A scientist looking into the effect of smoking on heart disease has chosen a large random sample of smokers and of nonsmokers. She plans to study these two groups for 5 years to see if the number of heart attacks among the members of the smokers' group is significantly greater than the number among the nonsmok- ers. Such a result, the scientist feels, should be strong evidence of an association between smoking and heart attacks. Given that
(a) older people are at greater risk of heart disease than are younger people; and
(b) as a group, smokers tend to be somewhat older than nonsmokers, would the scientist be justified in her conclusion? Explain how the experi- mental design can be improved so that meaningful conclusions can be drawn.