Perils of single life and benefits of marriage

Abstract It has long been observed that married persons experience more favorable mortality than unmarried persons. Hypotheses about possible explanations for mortality differentials by marital status fall into three general categories: (1) hypotheses regarding errors in the data, (2) hypotheses related to the selection of persons into and out of marital states, and (3) hypotheses regarding environmental and behavioral factors associated with various marital states. In this paper the authors investigate the extent to which mortality differences by marital status are similar across populations and assess the importance of selection factors in producing the excess mortality of the single populations.