Investigating the Gender Gap in SIAM PrizesDecember 13, 2011
Alice B. Popejoy, Phoebe S. Leboy, Jim Crowley, and Pam Cook
Each year SIAM awards a variety of prizes---some for noteworthy career-long research contributions, some for outstanding research in specific areas. Some prizes are reserved for early- and mid-career people; some recognize service to the community. Each is important, serving in a sense to define the area in which it is awarded. Prizes, especially those for younger people, can have a huge impact on careers. Clearly, the selection of prize recipients should be done fairly and equitably; all deserving groups should have the same opportunity for representation.
In the last fifty years, SIAM's flagship prize for distinguished contributions to the mathematical sciences---the John von Neumann Lecture---has been given to only three women: Olga Ladyzhenskaya (1998), Nancy Kopell (2007), and Ingrid Daubechies (2011). Is this surprising? Considering the small proportion of women among applied mathematicians in the early decades of the award's history, it is perhaps unsurprising, and even expected, that so few women have been selected to give the von Neumann lecture. The von Neumann lecture is not exceptional in this respect; very few women have received any of the major SIAM research prizes.
To investigate the low proportion of women among prize recipients, SIAM joined six other societies* and the Association for Women in Mathematics in the AWARDS project, funded by a grant from the National Science Foundation to the Association for Women in Science. The goal was to understand why there is a gender gap in our prize program and, if the gap cannot be attributed entirely to the low proportion of eligible female candidates, to determine how the awarding of prizes can be improved. An improved process would encourage objective decisions by selection committees, thereby supporting recognition of women for their research. Starting from the assumption that SIAM awards prizes to men and women objectively, we needed to look at the proportion of women receiving the prizes and compare it to the pool of eligible candidates---that is, to the proportion of active women researchers in the field.
Given the many fields in which SIAM members work, defining pools of eligible candidates for prizes is problematic. Therefore, we used three different measures of possible availability pools, with varying levels of relevance: percentage of PhDs awarded to women in mathematics 2035 years ago, percentage of female doctoral tenured faculty in mathematics departments in the last two decades, and percentage of female doctoral full-time faculty in applied mathematics and statistics departments.
To estimate the proportion of women among recipients of doctoral degrees in mathematics in the last few decades, we used the earliest available data on gender from the American Mathematical Society's Annual Survey of math departments and calculated the proportion of female PhD recipients in the United States from 1975 to 1990. Our estimate, that 15% of PhDs in this time frame were women, is most likely an overestimate of the availability pool: Not all recipients of PhDs in mathematics become SIAM members, some recipients of doctorates transfer out of the field after receiving their degrees, and many SIAM members have degrees in fields other than math (engineering, physics, or other STEM fields). In the last 20 years, 14% of tenured faculty in mathematics have been women; this is relatively consistent with the 15% doctoral recipient pool that would have fed into those positions.
These data include all faculty in mathematics departments reported to AMS. When only applied mathematics and statistics departments are included, 11% of doctoral full-time faculty members in the last 20 years have been women; more than 60% of them are tenured. Thus, when we look at all possible measures of eligible candidates for major awards, we find that the proportion of prizes given to women does not come even close to the proportion of eligible female candidates (Figure 1) and is less than half of the lowest possible cohort (Figure 2).
Figure 1. Proportion of eligible candidates for SIAM prizes (left) versus proportion of prize recipients by gender (right), based on average proportions of women (blue) and men (red) among PhD recipients, 19751991; tenured faculty in all mathematics departments included in the AMS Annual Survey, 19912009; and faculty in applied mathematics and statistics departments surveyed, 19912009. Figure 2 provides a breakdown of the measures of eligible candidates.
Figure 2. Percentage of women among: PhD recipients in all mathematics departments in the AMS Annual Survey, 19751991 (blue); tenured faculty in all math departments surveyed, 19912009 (red); faculty in applied mathematics and statistics departments surveyed, 19912009 (green); and recipients of research prizes, 19912009 (purple). Percentages for men complement those shown for women, as seen in Figure 1.
Psychology research suggests that implicit gender stereotypes held by both men and women lead us to associate men with careers and science, women with nurturing and the humanities. This could lead to an unconscious preference among prize selection committees for male candidates for research prizes. At a two-day workshop in Washington, DC, in June 2010, the participating societies explored this hypothesis and reviewed the records of each society. The society volunteers learned about implicit gender bias and discussed ways to make the selection process for their prizes more equitable. Strategies discussed include: ensuring that sufficient numbers of candidates have been nominated for a prize; broadening the candidate pool; having committee members agree on selection criteria before reviewing the candidate pool; ensuring that committee members understand their society's conflict-of-interest policy; promoting awareness of unconscious bias; and encouraging open discussion (by teleconference if not in person) among the committee members.
At SIAM, we are addressing this issue in several ways. First, to ensure a substantial pool of nominations, we have begun to promote prizes more actively. We now send a poster about the prizes to all academic departments on our mailing list; we also accept nominations earlier. Another step, planned for 2012, is the creation of committees whose task is to encourage individuals to submit nominations for prizes. One such committee will be named for each major prize; these committees, sometimes called "canvassing" committees, will operate completely independently of the selection committees. Finally, we are revising prize committee guidelines so that they provide more concrete information on standard procedures for making selections.
Our hope is that by enhancing the pools of candidates for SIAM prizes, and by ensuring that standard procedures are followed, we will enable our prize selection committees to choose the best people working in the field. A more diverse and representative set of prize recipients will be one welcome result.
*American Chemical Society, American Geophysical Union, American Mathematical Society, American Statistical Association, Mathematical Association of America, and Society for Neuroscience.
Carnes et al. (2005); Steinpreis, Anders & Ritzke (1999); Wenneras & Wold (1997); and Greenwald & Benaji (1995).
Alice Popejoy is the 20102012 public policy fellow at AWIS and works with Phoebe Leboy, a past president of AWIS, PI on the NSF AWARDS grant, and professor emeritus of biochemistry at the University of Pennsylvania School of Dental Medicine. SIAM executive director Jim Crowley attended the AWIS workshop. Pam Cook, SIAM secretary and a professor of mathematics at the University of Delaware, represented SIAM at the AWIS workshop, as did Suzanne Lenhart, a professor of mathematics at the University of Tennessee and a past president of AWM, and Juan Meza, provost at UC Merced.