Stephen Morgan Kim Weeden Cornell University
How do characteristics of classrooms and schools in the eighth, tenth, and twelfth grades affect later entry into science careers? How do young men and women envision the process of entering science careers, and how do their views shape their behavior while in middle school and high school? Analyzing survey data from the National Education Longitudinal Study (1988-2000) and the Education Longitudinal Study (2002-2006), the investigators will conduct three related projects. First, they will compare the high school graduating classes of 1992 and 2004, considering the differences across these two cohorts in the science courses they take in high school, in subsequent college entry, and in the selection of a science major in college. Second, using data on the high school graduating class of 1992, the investigators will model gender differences in the effects of engagement with science in the eighth grade and in high school on later entry into science careers, as of 2000. Third, using data on the high school graduating class of 2004, the investigators will model changes in the specificity and content of young men and women?s occupational plans, using heretofore unanalyzed verbatim responses about plans in the sophomore year of high school, the senior year of high school, and two-years following high school graduation. This project will assess (a) how plans for science careers coalesce and change as students develop more sophisticated understandings of the division of labor within science, the educational requirements of different science careers, and the costs and benefits of different science careers; (b) whether there are gender-specific trajectories in occupational plans, and (c) how these trajectories are mediated by family context or the high school curriculum.
Broader Impacts
Fierce international competition in science and engineering, emerging competition in information technology, and expanding demand for health care have led policy makers and workforce analysts to anticipate enormous shortfalls in the supply of personnel in STEM fields. To meet the twin challenges of staffing science positions and of ensuring diversity in those positions, it is critical to cultivate and nurture all of the available talent in the sciences. The largest untapped source of such talent consists of young women who have the demonstrated ability in middle and high school to prosper in scientific domains, but for various reasons forego scientific fields and careers. By using heretofore unanalyzed data and innovative estimation methods to understand how young women (and men) make decisions about whether or not to pursue science careers, the investigators will lay the foundation for new and effective interventions that will alleviate future shortages in science talent in the United States. The results of this research will help to pursue the nation?s goal of building a larger and stronger pool of scientists in the service of economic development and scientific progress.
The main results of the project are: (1) gender differences in college major selection remain large, even in an era when the percentage of young women who enter college is ten percent higher than for young men; (2) occupational plans are the uniformly strongest predictors of college major selection among available variables measured in high school, and these associations cannot be explained away by tested performance in math and reading, high school coursework, or work-family orientation; and (3) gender differences in the associations between occupational plans and the selection of college majors are consistent with prior research on STEM attrition, even affecting the selection of majors that are gateways into doctoral-track medicine. The first result is well known and has been demonstrated in recent research by other scholars. The second result is surprisingly novel. No research has shown that the occupational plans of high school students are the strong predictors of initial college major selection among characteristics of students typically measured before college. We are only able to demonstrate their predictive power because we coded for substantively important differences in these plans based on the verbatim responses of 15,000 students. In the context of a study focused on STEM, this means capturing distinctions between occupational plans to enter science and engineering occupations, doctoral-track medical occupations, non-doctoral track medical occupations, and combinations of occupations that could express uncertainty in occupational plans. Our third conclusion, although also novel, is nonetheless not inconsistent with "leaky pipeline models" of college major selection. Our measured variables leave large portions of the gender gaps in college major unexplained, giving room for the mechanisms identified in the higher education STEM literature to enter with full force after young women and men enroll in postsecondary education. We also show that the occupational plans of young women are less predictive of initial college major selection, suggesting that such mechanisms may well disrupt the educational and occupational plans of young women more than young men. Notably, our results indicate that it is not merely that young women are disproportionately leaving science, but also that larger proportions of young women who, as seniors in high school, show interest in entering high-status science and medical occupations are eschewing or abandoning majors that lead to these occupations in favor of science-related majors that are gateways to non-doctoral track clinical and health sciences occupations: these students are not "lost" to science task-niches, but they are lost to the workforce of concern to those who fret about American scientific competitiveness. What are the implications of our results for policies intended to plug leaks in the science pipeline and ensure that talented young women are just as likely as talented young men to enter the scientific workforce? The first lesson is that by the time young men and women enroll in colleges and universities, gender-differentiated occupational plans have not only already been formed, but also have begun to channel young men and young women who have interests and abilities in science into different educational pathways. Although much of the pipeline literature and related policy efforts focus on proximate transitions to, and within, science careers, our results suggest that much earlier interventions are warranted. Our results also call for attention to gender-differentiated processes early in the college career that lead to greater attrition of women who, at one time, had aspirations to enter high-status STEM or doctoral-track medical careers, but who wind up in majors that are gateways to non-doctoral-track health and clinical science occupations. Additional research, and likely additional data, will be necessary to identify the mechanisms underlying this aspirational leveling and how to mitigate it, but our research at least pinpoints a critical point at which interventions might be most effective.
