A pollster wants to estimate the difference between the proportions of men and women who favor a particular national candidate, using a 99%...

A pollster wants to estimate the difference between the proportions of men and women who favor a particular national candidate, using a 99%   confidence interval of width 0.05. Suppose the pollster has no prior information about the proportions. If equal numbers of men and women are to be polled, how large should the sample sizes be?