3
Examining Persistence and Attrition

CHAPTER HIGHLIGHTS

Women who start out on the path toward a career in academic science and engineering leave it for other fields at higher rates than their male counterparts. While there are field differences in pattern of attrition, more women than men leave at nearly every stage of the career trajectory. Fewer high school senior girls than boys state a desire to major in science or engineering in college. Girls who state such an intention are likelier than comparable boys to change their plans before arriving at college. Once in college, women and men show a similar persistence to degree, but women science and engineering majors are less likely than men to enter graduate school.

Women who enter graduate school in science and engineering are as likely as men to earn doctorates, but give a poorer rating to faculty-student interactions and publish fewer research papers than men. Many women graduate students report feelings of isolation. More women than men report plans to seek postdoctoral positions. Among postdoctoral scholars, women report lower satisfaction with the experience, and women are proportionately underrepresented in the applicant pools for tenure-track faculty positions.

It appears that women and men faculty in most fields who are reviewed receive tenure at similar rates. There is substantial faculty mobility prior to the tenure case, when some tenure-track ladder faculty move between institutions and others leave academe. Mo-



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 50
Beyond Bias and Barriers: Fulfilling the Potential of Women in Academic Science and Engineering 3 Examining Persistence and Attrition CHAPTER HIGHLIGHTS Women who start out on the path toward a career in academic science and engineering leave it for other fields at higher rates than their male counterparts. While there are field differences in pattern of attrition, more women than men leave at nearly every stage of the career trajectory. Fewer high school senior girls than boys state a desire to major in science or engineering in college. Girls who state such an intention are likelier than comparable boys to change their plans before arriving at college. Once in college, women and men show a similar persistence to degree, but women science and engineering majors are less likely than men to enter graduate school. Women who enter graduate school in science and engineering are as likely as men to earn doctorates, but give a poorer rating to faculty-student interactions and publish fewer research papers than men. Many women graduate students report feelings of isolation. More women than men report plans to seek postdoctoral positions. Among postdoctoral scholars, women report lower satisfaction with the experience, and women are proportionately underrepresented in the applicant pools for tenure-track faculty positions. It appears that women and men faculty in most fields who are reviewed receive tenure at similar rates. There is substantial faculty mobility prior to the tenure case, when some tenure-track ladder faculty move between institutions and others leave academe. Mo-

OCR for page 50
Beyond Bias and Barriers: Fulfilling the Potential of Women in Academic Science and Engineering bility patterns differ between women and men; men who move prior to tenure tend to leave academe, while women tend to enter adjunct positions. For women faculty members, feelings of isolation, lack of respect of colleagues, and difficulty in integrating family and professional responsibilities are major factors in attrition from university careers. For universities, faculty attrition presents a serious loss both economically and in morale. FINDINGS 3-1. There is substantial attrition of both men and women along the science and engineering educational pathway to first academic position. The major differences between the patterns of attrition are at the transition points: fewer high school girls intend to major in science and engineering fields, more alter their intentions to major in science and engineering between high school and college, fewer women science and engineering graduates continue on to graduate school, and fewer women science and engineering PhDs are recruited into the applicant pools for tenure-track faculty positions. 3-2. Productivity does not differ between men and women science and engineering faculty, but it does between men and women graduate students and postdoctoral scholars. Differences in numbers of papers published, meetings attended, and grants written reflect the quality of faculty-student interactions. 3-3. There is substantial faculty mobility between initial appointment and tenure case. Faculty at Research I universities are half as likely as the overall population of faculty to move to other types of academic institutions. Men and women hired into tenure-track positions had a similar likelihood of changing jobs, but men were twice as likely to move from academia to other employment sectors (15.3% of men and 8.5% of women) and women were 40% more likely to move to an adjunct position (9.2% of men and 12.7% of women). 3-4. Overall, men and women science and engineering faculty who come up for tenure appear to receive it at similar rates. Differences in the rate at which men and women receive tenure vary substantially by field and by race or ethnicity. For example, in social sciences women are about 10% less likely than men to be awarded tenure. African American women science and engineering faculty were 10% less likely than men of all ethnicities to be awarded tenure.

OCR for page 50
Beyond Bias and Barriers: Fulfilling the Potential of Women in Academic Science and Engineering 3-5. As faculty move up in rank, differences between men and women become apparent in promotions, awards, and salary. 3-6. No organization addresses the concerns of minority-group women; scientific and professional society committees address either women or minorities; most data are collected and analyzed by sex or by race or ethnicity. 3-7 Policy analyses of the education, training, and employment of scientists and engineers are hampered by data collection inadequacies, including lack of data, inability to compare data among surveys, difficulty in constructing longitudinal cohorts, difficulty in examining sex and race or ethnicity, and lags in the reporting of data. RECOMMENDATIONS 3-1. Efforts to increase the number of women in science and engineering should be focused on both recruiting and retention. Professional societies should work to recruit high school students to science and engineering careers. Colleges and universities should work to recruit women and minority students to science and engineering majors, to graduate school, and to faculty positions. University leaders and faculties need to work together to identify and remedy issues that address faculty retention. 3-2. Recruiting for faculty positions needs to be an active process that consciously develops and reaches out to women and minority-group scientists. Deans and department chairs and their tenured faculty should expand their faculty recruitment efforts to ensure that they reach adequately and proactively into the existing and ever-increasing pool of women candidates. 3-3. We need to understand more about faculty turnover. Universities should collect department data and scientific and professional societies should track discipline-wide turnover; the data should be collected annually and shared so that turnover dynamics can be understood and appropriate policies can be developed to retain faculty. 3-4. Changes should be made in the type of data that are collected on minority-group women and efforts should be made to ensure that the data are comparable across surveys and studies. Specifically, the National Science Foundation (NSF) Survey of Doctorate Recipients needs to be made more robust to allow for analysis of the small numbers of women of color. Other national surveys must collect data in a way that permits multiple demographic comparisons. Federal agencies and pro-

