Refer to the Proceedings of the 5th AMC Workshop on Computers & Communication Security (October 2012) study of the vulnerability of relying party (RP) Web sites, Exercise 8.59 (p. 465). Recall that the Web sites were categorized as server-flow or client-flow Web sites. The researchers want to estimate the true difference between the proportion of server-flow Web sites vulnerable to attack and the corresponding proportion of client-flow Web sites with a 95% confidence interval. They want to know how many Web sites of each type they need to sample to obtain an estimate that is no more than .15 from the true difference.
a. Identify the parameter of interest for this study.
b. What is the desired confidence level?
c. What is the desired sampling error?
d. Find the sample size required to obtain the desired estimate. Assume n1 = n2.
e. Repeat part d, but now assume that twice as many Web sites will be sampled from the population of server-flow Web sites as from the population of client-flow Web sites, i.e., assume n1 = 2n2.