6
The Academic Career

Faculty, Unfaculty, and Changes in the Academy

…the classic profile of the academic career is cut to the image of the traditional man with his traditional wife.

—Arlie Russell Hochschild, Inside the Clockwork of Male Careers, 19751

Academic science is the model for professional science. To rise in this system, one must climb an extraordinarily narrow ladder: from graduate student to postdoctoral fellow to research associate to assistant professor (or principal investigator). The majority of women in science have never completed that rise. They have remained research associates attached to the principal investigator for most or all of their working lives. The cause of arrest is multiple and it has a history.

—Vivian Gornick, Women in Science, 19902

INTRODUCTION

In this chapter we examine the careers of doctoral scientists and engineers in academia. Our analysis of the academic sector is far more detailed than those of other sectors for several reasons. First, doctoral scientists and engineers are traditionally trained to work in academia. Although the proportion of scientists and engineers working in academia has been declining since 1970, the academic sector remains the single largest employer of doctoral scientists and engineers. Second, the conduct of basic scientific research in the United States is intertwined with the higher education system. Institutions of higher education traditionally attract the best scientists and provide them with the most resources and rewards (Clark 1995; Wolfle 1972). As documented by Fox (1996), the evolution of science and the evolution of higher education have been reciprocal developments in the United States. In Wolfle’s words, academia is “the home of

1  

Hochschild (1975).

2  

Gornick (1990:81).



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 123
From Scarcity to Visibility: Gender Differences in the Careers of Doctoral Scientists and Engineers 6 The Academic Career Faculty, Unfaculty, and Changes in the Academy …the classic profile of the academic career is cut to the image of the traditional man with his traditional wife. —Arlie Russell Hochschild, Inside the Clockwork of Male Careers, 19751 Academic science is the model for professional science. To rise in this system, one must climb an extraordinarily narrow ladder: from graduate student to postdoctoral fellow to research associate to assistant professor (or principal investigator). The majority of women in science have never completed that rise. They have remained research associates attached to the principal investigator for most or all of their working lives. The cause of arrest is multiple and it has a history. —Vivian Gornick, Women in Science, 19902 INTRODUCTION In this chapter we examine the careers of doctoral scientists and engineers in academia. Our analysis of the academic sector is far more detailed than those of other sectors for several reasons. First, doctoral scientists and engineers are traditionally trained to work in academia. Although the proportion of scientists and engineers working in academia has been declining since 1970, the academic sector remains the single largest employer of doctoral scientists and engineers. Second, the conduct of basic scientific research in the United States is intertwined with the higher education system. Institutions of higher education traditionally attract the best scientists and provide them with the most resources and rewards (Clark 1995; Wolfle 1972). As documented by Fox (1996), the evolution of science and the evolution of higher education have been reciprocal developments in the United States. In Wolfle’s words, academia is “the home of 1   Hochschild (1975). 2   Gornick (1990:81).

OCR for page 123
From Scarcity to Visibility: Gender Differences in the Careers of Doctoral Scientists and Engineers science” (Wolfle 1972). Third, indicators of career attainment and rewards are more public and more uniformly defined in academia than in other sectors, allowing researchers to more readily collect detailed data on career outcomes. The position of women in the academic sector is also critical because it is within academia that future generations of scientists and engineers are trained. Frieze and Hanusa (1984) discuss a variety of reasons why female faculty may be especially important as role models and mentors for female graduate students. The presence of more than a token number of women on the faculty of graduate programs may be important both for recruiting new generations of women to graduate programs and for retaining them once they enroll in graduate education. Accordingly, in the analyses that follow we give special attention to scientists and engineers working in Research I universities and medical schools. Not only do these locations provide the majority of doctoral and postdoctoral training, but they are also the most conducive organizational contexts for a prestigious research career. For women to have an equal standing with men in science and engineering, it is essential that they gain parity within the most prestigious academic locations. While our focus in this chapter is on scientists and engineers with full-time employment in academia, it is important to keep in mind that a greater proportion of women than men are part time employees in academia, as shown in Chapter 4. In the rest of this chapter, unless otherwise noted, we restrict our analysis to the full-time labor force. FULL-TIME EMPLOYMENT IN ACADEMIA3 From 1973 to 1995, the percent of the combined male and female doctoral labor force that worked in academia decreased from 51 percent of all scientists and engineers to 40 percent. In 1973, 5 percentage points more men than women were working full time in academia, as shown by the two sets of bars on the left hand side of Figure 6–1. The relative decline in academic employment that occurred after 1973 was more rapid for men than for women, so that by 1995 three percentage points more women than men held full-time academic jobs. While our findings may appear to contradict past research that found women to be over-represented in academia (Zuckerman and Cole 1975 and the literature cited therein), keep in mind that we are considering men and women as a percent of the entire labor force, not as a percent of those working full time. If we consider only those in the full-time labor force (i.e., excluding those who are 3   See Appendix Tables D-1-D-2 for further information.

