Abstract
This study builds on an earlier finding from the May 1980 Current Population Survey that one-third of full-time dual-earner couples with children in the United States include at least one spouse who works other than a regular day shift. Using the same data source, the relevance of husbands’ and wives’ job characteristics (occupation and industry) on their shift work status are considered, and the nature of the association between husbands’ and wives’ work shifts is explored. Four alternative models are initially posed and tested with log-linear analysis; these models vary in the extent to which a spouse’s shift work status is contingent upon the job characteristics of a husband or wife. None of these models fit the data. A modification of the simplest of the four models (using forward selection) is the best fitting model. It is not symmetrical: the wife’s shift is contingent upon both her occupation and industry and that of her husband, but the husband’s shift is contingent only upon his occupation and industry. There is a relationship between husband’s and wife’s shift, the direction of which depends upon the occupations of both spouses. This study demonstrates the importance of taking a “couple” perspective on shift work among married persons, and the need for models that include interaction effects.