Despite an upward trend in employment, women are underrepresented in many areas of science, technology, engineering, and mathematics (STEM): in 2011 the U.S. Department of Commerce found that women held only 24% of STEM jobs, compared to 48% of jobs overall, and that women with STEM degrees were less likely than men with STEM degrees to work in these fields. A 2015 analysis by the National Science Foundation recently reported that 29% of science and engineering jobs are held by women. Explanations that attribute the discrepancy to differences in aptitude or interest have been disproven, yet the problem remains: men outnumber women in STEM employment, particularly in high-ranking positions.
What causes the "leaky pipeline" for women in STEM, and what can be done to retain talented women? On March 24, 2015, the Academy's Science Alliance and the New York Chapter of the Association of Women in Science (AWIS) hosted a panel discussion on Policies that Influence Gender Balance in STEM to explore the varied ways individuals, universities, and funding agencies can work to bolster women in STEM fields. Moderated by Bamini Jayabalasingham, former president of AWIS MetroNY Chapter, the event featured panelists Kelly Mack of the American Association of Colleges & Universities, Sandra K. Masur of Mount Sinai School of Medicine, and Beth McCormick of Pfizer.
STEM careers do not pose unique barriers to female advancement: women face exclusionary cultures, unconscious biases, and lack of mentorship in many fields. McCormick argued, however, that such difficulties are often more acute in STEM fields. The degree of specialization required for STEM jobs makes it harder to find female candidates, because there are fewer people of either gender with the particular skill set needed. It is therefore easier to advance women in fields such as sales or marketing that require more broadly applicable training. Mack pointed to the inflexibility of many STEM jobs (requiring a physical presence at a lab bench, for example) as putting women at a disadvantage. Unconscious biases, such as picturing the typical scientist as the "man in the white coat," may also be more ingrained in STEM, Masur added.
The discussion next turned to policy initiatives that have been put in place to support gender balance in STEM fields, as well as to panelists' visions for future policies. Jayabalasingham referenced recent efforts in Europe to enforce gender quotas on corporate boards, and inquired whether that strategy could also work in STEM. Mack replied that while quotas can be a good start, "numbers aren't enough." Other factors, such as quality of life and opportunities for advancement, must also be considered when evaluating the success of gender balance initiatives. For example, in the life sciences, where more than half of baccalaureate recipients are female, it is tempting to think that balance has been achieved—but a much lower proportion of women advance to faculty and leadership positions.
According to Mack, those numbers will not change until there is a shift in the culture of academia, which may be difficult to achieve in such an inveterate system. It would require people who benefit from the current structure to acknowledge its deficiencies and be open to change. "We're asking people to let go of their implicit biases," she explained. "Those things don't change overnight." She emphasized the need for patience: until the culture changes, policies will improve the outlook for women but will not alter the root causes of gender imbalance. McCormick drew on her experience in the corporate world to present an optimistic view: "There's a general understanding that [diversity] is smart for business, smart for science," she said. Leaders play a key role in setting the tone for inclusivity at organizations, as does training that helps employees recognize their own biases and teaches skills for supporting underrepresented groups.
Several institutions have sought to implement family-friendly policies to improve work–life balance for women in STEM. However, these policies vary widely, and it is difficult to measure their effectiveness. Mack described an AAC&U initiative to assign institutions a "family friendliness" score using a mathematical framework that accounts not only for policies in place but also for those policies' inclusiveness of diverse demographic groups and their needs. An inclusive policy, for example, might recognize that some families need assistance with child care, while others require elder care. The score also incorporates features that are harder to quantify, such as the culture of the institution and whether workers take advantage of its policies. The AAC&U researchers will track the metrics with surveys aimed at gauging how well faculty understand a policy, whether faculty can use the policy in practice, and whether any stigmas perceived to be associated with using benefits could cause women to hesitate before doing so.
Family-friendly policies include family and medical leave benefits as well as the option to "[stop] the tenure clock," which relaxes the pressure on faculty who become parents to obtain grants or produce publications in the year after a birth or adoption. Masur described an intriguing policy that Stanford School of Medicine is testing with funding from a Sloan Foundation grant: faculty earn credits for time spent on non-research tasks, such as teaching or serving on committees, and can redeem the credits for services such as meal deliveries or housecleaning. This program builds on research demonstrating that female faculty in dual-career couples spend more time than their male counterparts on housework, leaving less time available for work duties.
While promising, policies like Stanford's may not be feasible for all universities. Here, funding agencies can play a part in retaining women in STEM with grants aiming to transform institutional culture, such as those awarded by the NSF ADVANCE Program; with grants for individuals, such as NIH reentry awards; and with grants giving principal investigators latitude to pay for extra lab staff during periods of family leave or for child care at scientific meetings. Funding agencies can also include language that alerts reviewers to unconscious biases, and help change academic culture by withholding funding from institutions that ignore diversity issues. Such actions, Mack pointed out, could force schools to reexamine their institutional culture and remedy work–life balance issues that disproportionately affect women.
When asked what individuals can do to advance the cause of women in STEM, each speaker highlighted a different approach. Masur emphasized the advantages of working through groups such as the AWIS or the American Society for Cell Biology's Women in Cell Biology committee, which she chairs. Such groups not only provide access to mentoring support and career development advice, but also offer a more visible means to call for policy and legislative changes. "We do [things] much more effectively in groups than we do as individuals," she said. McCormick countered that individuals can also act as leaders and mentors, building a "critical mass" of women in STEM that could "change the conversation." Individuals can also be involved in changing policy more directly, Mack argued, by contacting elected officials to raise concerns.
The need to balance work and family in a STEM career is not unique to women, and in fact many men take advantage of tenure clock and family leave policies. Thus, McCormick stressed the importance of framing the discussion as pertinent to the needs of families, not just those of women. Though agreeing with McCormick, Mack pointed to AAC&U data suggesting that family-friendly policies do not always benefit men and women equally; after "stopping the tenure clock," for example, men tended to return to work with more grants and publications than did women, perhaps because women more often acted as the primary caregiver at home. Masur noted that some institutions require faculty to be responsible for 50% or more of a child's care to be eligible for the tenure clock policy, a requirement that promotes gender equality by targeting faculty with the greatest child care burdens. Indeed, changing gender expectations in society may have the greatest impact on gender balance in STEM, according to McCormick. "The notion of shared responsibility at home is really going to move the needle on women's ability to continue to advance their careers," she said.
To highlight the role data collection can play in shaping policy, the event also featured a presentation by Eric Sweet, a postdoctoral fellow at Mount Sinai School of Medicine and a representative of the NY Citywide Postdoc Council. Sweet spoke about his experience leading Mount Sinai's annual survey of postdocs, which evaluates how institutional policies affect postdoc scientists' lives. For those seeking to conduct similar surveys, he offered advice on advertising the survey, designing appropriate questions, and making the most of data collected. He pointed to recent changes in postdoctoral mentoring programs at Mount Sinai as an example of how survey data can influence policy. Taking a cue from the evening's theme, Sweet also showed how survey data can shed light on issues faced by women in STEM. For example, survey results revealed that female postdocs rely more on child care than do male postdocs. He emphasized that delivering these kinds of data to administrators and policy makers will play a large role in improving quality of life for scientists of both genders.