OCR for page 50
Beyond Bias and Barriers: Fulfilling the Potential of Women in Academic Science and Engineering fessional societies must report data so that the particular experiences of minority-group women can be understood and tracked and appropriate policies can be developed. 3-5. Universities should collect data annually on education and employment of scientists and engineers by sex and race or ethnicity using a standard scorecard format (Box 6-8). Data should include the number of students majoring in science and engineering disciplines; the number of students graduating with a bachelor’s or master’s degree in science and engineering fields; postgraduation plans; graduate school enrollment, attrition, and completion; postdoctoral plans; number of postdoctoral scholars; and data on faculty recruitment, hiring, turnover, tenure, promotion, salary, and allocation of institutional resources. The data should be made publicly available. 3-6. Scientific and professional societies should collect and disseminate field-wide education and workforce data with a similar scorecard. Women who start on the path toward a career in academic science leave that path in favor of other fields at a higher rate than their male colleagues. In this chapter, we will analyze sex differences in science and engineering education and career trajectories and rates of departure from the academic science track in favor of careers in other sectors. The decision to pursue a particular career path is a choice, but certainly not an arbitrary one. Forces other than individual preference or scholastic aptitude and preparation affect choices about career paths and appear to be driving women into careers outside of academic research. Not everyone who pursues a scientific education wants to be an academic scientist; 59% of science and mathematics, 55% of social science, and 28% of engineering graduate students say that they are preparing to become college or university faculty members or to seek postdoctoral research or academic appointments.1 In the United States, fewer than half of all people with PhDs in science and engineering are employed in the academic sector (Figure 3-1). As discussed in Chapter 2, social expectations and stereotypes regarding what it means to be a scientist or engineer influence career choices. Men benefit from a series of accumulated advantages: the implicit assumption that men can be academic scientists and engineers, the encouragement they 1 MT Nettles and CM Millett (2006). Three Magic Letters: Getting to PhD. Baltimore, MD: Johns Hopkins University Press. This study followed a sample of 9,036 graduate students from 21 of the major US doctorate-producing institutions from 1996 to 2001.

OCR for page 50
Beyond Bias and Barriers: Fulfilling the Potential of Women in Academic Science and Engineering FIGURE 3-1 Occupations of science and engineering PhDs by sector, 2002. SOURCE: National Science Foundation (2004). Women, Minorities, and Persons with Disabilities in Science and Engineering, 2004. Arlington, VA: National Science Foundation. receive to pursue academic careers, and role models provided by men who have successful academic careers. Women often suffer from a series of accumulated disadvantages, so when they make career choices, they choose from a set of options different from that of their male counterparts.2 Research shows that the more ways in which a person differs from the norm, the more social interactions affect choices; thus, the interlocking effects of 2 V Valian (1998). Why So Slow? The Advancement of Women. Cambridge, MA: MIT Press; MA Mason and M Goulden (2004). Marriage and baby blues: Redefining gender equity in the academy. Annals of the American Academy of Political and Social Science 596 (1):86-103; D Ginther (2006). The economics of gender differences in employment outcomes in academia. In Biological, Social, and Organizational Components of Success for Women in Academic Science and Engineering. Washington, DC: The National Academies Press.

OCR for page 50
Beyond Bias and Barriers: Fulfilling the Potential of Women in Academic Science and Engineering sex and race can further restrict career options.3 An analysis by the Education Trust4 found that 93 of every 100 white kindergartners would graduate from high school, 65 would complete some college, and 33 would obtain a bachelor’s degree. The corresponding numbers for black kindergartners were 87, 50, and 18, respectively. Of 100 Hispanic and Native American kindergartners, only 11 and 7, respectively, would earn a bachelor’s degree. There is no linear path to a degree. The default ‘pipeline’ metaphor … is wholly inadequate to describe student behavior [which] moves in starts and stops, sideways, down one path to another and perhaps circling back. Liquids move in pipes; people don’t. —Cliff Adelman, in The Toolbox Revisited: Paths to Degree Completion From High School Through College (2006)5 The question is where are differences in decision making manifested between men and women? The cohort of high school graduates who are now of an age to be assistant professors (assuming a direct educational path and no stop-outs) would have been seniors in the mid-1980s (Box 3-1 for a description of lagged cohort analysis). For this cohort, specific differences exist between the rates at which men and women chose and persevered in science and engineering education and careers.6 In 1982, high school senior girls were half as likely as boys to plan a science or engineering major in college. This difference was compounded by girls’ rate—2.4 times higher than that of boys—of attrition from the science and engineering educational trajectory during the transition from high school to college. During college, women and men showed similar perseverance to degrees in science and engineering fields. The other substantial difference in education and career attrition or perseverance between men and women in the cohort occurred during the transition from graduate school to tenure-track positions (Figure 1-2). 3 CSV Turner (2002). Women of color in academe: Living with multiple marginality. Journal of Higher Education 73(1):74-93. 4 Education Trust, Inc. (2002). The Condition of Education, 2002. Data were from surveys conducted by the US Department of Education and the US Department of Commerce Bureau of the Census, March Current Population Surveys, 1971-2001. 5 Available from the US Department of Education at http://www.ed.gov/rschstat/research/pubs/toolboxrevisit/toolbox.pdf. 6 Y Xie and KA Shauman (2003). Women in Science: Career Processes and Outcomes. Cambridge, MA: Harvard University Press.

