In some clinical trials, the measure for the primary endpoint is repeated occurrences of the same or related types of event. Poisson regression, Anderson-Gill(AG), and the negative binomial(NB) model are commonly used methods for analyzing such data. Signorini (1991) proposed a method to estimate sample size for Poisson regression, but the adequacy of this estimation for analyses using the AG and NB models have not been reported. We assess the operating features for detecting treatment differences using these statistical methods with the sample size estimated by Signorini's method. Based on our simulations, Signorini's method is also appropriate in the sample size estimation based on NB and AG models. Furthermore, we find only small discordant results among these three analyses. The specific analysis method rarely impacts the conclusion drawn from a study.