OCR for page 123
From Scarcity to Visibility: Gender Differences in the Careers of Doctoral Scientists and Engineers FIGURE 6–1 Percent of the doctoral labor force that is working full time in academia and percent of the full-time labor force that is working in academia, by sex and year of survey. part time, unemployed, retired, or out of the labor force), women are substantially more likely to be in academic positions, as shown by the bar graphs on the right. In 1973, 68 percent of women in the full-time labor force held academic jobs, compared to only 56 percent of the men. By 1995, this 12 point difference decreased to 7 points. Thus, over this 22 year period, full-time employment decreased 18 points for men and 38 points for women. The net effect of the increasing proportion of Ph.D.s who are women and the greater proportion of women than men in academic jobs is a steady increase in the percent of all full-time academic jobs that are held by women. The gray circles in Figure 6–2 show that 8 percent of all full-time academics were women in 1973, increasing to 23 percent in 1995. While this increase is driven largely by increasing numbers of women in science and engineering, the increase of 15 percentage points exceeded the growth of women as a percent of all Ph.D.s (shown by squares) and as a percent of all scientists and engineers who are working full time (shown by triangles). This reflects a combination of an increasing proportion of women working in academia and the possibility that the attrition of women from academic jobs has decreased.

OCR for page 123
From Scarcity to Visibility: Gender Differences in the Careers of Doctoral Scientists and Engineers FIGURE 6–2 Women as a percent of all Ph.D.s, as percent of all full-time scientists and engineers, and as percent of all full-time academic scientists and engineers, by year of survey. Field Differences in Full-Time Academic Employment Women are proportionally more likely than men to be in academic jobs in all fields except the social and behavioral sciences. This is shown in Figure 6–3, which plots differences between the percent of men with academic jobs Ph.D.s (as a percent of men who are working full time) and the corresponding percent of women. Positive values indicate a greater proportion of full-time women than full-time men are working in academia. In 1973 for all fields combined, 12 percentage points more women than men in the full-time labor force were employed in academia. By 1995 the difference was reduced to 6.5 points. Within fields, we find that even though the largest proportion of female Ph.D.s are found in the social and behavioral sciences, this is the only field with a greater proportion of men than women in academic jobs. Women in engineering and the life sciences are the most likely to be academic, with little change over time. In mathematics and the physical sciences, gender differences in full-time academic employment have nearly disappeared by 1995. While there is an increasing representation of women in each field, substantial variation exists across fields in the proportions, numbers, and

OCR for page 123
From Scarcity to Visibility: Gender Differences in the Careers of Doctoral Scientists and Engineers FIGURE 6–3 Gender difference in the percent of Ph.D.s working full time who have academic jobs, by year of survey. NOTE: Positive values indicate women are proportionally more likely to have academic positions. rates of increase of female academics, as shown by Figures 6–4 and 6–5. In the life and social/behavioral sciences, the percent of full-time academics who are women increased by nearly 20 percentage points from 1973 to 1995. As a result of the greater overall increase in the number of life scientists during this period, by 1995 there were more women in the life sciences than the social and behavioral sciences. In other fields, the increase in the percent of women was only between 6 and 7 points. Even by 1995, women were only 6 percent of the full-time academic work force in engineering, with less than 2,000 full-time female engineers. In mathematics and the physical sciences, women’s representation exceeded 10 percent by 1995, but in mathematics the number reached only 2,000 and in the physical sciences just over 4,000. In the life sciences, the percent of women approached 30 percent by 1995. The rapid change in the percent of academic positions held by women is largely the result of increases in the proportion of new Ph.D.s who are women. The effects of the more recent entry of women are seen by comparing Figure 6–6, which plots the percent women among those who re-