OCR for page 50
Beyond Bias and Barriers: Fulfilling the Potential of Women in Academic Science and Engineering CONTROVERSIES BOX 3-1 Models of Faculty Representation Most analyses of career trajectories of women scientists and engineers use a pipeline analogy, positing that women are underrepresented at senior levels of academe because they are disproportionately “lost” along the journey from interested high school student to tenured faculty. However, analyses must take into account the number of years it takes for a person to progress from a newly attained PhD to a tenured faculty position. There is a lag between earning a degree and advancing to the next level and “without considering lag time, we are left with erroneous conclusions about what the distribution of women faculty should be without enough information about what the available pool of women is.”a Senior-level academics attained their PhDs a number of years before reaching the level of full professor. One study reports that in 2002 the middle 50% of full professors in physics earned their doctorates in 1967-1980.b Therefore, in considering the representation of women in this faculty rank, it is most appropriate to consider that representation in terms of the cohort of PhDs granted in 1967-1980. Similarly for associate professors the appropriate cohort (again using the example of physics) is 1984-1991 and for assistant professors (the “entry level” of the professoriate) it is 1991-1997. That is what is meant by considering “lag time.” Although the specific length of the lag time may vary from field to field (based on such factors as number of postdoctoral fellowships required before receiving a faculty appointment), the general principle applies in fields other than physics. When lag time is considered, one notices that when the current cohort of senior faculty received their doctorates there were fewer women in the pool than there are now. In some fields, that almost completely explains the low numbers of women in senior faculty positions. For instance in physics, in 2005 5% of full professors were women; in 1967-1980 (when the current cohort of full physics professors would have attained their PhDs) an average of 4% of PhDs were awarded to women. At the associate professor level, 11% were women in 2005; and in 1984-1991 (the appropriate year range for this cadre) 9% of PhDs went to women. At the assistant professor level, 16% were women in 2005; and in 1991-1997 (the appropriate year range for this cadre) 12% of PhDs went to women.c Similar findings are not confined to the discipline of physics. Using a similar type of analysis a National Research Council panel reported, in a general non-discipline-specific finding, that “much, but not all, of the difference in men and women in their success in becoming faculty is due to differences in the stage of their career.”d The panel predicted, in the coming decades, increases in the percentages of female faculty. However, other work presents an alternative view. Nelson, in a study of faculty representation at “top 50” science and engineering schools, reports that “in most science disciplines studied, the percentage of women among recent PhD recipi    aR Ivie and KN Ray (2005). Women in Physics and Astronomy, 2005. College Park, MD: American Institute of Physics, http://www.aip.org/statistics/trends/reports/women05.pdf.    bIvie and Ray (2005), ibid.    cIvie and Ray (2005), ibid.    dNational Research Council (2001). From Scarcity to Visibility: Gender Differences in the Careers of Doctoral Scientists and Engineers. Washington, DC: National Academy Press.

OCR for page 50
Beyond Bias and Barriers: Fulfilling the Potential of Women in Academic Science and Engineering ents is much higher than their percentage among assistant professors, the typical rank of recently hired faculty.”e Nelson finds further, that even in fields where women earn more PhDs than men (such as biology), “white males maintain their hold on the vast majority of assistant professor positions.”f Similar findings were reported by Myers and Turner, who found the disparity between the number of female PhD recipients and the number of female assistant professors to be especially acute for underrepresented minority groups.g Such findings indicate that qualified female candidates exist, but in many fields they are not being recruited into the tenure-track applicant pool in proportion to their presence in the PhD pool and suggest that the lag model is insufficient to account for the current underrepresentation of female faculty. The usefulness of the lag model discussed above depends on the validity of the pipeline model itself, a validity that has been questioned by some. The traditional pipeline model assumes a one-way flow in career progression, suggesting that once a person leaves science it is not possible to return. Work by Xie and Shauman challenges this paradigm, arguing that “exit, entry and reentry are real possibilities. Many persons, especially women, become scientists through complicated processes rather than by just staying in the pipeline.”h Others, including the Building Engineering and Science Talent (BEST) Initiative (Box 1-2) and the Human Frontier Science Program, have developed new paradigms for education, training, and career paths in the natural sciences.i Women may be more likely to pursue career paths that are not accounted for in traditional models of representation. Efforts should be made to be cognizant and supportive of those different career paths, and, in considering faculty representation, it is important to consider pathways beyond the pipeline paradigm. Xie and Shauman argue that the underrepresentation of women in science and engineering is “a complex social phenomenon that defies any attempt at simplistic explanation.” They note the “complex and multifaceted nature of women scientists’ career processes and outcomes” and suggest that increasing “women’s representation in science/engineering requires many social, cultural and economic changes that are large-scale and independent.” Clearly the pipeline model is important but, by itself, it is not sufficient to address underrepresentation. A National Research Council panelj found that, “while the most important    eDJ Nelson (2005). A National Analysis of Diversity in Science and Engineering Faculties at Research Universities. Available at: http://cheminfo.chem.ou.edu/%7Edjn/diversity/briefings/Diversity%20Report%20Final.pdf.    fNelson (2005), ibid.    gSL Myers and CS Turner (2004). The effects of PhD supply on minority faculty representation. American Economic Review 94(2):296-301.    hXie and Shauman (2003). Women in Science: Career Processes and Outcomes. Cambridge: Harvard University Press.    iThe BEST Initiative (2004). The Talent Imperative: Diversifying America’s Science and Engineering Workforce. Available at http://www.bestworkforce.org/PDFdocs/BESTTalentImperativeFINAL.pdf; European Science Foundation (2002). Towards a new paradigm for education, training, and career paths in the natural sciences. European Science Foundation Policy Briefing 16, http://www.esf.org/publication/139/ESPB16.pdf#search=%22Torsten%20Wiesel%20training%20paradigm%22.    jNational Research Council (2001), ibid.

