4
Labor Force Participation

The Attrition of Women from the S&E Labor Force

…receipt of a doctorate in S&E does not imply full participation in these fields.

—J.Scott Long and Mary Frank Fox, Annual Review of Sociology, 1995.1

INTRODUCTION

Increases in the number of women among new Ph.D.s do not translate directly into increases in the proportion of women in the science and engineering labor force. Each new cohort of Ph.D.s represents only a small fraction of the total number of scientists and engineers. This is shown in Figure 4–1, which compares the growth in the percent of women among new Ph.D.s, to increases in the labor force, and finally among those employed full time in S&E. The proportion of women among new Ph.D.s, shown by the darkest bars, increased by 20 percentage points from 13 percent in 1973 to 33 percent in 1995, while the proportion of women in the labor force increased more slowly from 8 percent in 1973 to 21 percent in 1995. The proportion of women in the S&E labor force must increase slowly as older, predominantly male cohorts retire and are replaced by new cohorts that have a greater proportion of women. Hargens and Long (2000) demonstrate how demographic factors, such as the gender composition of new cohorts and the age distribution of currently employed

1  

Long and Fox (1995).



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 63
From Scarcity to Visibility: Gender Differences in the Careers of Doctoral Scientists and Engineers 4 Labor Force Participation The Attrition of Women from the S&E Labor Force …receipt of a doctorate in S&E does not imply full participation in these fields. —J.Scott Long and Mary Frank Fox, Annual Review of Sociology, 1995.1 INTRODUCTION Increases in the number of women among new Ph.D.s do not translate directly into increases in the proportion of women in the science and engineering labor force. Each new cohort of Ph.D.s represents only a small fraction of the total number of scientists and engineers. This is shown in Figure 4–1, which compares the growth in the percent of women among new Ph.D.s, to increases in the labor force, and finally among those employed full time in S&E. The proportion of women among new Ph.D.s, shown by the darkest bars, increased by 20 percentage points from 13 percent in 1973 to 33 percent in 1995, while the proportion of women in the labor force increased more slowly from 8 percent in 1973 to 21 percent in 1995. The proportion of women in the S&E labor force must increase slowly as older, predominantly male cohorts retire and are replaced by new cohorts that have a greater proportion of women. Hargens and Long (2000) demonstrate how demographic factors, such as the gender composition of new cohorts and the age distribution of currently employed 1   Long and Fox (1995).

OCR for page 63
From Scarcity to Visibility: Gender Differences in the Careers of Doctoral Scientists and Engineers FIGURE 4–1 The percent of women among new Ph.D.s, among all scientists and engineers available to work in science and engineering, and among those working full time, by year of survey. NOTE: Labor force includes those who are not employed or working part time. scientists, limit the rate of change in the gender composition of the academic labor force. For example, even if women were three-fourths of the new Ph.D.s, it would take decades before gender parity would be achieved in the labor force. Obviously, with women making up far less than half of the new Ph.D.s, changes will be even slower. Second, after completion of the doctorate, a greater proportion of women than men do not attain full-time careers in science and engineering. While doctoral scientists and engineers have low rates of unemployment compared to the total U.S. labor force, women with doctorates are substantially more likely than men to be unemployed, employed part time, employed outside of S&E, or not be in the labor force. This is reflected by the white bars in Figure 4–1, which show that the percent of full-time workers in S&E who are women increased from 6.5 percent to 20 percent. Differences between men and women in labor force participation add up to less accumulated work experience and less valuable experience for women over the course of their careers, a factor that is important for understanding the gender differences in career outcomes that are described in later chapters. This chapter begins by examining the sex composition of the scientific