OCR for page 123
From Scarcity to Visibility: Gender Differences in the Careers of Doctoral Scientists and Engineers FIGURE 6–4 Percent of the full-time academic labor force that is female, by field and year of survey. FIGURE 6–5 Number of women working full time in academia, by field and year of survey. ceived their Ph.D.s 11 or more years ago, to Figure 6–7 for those with doctorates within 10 years of the survey. While there were increases in the percent of women among those with older degrees, these changes are substantially smaller than for those with more recent degrees. Among older academics, the presence of women grew most rapidly in the social and behavioral sciences, but in 1995 women still represented less than 25 percent of the total among older social and behavioral scientists. When we consider more recent Ph.D.s, the increases and overall levels are much

OCR for page 123
From Scarcity to Visibility: Gender Differences in the Careers of Doctoral Scientists and Engineers FIGURE 6–6 Percent of the full-time academic labor force that is female for those who received their Ph.D.s more than 10 years ago, by field and year of survey. FIGURE 6–7 Percent of the full-time academic labor force that is female for those who received their Ph.D.s in the last 10 years, by field and year of survey.

OCR for page 123
From Scarcity to Visibility: Gender Differences in the Careers of Doctoral Scientists and Engineers larger. By 1995, nearly 40 percent of the younger academics in the life and social/behavioral sciences were women. An important implication of the more recent entry of women than men is that women are concentrated among the younger members of the faculty and research staffs. This has important consequences for understanding the presence of women among those who are tenured and have higher ranks, a topic which is discussed in detail below. Consequently, it is useful to consider the age distribution of men and women in academia more thoroughly. THE AGE STRUCTURE IN ACADEMIA The average academic woman received her degree more recently than the average academic man. Moreover, the difference between the average career age (i.e., years since the Ph.D. was received) for men and women is increasing. In 1973, the mean career age for women was 9.5 years and 11.1 years for men; in 1979, 8.9 years for women and 12.7 years for men; in 1989, 10.6 years for women and 15.9 years for men; and in 1995, 11.2 years and 17.0 years. The effects of changes in the growth of academia and the entry of women can be seen with a population pyramid (see Shyrock and Siegel 1973:236–245 for details). A population pyramid is a pair of horizontal histograms, one for men and one for women, with each bar representing the percent in an age group. Typically, the length of the bars corresponds to the percent in a given age-sex group (e.g., women aged between 1 and 3) relative to the size of the total population. Alternatively, a within sex pyramid can be used in which percentages for men are computed on the basis of the number of men in the population and the percentages for women are based on the number of women. A within sex pyramid is useful when there are large differences in the overall number of men and women, such as in academia. The shape of a pyramid reflects the number of each sex entering the population (e.g., new Ph.D.s) and the number leaving the population through death or retirement. For example, if the same number of new Ph.D.s were hired each year and there was no attrition until the age of retirement, the pyramid would be a rectangle. Or, if the size of new cohorts is increasing with increasing attrition among older members of the population, the pyramid would be triangular. Figure 6–8 contains within sex population pyramids for academic scientists in 1973 and 1995. Consider the age profile for women in 1973 (Panel A). Nearly 30 percent of female academics were within 3 years of their Ph.D. and over half were within 6 years. A substantially smaller number of women were found between ages 7 and 21, with 9 percent more men than women at these ages. The dark half of the pyramid for men has a narrower base and more area at the top, reflecting the greater proportion of men who are older. By 1995 (Panel B), the age structure of

OCR for page 123
From Scarcity to Visibility: Gender Differences in the Careers of Doctoral Scientists and Engineers FIGURE 6–8 Sex specific distribution of career ages of scientists in the full-time academic labor force. NOTE: Percentages are sex specific. For example, in 1973, 20 percent of the men were 1–3 years from the Ph.D.

