Suppose that a student wants to gather data to investigate whether husbands tend to be older, on average, than their wives. Th e student obtains a sample of 200 marriage licenses that report the ages of both the husband and wife. The student randomly selected 100 of the marriage licenses and recorded the husbands’ age from the license, and then the student recorded the wife’s age from the other 100 licenses.
a. Explain why a paired analysis would not be appropriate for the student’s data.
b. How could the student’s data collection method from the 200 marriage licenses be improved so the data would be paired?
c. Why is it reasonable to believe that pairing would be helpful in this situation?