A Table of Exact Sample Sizes for Use with Fisher's Exact Test for 2 x 2 Tables

We studied systematically the problem of sample size for Fisher's exact test of independence in 2 x 2 tables, assuming that neither of the two margins is fixed. In this case, we present tables to calculate sample sizes necessary to achieve 50% and 90% power with the significance level equal to 1% or 5%. These tables can be used to determine the sample size necessary to detect gametic-phase disequilibrium between two genetic loci in a population genetics survey. A one-sided test and two double-sided tests are considered corresponding to the two most commonly used interpretations of observed significance.