riage and the presence of children. The NRC (2001a) found some evidence that being married with young children helped men but hurt women in terms of their academic career. The size of this effect had been shown to increase for men and to decrease for women. Xie and Shauman (2003) and Mason and Goulden (2002) found that marriage and family also negatively affected women pursuing science and engineering careers.
Toutkoushian (1998a:515) laid out an hypotheses as to why the effect of marital status on faculty salary might differ by gender: on the supply side, since women “often bear the majority of child-rearing responsibilities in American society, married women may be more likely than married men to interrupt or reduce their time allocation to their career,” or “married women may accept lower wages in order to find employment at the same institution as their spouses.” On the demand side, “institutions may make higher salary offers to married men than to married women on the premise that married men are typically the breadwinners of the family and thus have a greater need for higher salaries.” Using NSOPF:93 to analyze faculty salaries, Toutkoushian found that the return on marriage for men was statistically significant and positive, but there was no corresponding return for women.
Sax et al. (2002:426) focused on the role of family-related variables in research productivity. Specifically, they asked: “Do marriage, children, aging parents, and other family-related factors influence faculty research productivity?” and “Is the nature of family-related factors dependent on gender or tenure status?” They analyzed data from the 1998-1999 Higher Education Research Institute Faculty Survey. They found, first, that male faculty were more productive than women, when compared at increasing levels of output over 2 years, i.e., a greater percentage of women than men produced zero publications, while a greater percentage of men than women produced five or more publications. However, Sax et al. found that “family variables contributed little or nothing to the prediction of faculty research productivity. More important were professional variables such as academic rank, salary, orientation toward research, and desire for recognition” (p. 435). Sax et al., hypothesized the lack of effect may have resulted because women who had children were able to do more with their limited time and reduce their time in activities outside of work and home (i.e., leisure time).
Perna (2003a:2) used the NSOPF:99 “to examine the ways in which parental status, marital status, and the employment status of the spouse are related to two outcomes, tenure and promotion, among college and university faculty.” In an earlier study drawing on data from the NSOPF:93, Perna (2001c, cited in 2003a:3) “found that parental and marital status were related to employment status among junior faculty and that the relationships were different for women than for men. Men appeared to benefit from having children, as men with at least one child were less likely to hold a full-time, non-tenure track position than they were to hold a full-time, tenure track position.” In this study, Perna found