OCR for page 50
Beyond Bias and Barriers: Fulfilling the Potential of Women in Academic Science and Engineering factor affecting gender differences in faculty status is the age of a scientist or engineer, there are important differences related to field, type of institution, and other variables.” A study by Kuck and colleagues highlights one of the other factors: the significance of the institution from which a person received their PhD as a factor in women’s likelihood of attaining a tenure-track position in chemistry. Kuck and colleagues examined hiring patterns in the 50 top-rated chemistry departments. They found that among the 50 departments, 10 schools supplied 60% of the younger faculty members, while only 32% of the faculty came from the other 40 schools.k The 10 top faculty-supplying schools were, with a few exceptions, also the top-rated graduate schools. In other words, “a small group of schools contributed a disproportionate number of younger faculty.” Postdoctoral placements also play a role in attaining tenure-track positions. Kuck and colleagues report that hiring of chemistry faculty by the top 50 universities is tracking the growth of women in postdoctoral appointments. Those who hold appointments at the top five suppliers of faculty are more likely to be preferentially hired by a top-50 department. Such findings demonstrate the influence of the PhD or postdoctoral institution on future career prospects and suggest that, when looking at faculty representation, it may be important to look at the pool of doctorates and postdoctorates from only a select subset of research universities.    kVJ Kuck et al. (2004). Analysis by gender of the doctoral and postdoctoral institutions of faculty members at the top-fifty ranked chemistry departments. Journal of Chemical Education 81(3):356-363. That type of analysis is useful for broad-brush policy development, but very specific differences by field must be acknowledged. Over the past decade, there have been significant changes, including increases in the numbers and proportion of girls taking high-level science and mathematics classes in high school and increases in graduate school enrollments and degrees. Research on underrepresentation in science and engineering focuses on the two categories of sex and race or ethnicity in large part because the data are collected by sex or race or ethnicity. As a consequence, minority-group women tend to disappear in analyses.7 Where possible, in the analysis of persistence and attrition in science and engineering education 7 See, for example, CB Leggon (2006). Women in science: Racial and ethnic differences and the differences they make. Journal of Technology Transfer 31:325-333.

OCR for page 50
Beyond Bias and Barriers: Fulfilling the Potential of Women in Academic Science and Engineering and academic careers, this report includes data on minority-group women broken out by race and ethnicity.8 COURSE SELECTION IN HIGH SCHOOL Rigorous study in high school is the best predictor of persistence to a degree in college.9 Advanced mathematics study appears to be an additional important factor in preparing students for college and can substantially narrow differences between racial and ethnic groups.10 The gender gap in science and mathematics courses taken in high school has narrowed over the last decade (Table 3-1). Since 1994, girls have been as likely as boys to complete advanced mathematics courses, including Advanced Placement or International Baccalaureate calculus.11 Also since 1994, girls have been more likely than boys to take advanced biology and chemistry. Physics is the only advanced science subject in which boys continue to complete courses at higher rates than girls, although the difference is small. African Americans and Hispanics were less likely than whites to complete advanced mathematics and science courses in high school. In an analysis of the National Educational Longitudinal Survey, Hanson found variability in attitudes toward science among women.12 For ex- 8 The committee acknowledges that there are different experiences within racial and ethnic groups. These are addressed in more detail in the National Science Foundation’s Women, Minorities, and Persons with Disabilities in S&E reports, http://www.nsf.gov/statistics/wmpd/; BEST reports, http://www.bestworkforce.org; NAS/NAE/IOM (2006). Biological, Social, and Organizational Components of Success for Women in Academic Science and Engineering. Washington, DC: The National Academies Press; G Campbell, R Denes, and C Morrison (1999). Access Denied: Race, Ethnicity and the Scientific Enterprise, New York: Oxford University Press; National Research Council (1992). Science and Engineering Programs: On Target for Women? Washington, DC: National Academy Press; National Research Council (1991). Women in Science and Engineering: Increasing Their Numbers in the 1990s: A Statement on Policy and Strategy. Washington, DC: National Academy Press; National Research Council (1989). Everybody Counts: A Report to the Nation on the Future of Mathematics Education. Washington, DC: National Academy Press. 9 LJ Horn and L Kojaku (2001). High School Academic Curriculum and the Persistence Path Through College: Persistence and Transfer Behavior of Undergraduates 3 Years after Entering 4-Year Institutions (NCES 2001-163). Washington, DC: US Department of Education. 10 C Adelman (1999). Answers in the Toolbox: Academic Intensity, Attendance Patterns, and Bachelor’s Degree Attainment (PLLI 1999-8021). Washington, DC: US Department of Education; G Orfield (2005). Dropouts in America: Confronting the Graduation Rate Crisis. Cambridge, MA: Harvard Education Press. 11 National Science Board (2006). Science and Engineering Indicators, 2006. Arlington, VA: National Science Foundation, Appendix Table 1-17. 12 SL Hanson (2004). African American women in science: Experiences from high school through the post-secondary years and beyond. NWSA Journal 16(1):96.

