Question

Telephone pollsters want to do a survey to get opinions on the following question: “Does the performance of a colleges’ football team have an impact on the public’s perception of the prestige of the university?” Suppose you were designing the survey and wish to estimate the difference in the proportion of faculty and students that answer “yes” to this question, with a 90% confidence interval that is no wider than 0.10. If you plan to select 10 times as many students as faculty, how many faculty and students should be included in the sample? Note: the pollsters don’t have time to do a pilot study and have not seen a study on this previously published so they want you to give them the maximum numbers they would need.

Answer 1