OCR for page 123
From Scarcity to Visibility: Gender Differences in the Careers of Doctoral Scientists and Engineers academia had changed dramatically as a result of the rapid entry of women and the end of growth in the size of new cohorts of men. For men, the age pyramid from age 1 to 27 is nearly uniform with roughly 3 percent of the male academics at each year from the Ph.D. The slight narrowing of the pyramid between ages 7 and 22 corresponds to the contraction of the academic labor market between 1973 and 1995. For women, the continuing increase in the entry of women is shown by the triangular shape of the distribution. Given that tenure and rank are time dependent, it is clear that women in 1995 must be found less frequently in advanced ranks, a topic considered below. Finally, it is interesting to note that the age structure for women in 1995 is very similar to the age structure for men in 1973. Since Figure 6–8 computes percentages within each sex, it does not reflect differences in the relative numbers of men and women. That is, it reflects rates of entry and exit from academia, but does not reflect the number. Since it is also important to understand how many women are at each age relative to men, Figure 6–9 computes percentages based on the entire population. For example, in Panel A we see that in 1973 women with Ph.D.s within the last 3 years represented 2.5 percent of all academic scientists (in Figure 6–8 we saw that these women represented nearly 30 percent of female academics). In 1973, the youngest group of men represented nearly 20 percent of all academics, while the youngest group of women represented less than 3 percent. By 1995, new female Ph.D.s grew to 4 percent of academics, while new men dropped to less than 8 percent. Overall, the slowed growth of academia is shown by young Ph.D.s dropping from over 20 percent in 1973 to 12 percent in 1995. Even with the rapid increase in the percent of women receiving Ph.D.s and entering academia, women are far from being half of the academic labor force, as shown by the much smaller area of the light gray bars compared to the dark gray bars. While new cohorts of Ph.D.s entering the academic marketplace are increasingly female, each new cohort is only a small proportion of those currently employed. Consequently, the move towards parity in the representation of women must occur slowly. While there has been a substantial increase in women with academic jobs, it remains to be determined whether there is a correspondingly large increase in the presence of women among all types of positions, ranging from full professors at elite research universities to visiting lecturers at two-year colleges. To this end, we begin by examining variations in the types of institutions in which men and women are employed. We then extend these analyses to consider variation in the types of jobs held by men and women in academia.

OCR for page 123
From Scarcity to Visibility: Gender Differences in the Careers of Doctoral Scientists and Engineers FIGURE 6–9 Distribution of career ages of scientists in the full-time academic labor force. NOTES: Panels A and B show the percent of the total population in a given age/sex category. For example, in 1973 18 percent of all scientists were men 1–3 years from the Ph.D.; 2.5 percent of all scientists were women 1–3 years from the Ph.D.

OCR for page 123
From Scarcity to Visibility: Gender Differences in the Careers of Doctoral Scientists and Engineers D-9 and D-10 for detailed information), these differences are largely due to differences in age structures across fields and types of institutions, with a strong correlation between the mean age of a particular group (e.g., women in engineering) and the percent in that group who are full professors. Logit Analyses of Academic Rank A multinomial logit was used to predict the proportion of men and women at each rank after adjusting for differences in field, type of institution, and, most importantly, age (Figure 6–35). While the full results of these analyses are given in Appendix Table D-12, our discussion focuses on the proportion of full professors, since results for assistant and associate professors duplicate the information in the section on tenure. As shown by the dark bars in Figure 6–36, there has been little change since 1979 in the observed over-representation of men among full professors. The age-adjusted differences, shown by the gray bars, show a decrease in the over-representation of men from 20 points in 1979 to under 10 points in 1995. Still, even after controlling for gender differences in career age, field of employment, and type of institution, men continue to have an almost 10 percentage point advantage in being full professors. The improved representation of women occurred in most types of FIGURE 6–35 Difference between men and women in the observed proportion of full professors and the adjusted proportions controlling for field, career age, and Carnegie type of institution, by year of survey. NOTES: Adjusted proportions are for 20 years after the Ph.D. Data were not available in 1973.

