Write an essay explaining how a random sample of a fairly small number n of people can indeed reliably measure the proportions p and 1-p pro or con on a given issue of a much larger population. Explain why, the smaller p or 1-p are, the larger the sample size n must be (both to ensure that the distribution of samples is normal, and to obtain a given margin of error) but why (other factors being the same) a bigger population doesn't mean having to take a bigger sample size n. Your essay should give the reader a feel for how big n must be (so they understand that n = 1000 for a population of 100 million is very reasonable). Also be sure to explain what a margin of error of plus or minus 2% actually means and why a margin of error of plus or minus 2% doesn't guarantee that the sample measurement is necessarily within 2% of the actual value.