Executive Summary

Twenty-five years ago, women were barely represented among doctoral scientists and engineers. Tens of new women Ph.D.s graduated each year in fields such as engineering, physics, chemistry and mathematics, as compared to hundreds of men. Female undergraduates in science and engineering stood a very small chance of ever encountering a woman professor of any description. A similar scarcity existed in industry.

All that has changed. In 1995, women were 32 percent of new science and engineering Ph.D.s and over 30 percent of faculty in many fields. Although they are nowhere close to half of doctoral scientists in most fields, they are a visible presence in the science and engineering workforce. This report traces the change from scarcity to visible presence.

The world did not sit still as women increased their presence in science and engineering. The structure of a scientific career changed, and continues to change, even as women’s presence has grown. The share of academic employment, once the predominant destination for new Ph.D.s, shrunk to less than half in most fields. As research funding grew in universities, so, too, did off-tenure track employment, a traditional employer for women in academia. Thus, although the representation of women expanded in all science and engineering fields, traditional mea-



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 1
From Scarcity to Visibility: Gender Differences in the Careers of Doctoral Scientists and Engineers Executive Summary Twenty-five years ago, women were barely represented among doctoral scientists and engineers. Tens of new women Ph.D.s graduated each year in fields such as engineering, physics, chemistry and mathematics, as compared to hundreds of men. Female undergraduates in science and engineering stood a very small chance of ever encountering a woman professor of any description. A similar scarcity existed in industry. All that has changed. In 1995, women were 32 percent of new science and engineering Ph.D.s and over 30 percent of faculty in many fields. Although they are nowhere close to half of doctoral scientists in most fields, they are a visible presence in the science and engineering workforce. This report traces the change from scarcity to visible presence. The world did not sit still as women increased their presence in science and engineering. The structure of a scientific career changed, and continues to change, even as women’s presence has grown. The share of academic employment, once the predominant destination for new Ph.D.s, shrunk to less than half in most fields. As research funding grew in universities, so, too, did off-tenure track employment, a traditional employer for women in academia. Thus, although the representation of women expanded in all science and engineering fields, traditional mea-

OCR for page 1
From Scarcity to Visibility: Gender Differences in the Careers of Doctoral Scientists and Engineers sures of status, such as becoming a full professor in a research university, did not expand proportionately. This report documents many dimensions of the changing representation of women in science and engineering. Using data from two NSF data bases—the Survey of Earned Doctorates for new Ph.D.s and the Survey of Doctoral Recipients for the S&E doctoral workforce—it brings together data on the educational background and demographic characteristics of three decades of new Ph.D.s and then examines their careers as described by the data. Although the report refers to explanations for the observed changes that are found in the literature, the focus is on the analysis of a very rich data set, not on establishing the root causes that give rise to the observed outcomes. It is the hope of the committee that this report will provide a common basis for decisionmakers in academia, industry and government to discuss whether the differences in career outcomes for women scientists and engineers are a matter for concern. It can identify areas where differences are greatest or most intractable, but it is left to policymakers to discuss what steps should be taken to narrow differences further. FINDINGS Degree Attainment and Educational Background From 1970 to 1995, there were significant advances in the entry of women into science and engineering. In the five broad fields considered (engineering, physical sciences, mathematical sciences, life sciences, and social/behavioral sciences) there were 350 percent more women among new Ph.D.s in 1995 than in 1973. In the social and behavioral sciences, women were just over half of the Ph.Ds in 1995 and in the life sciences they reached over 40 percent. Despite these strides toward equal representation in science and engineering, women are not anywhere close to being equally represented in all science and engineering fields. In 1995 they were 18 percent of bachelor’s degrees and 12 percent of Ph.D.s in engineering, compared to 50 percent and 40 percent, respectively, in the biological sciences. The move toward equal representation in doctorate achievement in science and engineering has been accompanied by growing similarities among men and women in background characteristics, such as parental educational background, type of baccalaureate institution, ranking of Ph.D. program, time to degree, and type of funding of graduate education. However, women are still less likely to obtain undergraduate degrees from Ph.D. granting institutions and more likely to take longer from time of baccalaureate to Ph.D.

