This paper proposes a sample size procedure for both equivalence and c
onventional tests for the comparison of two binomial proportions, base
d on the signed square root of the deviance. When the comparison is ba
sed on the odds ratio, I describe an alternate 'close' conditional exa
ct method that gives results that support those given by the deviance
method. I summarize the advantages of the deviance-based method and al
so show that in general equivalence situations the sample size estimat
e depends upon the measure of comparison selected, odds ratio, risk ra
tio or risk difference.