Women in Science: Career Processes and Outcomes (review)

The trend toward equality of educational and occupational attainment between women and men in the U.S. continues. However, buried beneath the aggregate statistics are what Yu Xie and Kimberlee Shauman refer to as "stubborn exceptions," particularly persistent gender inequalities in doctoral degrees in mathematics and some of the sciences and in the maintenance of scientific careers after the Ph.D. in these fields. By 2000 women accounted for nearly half of doctorates across academic fields; and while their representation in some sciences approximated this average (e.g., biology and biochemistry), their representation persisted as among the lowest in some engineering fields, physics, and mathematics. Across the sciences, women's career persistence and mobility after the degree fall well below their doctoral attainment levels.

The explanations for these persistent trends have eluded previous researchers, who have resorted to an intuitive metaphor — the leaking science pipeline. The pipeline metaphor is predicated on the assumption that the scientific career begins in middle and secondary school science and mathematics classes and persists in a necessary sequence of educational career transitions beginning with intentions to major in math/science in college and followed in order by majoring and graduating in math/science, attending graduate school, receiving the masters and Ph.D., attaining postdoctoral placements, and progressing through formal academic careers as scientist/professor. The leaking pipeline portrays the cumulative loss of women along the way without specifying the mechanisms that propel the loss.

Xie and Shauman offer a set of empirically derived explanations for this attrition and findings that contradict the pipeline metaphor. They do so by following a life course perspective and using rigorous multivariate demographic methods on microdata from multiple (17) longitudinal and census sources to predict gender differences. The life course perspective proposes that life transitions are interdependent across education, family, and work domains and that later transitions are contingent on (but not determined by) earlier transitions. Hence, the science pipeline does not operate in a social vacuum. To track this multidimensional and dynamic process, Xie and Shaumann concatenate a set of limited longitudinal datasets that permit the construction of synthetic cohorts that can be followed from middle school to postdegree career years.

They begin in the middle- and secondary school years by using 6 datasets to compare gender patterns of mathematical and science performance (National Longitudinal Study of the Class of 72; High School and Beyond Senior and Sophomore Cohorts, respectively; National Educational Longitudinal Study of 1988; and the Longitudinal Studies of American Youth, cohorts 1 and 2). In chapter 2, they find that gender differences in mathematical ability are minimal except at the upper extreme of the distribution, but that male students participate more in scientific curricula. In chapter 3, they find large gender differences among high school seniors in expectations to major in science and engineering in college, by a ratio of 2 males to 1 female. However, in chapter 4 they find that after entering college women are more likely than men to enter a science and engineering major after starting a nonscience major.

Chapters 5 and 6 track post-B.A. and M.A. degree career paths (using the Baccalaureate and Beyond Longitudinal Study and the New Entrants Surveys). Here, as life course theory would predict, things get more complicated. First, while women are more likely to major in some biological sciences, the majors in these fields are less likely overall than other science majors (e.g. engineering and physics) to pursue science and engineering careers. And, controlling for major, women are 25% as likely to work in science and engineering careers. Finally, all else equal, married women, and those with children, are less likely to continue science and engineering careers. Hence, gender segregation by major (biology versus engineering/physics) and familial roles hinder women's career progression.

Chapters 7-10 employ microdata from the census (PUMS 1960-90) and five other datasets to examine four post-degree career patterns: employment, geographic mobility, research productivity, and the status...