Question:

a) Compare and contrast stratified sampling and cluster sampling. Under what circumstances would we choose these techniques over a simple random sample?

b) A corporation has 30 outlets in the Northeast, 40 in the South, 60 in the Midwest and 45 in the West. Using a proportional allocation of a random sample of 20% of the outlets, the manager records the following proportion of stores expecting sales to increase over the next year.

Estimate the population proportion. Develop a 90% confidence interval for the population proportion?

c). A sales representative has 398 accounts. From a random sample of 56 of these accounts, she finds that 27 had placed an order within the last 30 days. Estimate the 90% confidence interval for the proportion of all accounts that placed an order within the last 30 days.

d). Explain why non-response error may bias the results of a survey