Go to www.surveysystem.com/sscalc.htm and take a look at their sample size, sampling error, level of confidence calculator.
a. If you have a problem where the population size is approximately 1 million and you want a sampling error of plus or minus 5 percent and a 95 percent level of confidence that the true population value will be within the sampling error specified, how big of a sample would you need?
b. What sample size if you wanted to have 99.5 percent confidence?

c. What sample size would you need with the specifications in (a) except you change the acceptable error to plus or minus 2.5%?

d. What do these calculations tell you about the nature of the relationship between sample size needed and sampling error? Same question for level of confidence?