OCR for page 63
From Scarcity to Visibility: Gender Differences in the Careers of Doctoral Scientists and Engineers and engineering labor force in our five broad fields. How do men and women differ in the percent who are working full time in S&E occupations? Of those Ph.D.s who are not working full time in S&E, are they working outside of science, working part time, unemployed and looking for work, or out of the labor force? We find that while there has been improvement since 1973, female Ph.D.s continue to be substantially less likely than men to be fully employed in scientific and engineering occupations. Second, we examine the reasons given by scientists and engineers for working part time and look at characteristics of scientists that are associated with less than full employment. We find that lack of full participation for women is most strongly related to familial status. Finally, we summarize the effects of the underemployment of women by comparing the average number of years of professional experience for men and women. Overall during the past 20 years, 10 percent of the potential professional work of female doctorates has been lost as a result of women being less likely to be fully employed. Before proceeding, it is important to emphasize that participation in the full time scientific and engineering labor force includes a variety of jobs that differ greatly in prestige, security, and remuneration. This chapter considers only the question of whether a Ph.D. scientist is working full time in S&E, not the equally important question of what kind of work. In later chapters, we consider only those scientists and engineers with full-time employment in S&E and explore differences between men and women in sector of employment, work activity, position, and salary. A Note on Terminology Before proceeding, it is helpful to define several terms that are used in this chapter. Labor force is defined as Ph.D.s in S&E fields who are living in the United States and are under the age of 75. The labor force includes both part time workers and those who are unemployed. Full-time labor force is defined as members of the labor force who are working full time in some area of science or engineering. Employed outside of S&E includes doctoral scientists and engineers who are working full time in occupations that are not directly related to S&E as defined by the Survey of Doctorate Recipients. Unemployed includes those without jobs who are seeking work. Out of the labor force are those who are not working and not seeking work. Underemployment includes part time workers and the unemployed.

OCR for page 63
From Scarcity to Visibility: Gender Differences in the Careers of Doctoral Scientists and Engineers THE FULL-TIME SCIENTIFIC AND ENGINEERING LABOR FORCE2 As the growth in the production of new Ph.D.s slowed since 1973, the full labor force of Ph.D. scientists and engineers increased 220 percent from 187,236 in 1973 to 412,497 in 1995. During this period, the entry of increasingly female cohorts gradually changed the gender composition of the full-time labor force, from 6.5 percent female in 1973 to 19.6 percent in 1995 (Figure 4–2). But, as documented in the last chapter, the proportion of female Ph.D.s differs widely by field, which affects the sex composition of the full-time labor force within fields. While there has been substantial movement towards parity, the full-time participation of women in the labor force remains far below 50 percent in all fields. Engineering: While the number of male Ph.D. engineers working full time doubled from 30,208 in 1973 to 69,013 in 1995, the corresponding number of women increased by a factor of 40, from a mere 82 in 1973 to 3,589 in 1995. Still, by 1995 women were only 5 percent of the full-time labor force. As shown in Table 4–1, in 1995 women were least represented in the largest subfields of electrical and chemical engineering, and were most strongly represented in the smaller fields of industrial engineering with 15 percent women and materials sciences with 10 percent women. FIGURE 4–2 The percentage of the full-time scientific and engineering labor force that is female, by field and year of survey. 2   See Appendix Tables B-1 and B-2 for additional data.

OCR for page 63
From Scarcity to Visibility: Gender Differences in the Careers of Doctoral Scientists and Engineers Mathematical Sciences: The number of female doctoral mathematicians working full time in S&E grew from 677 in 1973 to 3,728 in 1995. This increase of 450 percent can be compared to a 121 percent increase in male mathematicians from 11,639 in 1973 to 25,711 in 1995. Overall, women represented 5.5 percent of the mathematicians in 1973, which increased to 13 percent in 1995. As shown in Table 4–2, in 1995 women were most strongly represented in statistics and probability with 18 percent of the full-time labor force and in the rapidly growing field of computer science where they were 13 percent. Physical Sciences. The proportional representation of women in the physical sciences is similar to that in mathematics. In 1973, there were 1,637 women, which corresponded to 3 percent of the total Ph.D.s in the physical sciences working full time in S&E. By 1995 the number of women increased by 480 percent to 9,505 or 10.5 percent of the full-time labor force. During this same period, the number of male physical scientists increased only 50 percent from 52,168 in 1973 to 81,373 in 1995. As shown in Table 4–3, the largest increases overall, as well as for women, were in the large subfields of physics and chemistry, which represented 31 percent and 51 percent of the full time Ph.D.s in 1995, respectively. The largest proportional representation of women is in chemistry, which grew from being 4 percent female in 1973 to 14 percent female in 1995. While the second largest number of women is in physics, with nearly 1,500 women in 1995, the growth was from a mere 1.3 percent of the full-time Ph.D. labor force being female in 1973 to a still small 5 percent in 1995. Life Sciences: Overall, women have a larger representation in the full-time labor force in the life sciences than the preceding broad fields. In 1973, there were 4,598 women, 9.5 percent of the total, compared to 44,053 men. The proportion of women increased to 26 percent in 1995, as the number of women grew six fold to 29,885, while the number of men only doubled to 85,098. As shown in Table 4–4, women are found most often in the subfield of biological sciences, where they were 49 percent of the full-time labor force in 1995. They were least represented in the agricultural sciences with only 1 percent of the total in 1973, increasing to 12 percent in 1995. Social and Behavioral Sciences. Women are most highly represented in the social and behavioral sciences. As the number of Ph.D.s working full time increased 150 percent from 43,298 in 1973 to 107,216 in 1995, the representation of women grew from 12 percent to 33 percent. As shown in Table 4–5, women are most highly represented in anthropology with 40 percent of the full-time labor force, followed by 39 percent in the large