OCR for page 123
From Scarcity to Visibility: Gender Differences in the Careers of Doctoral Scientists and Engineers FIGURE 6–36 Differences between men and women in the adjusted proportion who are full professors, by Carnegie type of institution and year of survey. NOTES: The difference was 0 in Medical institutions in 1995. Data were not available in 1973. institutions, as shown in Figure 6–37. Most importantly, because of its implications for training future generations of scientists and engineers, the 25 point over-representation of men among full professors in Research I institutions in 1979 was reduced to 8 points by 1995, and the 35 point advantage in Medical institutions in 1979 was eliminated. Surprisingly, given the historical presence of women in undergraduate institutions, the over-representation of men in Baccalaureate institutions returned to 15 points after dropping in 1989. To examine the effects of having children, we included a variable indicating whether a scientist had children under the age of 18. The age of 18 was used since delays in promotion may result from the accumulated effects of children over the entire career. Figure 6–37 plots the difference in the adjusted percent of married female faculty who are full professors and the adjusted percent of full professors among those with children. Again we see that the effects of family differ for men and women. In 1979 women with young children were 12 points less likely to be full professors. The effect decreased to 5 points in 1989, where it remained in 1995. The lower probability of being a full professor for women with children may reflect a cost in productivity, delays in beginning their first faculty position, or an assumption on the part of the university or department that women with children are a poor risk. Unfortunately, the SDR does not provide information on when faculty began their current job; while

OCR for page 123
From Scarcity to Visibility: Gender Differences in the Careers of Doctoral Scientists and Engineers FIGURE 6–37 Effects of being married compared to having children on the probability of being a full professor, by sex and year of survey. NOTES: Negative values indicate children make being a full professor less likely. Data were not available in 1973. we have information on possible interruptions before the Ph.D., we do not have information on career interruptions after the degree. For men, the small negative effects of having children in 1979 and 1989 became larger and positive in 1995. Table 6–8 shows that having an interruption between the baccalaureate and the Ph.D. had a large positive effect for women, over 10 percentage points, with a smaller effect for men. Without additional information, we can only speculate on why this substantial effect occurs. One possibility is that interruptions for women are due to family obligations and that having these interruptions before the Ph.D. decreases the number of interruptions later in the career. For both men and women, these interruptions may also correspond to predoctoral research experience that makes postdoctoral fellowships less likely, thus speeding up the movement into faculty positions and eventually into more advanced academic ranks. Bayer and Astin (1975) found that career interruptions had a negative effect for women, but these were interruptions after the Ph.D. Zuckerman and Cole (Zuckerman and Cole 1975) suggested that these interruptions are due to familial obligations, which is consistent with our findings on the effect of having children. The table also shows a modest positive effect for both men and women for obtaining a degree from a more prestigious graduate program. While several studies have found that the effects of doctoral origins are insignificant for rank (Cole 1979:411; Hurlbert and Rosenfeld

OCR for page 123
From Scarcity to Visibility: Gender Differences in the Careers of Doctoral Scientists and Engineers TABLE 6–8 Effects of Time from Baccalaureate to Ph.D. and Prestige of Ph.D. on Adjusted Proportion Who Are Full Professors in the Twentieth Year of their Career   Women Men 1979 1989 1995 1979 1989 1995 More than 10 Years from Bachelor’s to Ph.D. 13.2 9.9 13.2 7.1 2.1 5.3 Prestige of Ph.D. –0.6 3.5 3.7 0.2 2.0 2.0 NOTE: Data were not available in 1973. 1992), Long, Allison, and McGinnis (1993) found a positive effect of doctoral origins on promotion to full professor. Summary on Academic Rank Many studies across many fields at different times using a myriad of control variables found evidence of substantial gender differences in academic rank (Ahern and Scott 1981; Astin and Bayer 1979; Cole 1979; Hurlbert and Rosenfeld 1992; Long, Allison and McGinnis 1993; Perrucci, O’Flaherty and Marshall 1983; Rosenfeld and Jones 1986, 1987; Sonnert 1990; Szafran 1984). Our results are consistent with these findings and provide evidence that gender differences in rank are found across fields and types of institutions. While overall percentages that do not control for any variables affecting rank show no improvement since 1979, controlling for age and other factors provides evidence of substantial improvement. However, gender differences still persist in this critical outcome for the academic career. A possible explanation for the remaining gender differences in rank attainment is that our analyses do not include controls for variables that past research has shown to affect rank. Rosenfeld and her colleagues (Hurlbert and Rosenfeld 1992; Rosenfeld and Jones 1986) found that rank advancement is related to institutional mobility and that women may have more constraints on their opportunities to change institutions. Bayer and Astin (1975) found a negative effect of time devoted to teaching, which is likely to more severely affect women than men. Last, and most importantly, we do not include measures of productivity. The information on productivity that we do have, which is discussed in the next section, is too aggregated over time to be used in predicting promotion.25 25   Since attaining rank leads to resources that enhance productivity, it is essential that publication data be for the period immediately before the decision for promotion or tenure is made.