OCR for page 1
From Scarcity to Visibility: Gender Differences in the Careers of Doctoral Scientists and Engineers Differences remain in the ways that men and women fund their education making it more likely that men are launched into research careers. Men are more likely to receive funding through research assistantships. Women are more likely than men to fund their graduate work by holding teaching assistantships in the physical sciences, mathematical sciences, and engineering—fields in which they are least well represented. Although the gender difference is narrowing, men are more likely than women to be married and to have children at the time they receive their Ph.D. Female Ph.D.s have children later in their career than do men. The increasing similarity of men’s and women’s educational background and demographic characteristics point to eventually narrowing career outcomes. Women, however, are systematically different from men in labor force participation, full time status, and the effect of children on labor force participation. These differences have an effect on career outcomes. Labor Force Outcomes The share of women in the S&E Ph.D. labor force has grown steadily between 1973 (9 percent) and 1995 (21 percent). Comparable figures for full-time scientists and engineers are 6.5 percent and 20 percent, respectively. There are broad differences in full time labor force representation across fields. Women made up almost a third of the social and behavioral science workforce in 1995. In engineering, they were only 5 percent. Their share was 10 percent and 11 percent in the physical sciences and mathematics, respectively. In the life sciences, women made up 26 percent of the doctoral workforce. The age distribution of women in the workforce differs significantly from that of men. Almost 50 percent of women had less than 10 years of experience, as compared to 30 percent of men. This reflects the recent growth in female Ph.D.s, but it also means that men are more likely to have the years of experience that go along with high status in their field. Doctoral women are less likely than men to be working full time in science or engineering. This difference has three components: full time work outside of S&E, part time work, and unemployment. Women are slightly more likely than men to work outside of science (2 percentage points) and considerably more likely to work part time (11 percent vs. 4 percent). Rates of unemployment have been low and declining over time for women in S&E, but the percentage of women who were unemployed and not seeking work rose from 3 percent to 4.6 percent between 1989 and

OCR for page 1
From Scarcity to Visibility: Gender Differences in the Careers of Doctoral Scientists and Engineers 1995. For men, this rate was 0.3 percent and 1.0 percent for the two survey years. Taken together, these sources led to 17 percent of women who were not employed full time in S&E in 1995 compared to 6 percent for men, despite a marked decline in women’s underemployment since 1973. Marriage and family are the most important factors differentiating the labor force participation of male and female scientists and engineers. Single men and single women participate equally in the workforce. Marriage and children are associated with increased rates of full-time employment for men, but declining rates for women. The negative effect of marriage and young children has declined for women over time. As predicted by a statistical model, single women were over 30 percentage points more likely to be working full time compared to women with small children (91 percent vs. 61 percent). By 1995, this predicted rate for women with young children had increased to 71 percent and the gap had narrowed to 22 percentage points. The impact of lower rates of labor force participation for women is that, at any point in her career, a woman on average has fewer years of work experience than a man who received his doctorate at the same time. The gap shrunk from 1979 to 1989, yet with 12 years of experience, the average woman had one year less of work experience than the average man (compared to 1.5 years less in 1979). The difference in work experience matters for career outcomes. Sector of Employment and Primary Work Activity Differences in the distribution of male and female scientists and engineers across sectors of employment shrunk between 1973 and 1995. In 1973, 8 percent of female scientists and engineers were employed in industry, while 26 percent of men were. In 1995, these percentages had risen to 26 and 37, respectively. The growth in the industry share came largely at the expense of academia, which employed 68 percent of women in 1973 and only 51 percent by 1995. For all fields, gender differences in distribution across sectors of employment narrowed or stayed essentially constant. By 1995, the largest differences in employment sector were in engineering, where 11 percent more men than women were employed in industry, and in the life sciences, where the gap was 7.5 percent. Primary work activity varies by employment sector. In academia, teaching and research are the primary activities. In industry, management and applied research are. Both men and women in academia report a shift to research as a primary work activity over the 1973–95 period. In the physical and life sciences, proportionately more women were engaged in teaching. Social/behavioral science was the only broad field where men reported teaching as their primary activity in higher proportion than

OCR for page 1
From Scarcity to Visibility: Gender Differences in the Careers of Doctoral Scientists and Engineers women. In business, there has been a steady decline in the percent reporting management as their primary work activity, but men are more likely than women to be engaged in management in all fields. Women predominate in professional services in the social/behavioral sciences (probably as counseling psychologists) and in applied research in the physical sciences and engineering. Regrettably, far less detailed data are collected about career paths in industry than in academia. We do know, however, that both men and women doctoral scientists and engineers have been shifting into jobs in industry rather than in academia and that there has been a narrowing in gender differences in sector of employment. Management is rarely a starting position for scientists and engineers in industry, so the continued dominance of men in management may reflect their greater seniority. Data do not permit us to learn if there is gender stratification within the general category of applied research. We do know far more about gradations in academia, and these findings are discussed below. Academic Careers The academic sector has a special importance in understanding differences in career outcomes for men and women in science and engineering. It is the sector where all Ph.D.s receive their initial training and, in most fields, it still employs the largest share of Ph.D.s. We look at the presence of women in academic science and engineering generally and then at differences among fields and types of institutions. Differences in career outcomes in what are called Research I Universities in the 1993 Carnegie classification of institutions of higher education are especially important in their dual role of employers and producers. Here we focus on their role as employers. Women are a growing presence in academic employment in all fields. Between 1973 and 1995, their representation almost tripled from 8 percent to 23 percent of full-time academic employment, even as the share of the academic sector in Ph.D. employment shrank. There is wide variation in the representation of women across fields. The life sciences and the social/behavioral sciences, which have shown the greatest increase of women among new Ph.D.s, have also seen the greatest increase in women’s representation in academic positions. This creates a difficulty, however, because academia has experienced very little overall growth in recent years and it may be more difficult for women to attain senior faculty positions than it was for men in the 1970s and 1980s when academic positions were growing.

