mean AUC and standard errors for the raw data and using this with pwr.t.test in R gives the following error.

Error in uniroot(function(n) eval(p.body) - power, c(2 + 1e-10, 1e+09)) : 
  f() values at end points not of opposite sign

Below are means(SE) for the two groups from the previous study:

Group A: 6300(35.6); n=6

Group B: 16750(57.9); n=6

Here's the code I run:

pwr.t.test(n = NULL , d = 119.84, sig.level =0.05 , power = 0.8, type = "two.sample")

Log transformation should be the way to go I reckon, but I'm not sure how to derive the log SD from the raw SD. Any help will be much appreciated.


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.