Sample size determination for testing nonzerodifference of two proportions in matched-pair design

Two approximate sample size formulae are proposed for testing the null hypothesis of nonzero rate difference of two proportions in matched pair design based onTango’s score test statistic.The formulae can be used to produce sample size estimates that guarantees a prespecified power of a hypothesis test at a certain significance level and controls the width of a confidence interval with a certain confidence level.Our empirical studies confirm that the proposed sample size formulae perform satisfactorily.A real example is used to illustrate our methods.