OCR for page 1
From Scarcity to Visibility: Gender Differences in the Careers of Doctoral Scientists and Engineers Men and women have become increasingly similar in their distribution among types of institutions. In 1973, the percentage of men in Research I institutions exceeded that of women by 11 percentage points. By 1995, this difference had shrunk to 5 points. Earlier, women were more concentrated in master’s only and in baccalaureate institutions. Although women are found in research universities in increasing numbers, those numbers are still low in the physical sciences and engineering. In 1995, they made up only 6 percent of the Research I academic workforce in engineering and only 11–12 percent in mathematics and the physical sciences. To the extent that increasing the numbers of the women in these fields requires women as role models, it will be far more difficult than in the life sciences and social/behavioral sciences, where women make up 26 percent and 37 percent, respectively. When we look more closely at the types of academic positions women hold, the semblance of growing equality fades. Men continued to hold a 14 percentage point advantage over women in tenure track positions at a time when these positions were declining as a percentage of academic jobs. Through a logit analysis, we find that much of this difference is due to the lower career age of women, suggesting that women are quite likely to be increasingly well represented in these positions with the passage of time. Career interruptions matter to the chance that a person will achieve tenure track status. Women with interruptions before receiving the Ph.D. are more likely to become faculty, while this variable has the opposite effect for men. Being married with young children had a large and negative effect for women in 1979, but that effect had disappeared by 1995. For men, the effect was positive and greater in 1995 than it had been in earlier survey years. A close analysis of achieving tenure yields similar findings to the analysis of tenure track status. Adjusting for differences in the age structure for men and women, men are still more likely to be tenured than women at any professional age. Although these differences have declined over time, they persist, especially in research universities. Some of the difference in women’s status in academia can be explained by lower productivity, as measured by publications. It seems clear, however, that differences in the positions held by women are likely the cause of lesser productivity, rather than the other way around. Finally, in all academic ranks and measures of status, women are least represented in Research I institutions. Both Research I and other institutions have increased the representation of women. Comparatively, however, the increases have been greater in non-Research I institutions.

OCR for page 1
From Scarcity to Visibility: Gender Differences in the Careers of Doctoral Scientists and Engineers Salary To the extent that salary is a characteristic of rank and status, we would expect average salaries for female science and engineering Ph.D.s, who are on average younger and of lower rank than men, to be lower. Here, however, we control for a large number of differing characteristics of men and women, such as years of experience and field. Gender differences, although declining over time, persist even in the face of such adjustment. Specifically, Overall, male doctoral scientists and engineers had about a 20 percent salary advantage over women, and this difference persisted between 1973 and 1995. Although sizeable, this gap is smaller than the gender gap in salaries for professionals and in the labor force generally. Women’s salaries reach a plateau when they achieve 20 years of experience, while men’s continue to rise. This was the case in years preceding 1995, but does not necessarily predict what will happen to recent Ph.D.s as they gain experience. In fact, there is evidence that the salary gap had narrowed somewhat for recent cohorts as they entered mid-career. There are significant salary differentials across fields, with salaries in engineering and mathematics being greater than those in the life and behavioral/social sciences—fields in which there are relatively more women. The same holds true for sector of employment: women are relatively more concentrated in academia, where the median salary is lower, than in industry, where more men are employed. Regression analysis permits all the factors described above to be controlled for simultaneously. Doing so cuts the gender gap in salaries in 1995 from 21 percent to 6 percent with the biggest effects resulting from career age and field. The effect of these controls varies by survey year and is smallest in the early survey years, suggesting a greater degree of gender discrimination then. Generally speaking, the more restricted the population, the smaller the gender gap in salaries. That is, the gap for all full professors (11 percent) is smaller than the gap for all tenure track faculty (20 percent) and even smaller when the comparison is made within field. It appears to be the case, however, that there are larger within-rank within-field gaps in Research I Universities. CONCLUSION This report presents an exhaustive analysis of the available data on men and women Ph.D. scientists and engineers from 1979 to 1995. Dur-

OCR for page 1
From Scarcity to Visibility: Gender Differences in the Careers of Doctoral Scientists and Engineers ing that period, the numbers of women grew in all fields, sectors of employment and faculty ranks. Regression analysis was used to identify statistically the independent effects of educational background, field choice, career and family experience on a variety of career outcomes for men and women. Disparities in those outcomes have narrowed over time, but they remain. What the report has not done is to get behind the numbers. It has not systematically investigated the web of decisionmaking by those who have the power to influence careers. Both men and women encounter such guidance and gatekeeping at all stages of their careers in science. Equally important, it has not delved into the decisionmaking of the men and women themselves. It has not examined the complex calculus that men and women must conduct as they balance the pursuit of a scientific career with the often competing demands of marriage, children, and geographical location. We have observed only the outcomes. So far, these outcomes indicate that women, although they have made great progress toward equality in science and engineering in the past 25 years, are still more likely than their male counterparts to be in positions of lower status and lower pay. It is the hope of the Committee that the careful documentation of progress and stasis provided in this report will be of help to those who wish to see more equal use of talented women scientists and engineers to the benefit of science and engineering generally.