(A) Null and alternative hypotheses (two-sided): H0: RR = 1 vs. H1: RR not = 1
(B) Set alpha level (e.g., a = .05)
(C) Test statistic is Chi-square(1, N = xx) = xx.xx, p = probability region in right tail of a chi-square probability function with 1 degree of freedom.
* Use Fisher's test when an expected value is less than 5.
** Can also use 95% confidence interval as "test statistic" at a = .05.
(D) Reject H0 when p < a