Abstract
Researchers have long wondered whether marital-status differences in mortality arise largely from selection mechanisms or from causal processes typically known as marriage protection. Unfortunately, many investigators have relied on aggregate patterns of mortality differentials—such as age schedules of excess mortality in the single population or the relationship between the level of excess mortality and the relative size of the single population—to make inferences about the relative importance of selection and causal processes. In this paper, a simple mathematical simulation model is used to demonstrate that many inferences derived from observed patterns are simply not justified. This finding highlights the importance of prospective data for assessing the relative importance of selection and causal factors in accounting for the excess mortality of the unmarried.
