There is a typo in the SAMPSIZEPWR documentation within Statistics Toolbox 6.0 (R2007a). The second example text reads:
"Compute the sample size n required to distinguish p =0.23 from p = 0.2 with a binomial test. The result is approximate, so make a plot to see if any smaller n values also have the required power of 0.5 napprox = sampsizepwr('p',.2,.26,.6)"
The input parameters do not match the values in the text.
Best Answer