OCR for page 123
From Scarcity to Visibility: Gender Differences in the Careers of Doctoral Scientists and Engineers However, past research suggests that productivity differences do not explain gender differences in promotion. Based on his own results and a review of the literature, J.Cole (Cole 1979:246) concluded: “Historically, productivity patterns simply will not explain the gender differences in academic rank. Other social and economic variables might explain these associations, of course, but in the absence of adequate data to test alternative hypotheses I tentatively conclude that there has been extensive sex discrimination in promotion opportunities over the past 40 years.” Later analyses by Long et al. (1993) included detailed, over time data on productivity and found that gender differences persisted after controls for productivity and many other variables. They concluded: “While these differences [in the rates of promotion for men and women] may be due to the exclusion of other variables, it is unclear what these variables might be. We believe that a more reasonable explanation is that women are expected to meet higher standards for promotion.” RESEARCH PRODUCTIVITY More than 50 studies in various fields show that women publish less than men. Moreover, correlations between gender and productivity have been roughly constant since the 1920s. The existence and stability of gender differences in productivity continue to be puzzling. —Jonathan R.Cole and Harriet Zuckerman, Advances in Motivation and Achievement, 198426 In a review article on gender differences in scientific productivity, Cole and Zuckerman (1984) estimated that men published 40 percent to 50 percent more than women. While our data are inadequate for a full analysis of factors determining gender differences in scientific productivity, we can provide some information that helps us to explain why the overall rate of productivity is greater for men than women. Past research has examined a large number of factors that may be determining the lesser productivity of women. These factors include ability, marriage and family, career interruptions, doctoral and postdoctoral training, the type and prestige of the academic employer, the organizational context of employment, processes of reinforcement, and discrimination. For detailed reviews, see Cole and Zuckerman (1984), Fox (1983), Long (1992), and Xie and Shauman (1998). The key to understanding the large observed gender differences in productivity is to control for the many differences between men and 26   Cole and Zuckerman (1984).

OCR for page 123
From Scarcity to Visibility: Gender Differences in the Careers of Doctoral Scientists and Engineers women in the types of positions and resources that they have. This approach was taken recently by Xie and Shauman (1998) who analyzed scientific productivity using national surveys from 1969, 1973, 1988, and 1993. Their first conclusion was that gender differences in productivity have declined. Second, they conclude that “gender differences in research productivity stem from gender differences in structural locations and as such respond to the secular improvement of women’s position in science.” That is, gender differences in productivity reflect differences in positions women have held, rather than differences in abilities or motivation. Given evidence of the larger effect of work context on productivity than of productivity on attaining a given position (see Allison and Long 1990 and the literature cited therein), the increasing entry of women into faculty positions in all types of institutions should lead to future decreases in gender differences in scientific productivity. Still, to the extent that differences in employment persist, differences in productivity can be expected to continue, albeit to a lesser degree. Figure 6–38 illustrates the degree to which gender differences in scientific productivity are associated with differences in the positions held by male and female scientists and engineers. Each bar indicates the percent more publications by the average male academic than the average female academic. The first bar considers all academic scientists and engineers in 1995 and shows that men have about 30 percent more publications than women. As we move to the right, we increasingly restrict the group of academics to make their characteristics more similar. Among those in Research I institutions, men are just under 25 percent more productive. However, earlier we showed that women were much more likely to have off-track positions which we would expect to be associated with lesser productivity. Restricting our comparison to only faculty we find that men are 13 percent more productive; among tenured faculty in Research I institutions, 8 percent more productive, and when comparing full professors in the life sciences, men are less than 5 percent more productive. Even with the limitations of our data, it seems clear that differences in structural position are a key factor in the lesser productivity of women in science and engineering. THE PRESENCE OF WOMEN IN ACADEMIC POSITIONS In prior sections we focused on changes in the relative proportions of men and women who have advanced to more secure and prestigious positions. By comparing the percent of all women who obtained a given status to the corresponding percent of all men, we are able to determine whether men and women have equal success in attaining each type of position. To the degree that such equity occurs, differences in the repre-