OCR for page 50
Beyond Bias and Barriers: Fulfilling the Potential of Women in Academic Science and Engineering TABLE 3-1 Percentage of High School Graduates Completing Advanced Coursework in Mathematics and Science, by Sex and Year of Graduation   1990 1994 1998 2000 Subject Men Women Men Women Men Women Men Women Mathematics Trigonometry/Algebra III 20.6 20.9 23.0 24.9 19.4 22.5 17.9 21.1 Precalculus/Analysis 14.4 13.0 16.3 18.4 23.1 22.9 25.4 27.9 Statistics and probability 1.2 0.8 2.0 2.1 3.4 4.0 5.8 5.6 Calculus 8.3 6.2 10.3 10.1 12.0 11.6 13.3 12.0 Science Advanced biology 25.7 29.2 31.5 37.8 33.8 40.8 31.5 40.5 Chemistry 43.8 46.1 47.5 53.3 53.3 59.2 58.1 66.8 Physics 24.9 18.3 26.7 22.5 31.0 26.6 35.6 31.5 SOURCES: US Department of Education, National Center for Education Statistics, National Assessment of Educational Progress, 1990, 1994, 1998, and 2000 High School Transcript Studies. Based on Table 1-8 in National Science Board (2006). Science and Engineering Indicators, 2006. Arlington, VA: National Science Foundation.

OCR for page 50
Beyond Bias and Barriers: Fulfilling the Potential of Women in Academic Science and Engineering TABLE 3-11 Average Start-up Packages for Assistant Professors in Selected Fields Starting in 2000-2001 at Public Research I Universities Field Start-up Equipment Other Supporta Moving Allowance Total Startup Biology $190,000 $27,000 $4,000 $221,000 Chemical engineering $225,000 $164,000 $5,000 $394,000 Chemistry/biochemistry $231,000 $14,000 $5,000 $250,000 Computer science $51,000 $35,000 $4,000 $90,000 Economics $6,000 $17,000 $5,000 $28,000 Geology $119,000 $0 $4,000 $123,000 Physics $156,000 $20,000 $4,000 $180,000 Political science $4,000 $5,000 $3,000 $12,000 Psychology $35,000 $9,000 $3,000 $47,000 Sociology $5,000 $4,000 $4,000 $13,000 aIncludes graduate student support ($140,000) and summer salary ($24,000) for chemical engineering; other disciplines also include support for postdoctoral scholars, renovations, and travel, but many schools left off such “other support” in the survey. SOURCE: University of Colorado at Boulder (2001). Faculty Recruitment and Retention Task Force Report, http://www.colorado.edu/AcademicAffairs/fac_recruit/fac_recruit.doc. There are costs associated with establishing search and recruitment committees and costs associated with relocation allowances, infrastructure, and support (for example, for laboratory renovations, offices, and equipment that might be required in support of new faculty). Those costs are included in the estimates discussed previously (and detailed in Tables 3-11 and 3-12). In addition, there is a substantial secondary cost associated with the loss of faculty and hiring of new faculty: that of research and grant productivity. In many cases, new faculty do not immediately bring the type of research-grant award support that productive, established faculty might. Callister reports that “it can take 10 years for a new faculty member in science or engineering to develop enough of a positive revenue stream from grants and to recoup start-up costs. If a faculty member leaves before startup costs are recovered, the university loses money and must start over again.”108 In monetary terms, that can be substantial. The UC-Boulder task force estimated that a productive faculty member “may bring about $100K per year” in external support to the university, external support that would take a new faculty member several years to generate.109 108 RR Callister (2006). The impact of gender and department climate on job satisfaction and intentions to quit for faculty in science and engineering fields. Journal of Technology Transfer 31:367-375. 109 University of Colorado (2001), ibid.

OCR for page 50
Beyond Bias and Barriers: Fulfilling the Potential of Women in Academic Science and Engineering TABLE 3-12 Start-up Costs Associated with New Professors     Private Research 1 Average (N) Private Nonresearch 1 Average (N) Public Research 1 Average (N) Public Nonresearch 1 Average (N) AA Physics and astronomy 395,746 (9) 147,944 (18) 320,932 (42) 169,491 (56) AA Biology 403,071 (14) 199,754 (26) 308,210 (38) 172,582 (55) AA Chemistry 489,000 (20) 221,052 (29) 441,155 (43) 210,279 (71) AA Engineering 390,237 (19) 152,101 (20) 213,735 (52) 112,875 (46) HA Physics and astronomy 563,444 (9) 254,071 (14) 481,176 (41) 248,777 (47) HA Biology 437,917 (12) 208,886 (22) 430,270 (37) 217,082 (49) HA Chemistry 580,000 (17) 259,348 (23) 584,250 (40) 284,269 (60) HA Engineering 416,875 (16) 209,057 (21) 259,494 (50) 146,831 (43) AP Physics and astronomy 701,786 (7) 90,000 (2) 740,486 (29) 359,783 (23) AP Biology 957,143 (7) 481,458 (12) 651,087 (23) 438,227 (31) AP Chemistry 983,929 (14) 532,046 (11) 989,688 (32) 550,349 (33) AP Engineering 1,441,667 (9) 326,694 (14) 408,443 (38) 223,292 (23) HP Physics and astronomy 1,000,000 (4) 418,333 (3) 1,110,577 (24) 455,882 (17) HP Biology 1,575,000 (5) 555,500 (10) 856,250 (16) 709,444 (27) HP Chemistry 1,172,222 (9) 575,000 (8) 1,187,115 (26) 648,913 (23) HP Engineering 1,807,143 (7) 452,000 (34) 472,086 (34) 254,597 (23) NOTES: Responses were tabulated from the Cornell Institute of Higher Research Institute Survey of Start-Up Costs and Laboratory Space Allocation Rules that was mailed to 3-5 chairs of selected biological science, physical science, and engineering departments at each reseach and doctoral university during the summer of 2002. AA: average start-up costs for new assistant profersors. HA: high-end start-up costs for new assistant professors. AP: aveage start-up costs for senior faculty. HP: high-end start-up costs for senior faculty. SOURCE: RG Ehrenberg, MJ Rizzo, GH Jakubson (2003). Who Bears the Growing Cost of Science at Universities? (Working Paper 9627). Cambridge, MA: National Bureau of Economic Research, http:www.nber.org/papers/c9627.

