You want to estimate the difference between two population proportions using an 80% confidence interval. You need to take samples that are large enough, so that the width of the confidence interval...

You want to estimate the difference between two population proportions using an 80% confidence interval. You need to take samples that are large enough, so that the width of the confidence interval does not exceed 5.0%. You wish to take two samples (one from each population) of the same size n. The populations have not been sampled before, and therefore you have no information about what the two proportions may be.

Using the most conservative estimates for the two population proportions, what is the minimum sample size you will need (for both of the samples)? (Round the result up to the nearest integer.)