OCR for page 123
From Scarcity to Visibility: Gender Differences in the Careers of Doctoral Scientists and Engineers FIGURE 6–38 Percent more publications by the average man compared to the average woman in increasingly similar groups of academic scientists in 1995. NOTES: Moving from left to right, each group is a subset of the prior group. For example, “Tenure-Track Faculty” are full time in Research I institutions. sentation of men and women are a function of the smaller number of women in academia. As a way to summarize our findings, as well as to show the growing presence of women in the academy, we consider the changing percent of academics who are women. Overall, we find that while there have been dramatic increases in the presence of women in all types of academic positions, women remain well below half of the total in all categories of positions with academia. Figure 6–39 displays the percent of academic scientists and engineers who are women for various employment statuses within academia. Panel A shows the results for those employed in Research I universities; Panel B presents the results for all non-Research I institutions combined; and Panel C plots the difference in the percent female in non-Research I universities compared to Research I institutions, where positive values indicate a greater presence of women in non-Research institutions. There has been an increase in the percent of women in all categories of academic employment, ranging from off-track positions to being full professors. This trend is driven by the increasing number of women with Ph.D.s and the corresponding increase in the number of women in academia. This is shown by the leftmost set of bars, which gives the percent of all full-time academic scientists and engineers who are women, combining all types of full-time employment. In 1995 women were nearly 20 percent of all academics in Research I universities and nearly 25 percent of those in all other types of

OCR for page 123
From Scarcity to Visibility: Gender Differences in the Careers of Doctoral Scientists and Engineers FIGURE 6–39 Percent of academic scientists who are women, by type of institutions, type of jobs, and year of survey.

OCR for page 123
From Scarcity to Visibility: Gender Differences in the Careers of Doctoral Scientists and Engineers institutions. As we move to the right, we see that women are found in varying proportions among different types of academic positions. Women are found most often in the least prestigious, least secure, and most poorly paid off-track positions, such as research associates and temporary instructors. Among tenure-track faculty, women are found most often among assistant professors. At the critical rank of full professor, women in 1995 are still less than 10 percent of the full professors in Research I universities and just 12 percent in other types of schools. Panel C shows that the advance in the representation of women in academia has occurred more slowly in Research I universities than in other types of institutions. The five panels of Figure 6–40 show that women make up very different proportions of the academic labor force in different fields. The first set of bars in each column is the percent of all full time academics who are women. Across fields in 1995, the percent women among all full time academics ranges from 6 percent female in engineering to 31 percent in the social and behavioral sciences. These overall differences across fields are also found when we examine the proportion of women in specific types of positions, such as off-track or tenured positions. In large part, as would be expected, the representation of women within fields of academia is largely dependent on the number of women obtaining degrees in those fields. Overall, there has been substantial improvement in the presence of women in academia. In all fields, women made up a substantially larger proportion of the academic labor force in 1995 than in 1973. Large field differences persist, with women found least frequently in engineering and most often in the life sciences and the social/behavioral sciences.

OCR for page 123
From Scarcity to Visibility: Gender Differences in the Careers of Doctoral Scientists and Engineers FIGURE 6–40 Percent of scientists in given types of positions who are female, by field and year of survey. NOTE: See Appendix Table D-11 for further details.

OCR for page 123
From Scarcity to Visibility: Gender Differences in the Careers of Doctoral Scientists and Engineers Still, at most, women make up only about 33 percent of the academic labor force in any field. With the proportionally greater entry of women than men into academia in recent years, the average career age of women is less than that of men. This accounts for a substantial amount of the greater representation of men among those with tenure and those with the rank of full professor. However, controls for gender differences in age and field do not eliminate the greater presence of men among those on the tenure track, with tenure, or promoted to full professor. While the presence of women in academia has shown notable improvements, women remain underrepresented in academic science and engineering.