OCR for page 50
Beyond Bias and Barriers: Fulfilling the Potential of Women in Academic Science and Engineering Because science and engineering faculty incur costs continuously, some researchers have suggested that the aggregate costs required by new faculty (and not merely the initial start-up costs) should be considered in analyzing the cost of faculty turnover. Joiner110 has suggested an economic model for calculating the cost of turnover based on net present value (NPV). This model is commonly used in business to project the value of projects. It views faculty as long-term investments by considering all positive and negative cash flows for faculty members over time. Applying the model to faculty costs allows projections of the yearly costs of faculty salary, fringe and personal benefits, supplies and equipment, facility renovation, and other factors that are typically part of the costs accrued by universities in support of faculty (either new or existing). At the same time, the positive cash flows provided by a faculty member to the university (grant support, clinical revenues, and so on) are estimated. In concert, those two parts of the NPV model yield an estimate of the net cost (or financial yield) of a faculty member to a university.111 Using the NPV model, one could estimate the length of time a faculty member must remain at an institution for the institution to see a financial return on its investment. From a strictly economic perspective, if a faculty member leaves an institution prematurely (before the NPV model shows a positive yield), the institution loses money. In essence the NPV model dictates that “a dollar today is worth more than a dollar tomorrow.”112 Existing faculty are likely to have a positive NPV, whereas new faculty are likely to show a negative net cost. Accordingly, this model suggests that it is in the best financial interest of the university to direct efforts at retaining faculty. Some effective retention practices are outlined in Box 3-7. CASE STUDY: CHEMISTRY113 To examine the issue of faculty recruitment in more detail, the committee focused on chemistry, a field with a relatively high proportion of women PhDs. Information on the age, sex, and training of chemistry faculty members was obtained from the American Chemical Society’s 2001 DGR. The study was limited to faculties in the departments of chemistry, chemistry and biochemistry, or chemical biology at 86 Research I institutions. Only 110 KA Joiner (2005). A strategy for allocating central funds to support new faculty recruitment. Academic Medicine 80(3):218-224. 111 Joiner (2005), ibid. 112 Joiner (2005), ibid. 113 This section is based on research commissioned by the committee from Valerie J Kuck, Visiting Professor, Seton Hall University (Retired, Bell Labs).

OCR for page 50
Beyond Bias and Barriers: Fulfilling the Potential of Women in Academic Science and Engineering EXPERIMENTS AND STRATEGIES BOX 3-7 The University of Washington Faculty Retention Toolkita “Faculty retention is critical to the health of a university department both for morale reasons and also for economic reasons …” Recognizing that, the University of Washington has developed a toolkit designed to assist department chairs in retaining faculty of all ranks. The toolkit contains nine specific measures that when applied together act to encourage faculty satisfaction and productivity. The measures are designed to be applied to all faculty in a department but are noted to be “particularly important to women and underutilized minorities.” The toolkit contains the following measures: Monitoring the health and welfare of departments. Avoid disparities in workload, resources, salary, and recognition. Departments should provide regular state-of-the-department reviews, monitor faculty workload, and establish a process of individual faculty review meetings. Transparency in operations including fair and open promotion and tenure guidelines. Encourage open communication in the tenure process. Committee members should rotate, and faculty should have access to the evaluation process. Creating a welcoming department climate. Professional isolation is a common reason for faculty attrition. Encourage the development of a common department community, including social activities and professional recognition programs. Mentoring. Mentoring can be used as a powerful tool for fostering a sense of community and for professional development, learning, and collaboration. Valuing diversity in the department. Not all faculty fit the traditional view of a professor. Criteria of excellence should be expanded to include diverse approaches and values, such as involvement in outreach activities or nontraditional approaches to research. Supporting career development of pretenure faculty. New and pretenure faculty are at the highest risk of attrition. Specific efforts should be made to support and retain new and pretenure faculty by providing recognition, mentoring, professional development opportunities, and balanced workloads. Encouraging midcareer professional development. Professional development activities should continue for midcareer faculty. They include mentoring, professional recognition, and providing support to encourage creativity. Faculty development programs, benefits, and resources. Provide ongoing development programs, such as workshops and seminars, to introduce new faculty to programs on campus and renew and reinvigorate existing faculty. Flexible and accommodating policies and practices. Flexible family leave policies, dual career partner hiring programs, and transition support programs can play important roles in faculty productivity and retention.    aUniversity of Washington (2003). ADVANCE Center for Institutional Change Faculty Retention Toolkit, http://www.engr.washington.edu/advance/resources/Retention/index.html.

