Problem
You are catching fish in a pond and measuring the length of fish. You want to know the mean length of the population of fish, but you also know you can never be sure you have caught and measured all the fish, so you decide to use confidence intervals. You catch fish 10 at a time, measure their lengths. and create 90% confidence intervals each time. Assume you have data from 100 samples of fish taken from the pond and have calculated the 90% confidence interval for each of these samples.
i. Why will you have different confidence intervals for each of your trials?
ii. How many of those 100 confidence intervals will contain the true population mean? Why?