A 1950s study found that married men were in better health than men of the same age who never married or were divorced, suggesting that the healthiest path is for a man to marry and never divorce [23].

a. Explain why there might be sampling bias in this study.

b. Suppose that marriage is generally bad for a man’s health. Explain how it could still be the case that men who marry and stay married are in better health than

(i) men who never marry and (ii) men who marry and then divorce.

c. What would have to be done for a controlled experiment?