Question: Compute the chi-square statistic, and assess the statistical significance for the relationship between bird ownership and lung cancer, based on the data in Exercise. State a conclusion about the relationship.

Exercise: In a study in Berlin, reported by Kohlmeier et al. (1992) and by Hand et al. (1994), researchers asked 239 lung-cancer patients and 429 controls (matched to the lung-cancer cases by age and sex) whether or not they had kept a pet bird during adulthood. Of the 239 lung-cancer cases, 98 said yes. Of the 429 controls, 101 said yes.

a. Construct a contingency table for the data.

b. Compute the risk of lung cancer for bird owners and for those who had never kept a bird.

c. Can the risks computed in part (b) be used as baseline risks for the populations of those who have and have not owned birds? Explain.

d. What is the relative risk of lung cancer for bird owners?

e. What additional information about the risk of lung cancer would you want before you made a decision about whether or not to own a pet bird?

f. Can these data be used to establish a causal connection between owning a bird and the risk of lung cancer? Explain.

g. For any study of lung cancer, the effect of smoking should be considered. How would you determine whether smoking might be a confounding variable in this study?