OCR for page 50
Beyond Bias and Barriers: Fulfilling the Potential of Women in Academic Science and Engineering TABLE 3-13 2001 Chemistry Faculty Members, by Country of Doctorate   Total Men Women All 2,476 2,218 261 (10.5%) Foreign PhD 305 284 21 (6.9%) SOURCE: American Chemical Society (2001). Directory of Graduate Research. Washington, DC: American Chemical Society. data on persons holding the rank of assistant, associate, or full professor were ascertained. Persons for whom there was no biographical information on training or rank were excluded from the study.114 The hiring data clearly show that chemistry faculty who have done their graduate work at Research I universities are overwhelmingly preferred; in addition, women faculty are drawn from a smaller pool of institutions than men. Of the 2,476 faculty members at the Research I institutions, 10.5% were female (Table 3-13). 12.3% of the faculty members earned their doctorates at a non-US institution; of these 6.9% were women—a smaller fraction than they were of all the faculty members. The top foreign institutions training the greatest number of future faculty members were Cambridge University, University College of London, and Oxford University. The median and average age of men faculty members were 49 years and 50 ± 11.8 years, respectively. The women faculty members were on average younger, with a median age of 42 years and an average age of 44 ± 9.2 years. It should be noted that a number of individuals did not give their date of birth (20 men and 11 females); therefore, they could not be included in these calculations. Since 1981 there has been an increase in the hiring/retention of women. A comparison of the number of men and women faculty members who received their doctorates during the same years indicates that the growth in the number of women faculty members has mirrored that of men who received their doctorate in the same time interval (Figure 3-7). In 2001, women held 18.3% of the positions at the rank of assistant professor and 17.9% of associate professor (Table 3-14) at Research I universities. A much lower percentage, 6.4%, of the full professor positions were held by women. Less than 4% of chemistry doctorates were found to hold faculty 114 The DGR contained the names of about 20 faculty members with no other information on their training or rank.

OCR for page 50
Beyond Bias and Barriers: Fulfilling the Potential of Women in Academic Science and Engineering FIGURE 3-7 Comparison of the number of men and women chemistry faculty members at RI institutions. SOURCE: American Chemical Society (2001). Directory of Graduate Research. Washington, DC: American Chemical Society. positions at Research I institutions. With the exception of the years 1971-1975, a higher percentage of men than women who earned chemistry PhDs ever were employed on Research I university faculties (Table 3-15). It appears that after all the efforts to increase the diversity of faculties, women with doctorates are still lagging behind men in attaining faculty positions at Research I institutions. There is a strong preference by Research I chemistry departments to hire graduates from a small subset of universities. Ten of the top 11 institutions were common to both men and women faculty (Table 3-16). Eleven TABLE 3-14 Chemistry Faculty, by Sex and Rank, 2001 Rank Total Men Women Assistant professor 464 379 85 (18.3%) Associate professor 408 335 73 (17.9%) Full professor 1,605 1,502 103 (6.4%) SOURCE: American Chemical Society (2001). Directory of Graduate Research. Washington, DC: American Chemical Society.

OCR for page 50
Beyond Bias and Barriers: Fulfilling the Potential of Women in Academic Science and Engineering TABLE 3-15 Proportion of Chemistry Doctorates Who Obtain Chemistry Faculty Positions at Research I Institutions, by Sex and Year of PhD Years Chemistry PhDs Granted Chemistry PhDs Who Obtain an R1 Faculty Position Women     1966-70 686 14 (2.0%) 1971-75 928 28 (3.0%) 1976-80 1,038 8 (0.8%) 1981-85 1,488 47 (3.2%) 1986-90 2,231 54 (2.4%) 1991-95 2,964 50 (1.7%) 1996-99 2,545 31 (1.2%) Men     1966-70 8,689 278 (3.2%) 1971-75 8,730 214 (2.5%) 1976-80 6,805 195 (2.9%) 1981-85 7,163 244 (3.4%) 1986-90 7,732 233 (3.0%) 1991-95 7,931 226 (2.8%) 1996-99 7,412 135 (1.8%) SOURCES: Chemistry PhDs: National Science Foundation (1966-2001). Survey of Earned Doctorates. Arlington, VA: National Science Foundation; data accessed through WebCASPAR. Chemistry Faculty: American Chemical Society (2001). Directory of Graduate Research. Washington, DC: American Chemical Society. departments graduated 54.6% of the US-trained men future RI faculty; Harvard University and the University of California, Berkeley, trained by far the most. For women, 11 departments graduated 51.7% of the US-trained women future RI faculty members, and Berkeley trained by far the most. During the years 1988-1997, women received 26.4% of the doctorates in chemistry. A lower proportion of women doctorates obtained faculty positions at Research I institutions than did men doctorates (Table 3-17). Of those Research I universities that hired more than 5 faculty, 4 hired above the pool, 7 hired at about the pool, and 19 hired substantially below the available pool of women chemistry PhD graduates. Programs designed to increase the representation of women chemistry faculty need to take into account cuts in the number of full-time faculty slots at doctorate-granting institutions, as demonstrated by the larger proportion but smaller number of women faculty (Table 3-18). This shrinkage of the tenure track is a general phenomenon. The academic employment of

OCR for page 50
Beyond Bias and Barriers: Fulfilling the Potential of Women in Academic Science and Engineering TABLE 3-16 Institutions Training the Greatest Number of Chemistry Faculty at Research I Institutions, by Sex and Year of PhD Men Faculty Membersa Women Faculty Membersb Institution Faculty Membersc Institution Faculty Membersc Harvard 179 Berkeley 32 Berkeley 175 California Institute of Technology 15 MIT 123 Harvard 10 California Institute of Technology 96 MIT 10 Wisconsin 92 Yale 10 Stanford 82 Cornell 10 University of Illinois, Urbana-Champaign 75 University of Illinois, Urbana-Champaign 8 Columbia 68 UCLA 8 Chicago 62 Stanford 7 Yale 52 Columbia 7 Cornell 51 Chicago 7 Total: 1,055 Total: 124 a54.6% of US-trained male faculty members. b51.7% of US-trained female faculty members. cNumber of PhDs trained at institution who subsequently hold faculty position at RI institution. SOURCE: American Chemical Society (2001). Directory of Graduate Research. Washington, DC: American Chemical Society. science and engineering PhDs increased from 118,000 in 1973 to 258,300 in 2003, full-time faculty positions grew more slowly than postdoctoral and other full- and part-time positions, and growth was slower than in the government and business sectors.115 CONCLUSION Individual efforts can have dramatic effects but sustained change is unlikely unless there is a transformation of the process by which students and faculty are educated, trained, recruited, and retained. To increase the numbers of women in science and engineering education and academic careers, policy action should focus on specific lever points: the transition to 115 National Science Board (2004). Science and Engineering Indicators, 2004. Arlington, VA: National Science Foundation, Table 5-6.

OCR for page 50
Beyond Bias and Barriers: Fulfilling the Potential of Women in Academic Science and Engineering TABLE 3-17 Number of Faculty Hired at Selected Research I Institutions, by Sex, 1988-1997 Hiring Institutiona Number of Faculty Hired Men Women Total % Women University of California, Berkeley 49 19 68 27.9 Harvard University 32 3 35 8.6 California Institute of Technology 27 6 33 18.2 MIT 25 0 25 0.0 Stanford University 23 5 28 17.9 University of Wisconsin, Madison 19 2 21 9.5 University of Illinois, Urbana-Champaign 18 2 20 10.0 Yale University 15 5 20 25.0 University of California, Los Angeles 13 4 17 23.5 University of Chicago 12 4 16 25.0 Columbia University 12 1 13 7.7 Cornell University 12 6 18 33.3 North Carolina State University 10 2 12 16.7 University of Texas, Austin 10 0 10 0.0 Northwestern University 8 1 9 11.1 University of Pennsylvania 8 0 8 0.0 University of Arizona 7 0 7 0.0 University of Michigan, Ann Arbor 7 1 8 12.5 University of Minnesota 7 2 9 22.2 Ohio State University 6 0 6 0.0 University of California, Irvine 5 1 6 16.7 University of California, San Diego 5 1 6 16.7 Princeton University 4 2 6 33.3 University of Colorado, Boulder 3 3 6 50.0 Pennsylvania State University 5 0 5 0.0 Purdue University 5 0 5 0.0 University of Southern California 5 0 5 0.0 Rochester University 4 1 5 20.0 Texas A&M University 4 1 5 20.0 Iowa State University 3 2 5 40.0 aOnly Research I universities that produced more than 5 faculty members are included. SOURCE: American Chemical Society (2001). Directory of Graduate Research. Washington, DC: American Chemical Society. college, graduate school faculty interactions, application and recruitment to faculty positions, and retention of faculty. Increasing the number of women and underrepresented minority-group faculty substantially will require assistance from faculty, individual departments, and schools; oversight and leadership from provosts and presidents; and sustained normative pressure, possibly from external sources. As dis-

OCR for page 50
Beyond Bias and Barriers: Fulfilling the Potential of Women in Academic Science and Engineering TABLE 3-18 Women PhD Chemists Working Full-Time at PhD-Granting Institutions, by Rank and Sex, 1990-2005   Percent Women Total Number of Women 1985 1990 1995 2000 2005 1985 1990 1995 2000 2005 Full professor 3.1 4.3 5.3 7.9 10.6 1,655 1,623 1,892 1,696 1,274 Associate professor 9.2 12.2 14.5 18.0 23.0 564 517 615 534 414 Assistant professor 12.1 18.4 22.4 25.2 26.0 431 511 557 563 389 Instructor, adjunct 23.4 30.8 40.4 39.9 37.0 141 133 203 271 167 Research appointment 25.3 20.5 22.5 24.0 19.8 225 728 1,153 883 359 Other nonfaculty N/A 27.6 26.9 30.6 30.0 N/A 225 405 310 172 No ranks 0.0 0.0 23.5 10.0 0.0 11 7 17 13 3 Total 8.3 12.8 15.9 18.5 18.7 3,058 3,744 4,842 4,270 2,844 NOTE: N/A indicates data not available. SOURCE: American Chemical Society (2006). Directory of Graduate Research. Washington, DC: American Chemical Society.

OCR for page 50
Beyond Bias and Barriers: Fulfilling the Potential of Women in Academic Science and Engineering cussed in the previous chapter, the first step is to understand that women are as capable as men of contributing to the science and engineering enterprise. As discussed in the next chapter, the science and engineering community needs to come to terms with the biases and structures that impede women from realizing their potential. The data show that policy changes are sustainable only if they create a “new normal,” a new way of doing things. The community needs to work together, across departments, through professional societies, and with funders and federal agencies, to bring about gender equity so that our nation can perform at its full potential.