OCR for page 63
From Scarcity to Visibility: Gender Differences in the Careers of Doctoral Scientists and Engineers TABLE 4–1 Numbers of Engineers Working Full Time in S&E, by Sex, Subfield, and Year of Survey   1973 1979 # Men # Women % Women # Men # Women % Women Biomedical 307 2 0.6 778 10 1.3 Chemical 4,835 13 0.3 6,696 33 0.5 Electrical 7,317 17 0.2 9,578 53 0.6 Industrial 720 4 0.6 1,007 12 1.2 Materials Science 146 0 0.0 599 12 2.0 Other 16,883 46 0.3 23,860 218 0.9 Total 30,208 82 0.3 42,518 338 0.8 TABLE 4–2 Numbers of Mathematicians Working Full Time in S&E, by Sex, Subfield, and Year of Survey   1973 1979 # Men # Women % Women # Men # Women % Women Computer Science 540 30 5.3 1,330 92 6.5 Probability and Statistics 2,233 130 5.5 3,710 318 7.9 Mathematics 8,866 517 5.5 10,871 780 6.7 Total 11,639 677 5.5 15,911 1,190 7.0 TABLE 4–3 Numbers of Physical Scientists Working Full Time in S&E, by Sex, Subfield, and Year of Survey   1973 1979 # Men # Women % Women # Men # Women % Women Astronomy 930 64 6.4 1,531 89 5.5 Physics 16,424 220 1.3 20,603 427 2.0 Chemistry 28,936 1,236 4.1 34,990 2,105 5.7 Oceanography 523 6 1.1 888 37 4.0 Geosciences 5,355 111 2.0 7,287 292 3.9 Total 52,168 1,637 3.0 65,299 2,950 4.3

OCR for page 63
From Scarcity to Visibility: Gender Differences in the Careers of Doctoral Scientists and Engineers 1989 1995 # Men # Women % Women # Men # Women % Women 1,314 84 6.0 1,729 135 7.2 9,410 257 2.7 9,574 504 5.0 12,595 213 1.7 16,422 601 3.5 901 75 7.7 1,590 279 14.9 1,587 138 8.0 3,304 366 10.0 32,631 791 2.4 36,394 1,704 4.5 58,438 1,558 2.6 69,013 3,589 4.9 1989 1995 # Men # Women % Women # Men # Women % Women 3,670 397 9.8 6,501 933 12.6 5,067 798 13.6 5,165 1,175 18.5 13,610 1,304 8.7 14,045 1,620 10.3 22,347 2,499 10.1 25,711 3,728 12.7 1989 1995 # Men # Women % Women # Men # Women % Women 2,310 175 7.0 2,578 202 7.3 25,577 982 3.7 26,977 1,473 5.2 39,788 4,426 10.0 39,773 6,251 13.6 1,336 132 9.0 1,537 212 12.1 9,857 809 7.6 10,508 1,367 11.5 78,868 6,524 7.6 81,373 9,505 10.5

OCR for page 63
From Scarcity to Visibility: Gender Differences in the Careers of Doctoral Scientists and Engineers TABLE 4–4 Numbers of Life Scientists Working Full Time in S&E, by Sex, Subfield, and Year of Survey   1973 1979 # Men # Women % Women # Men # Women % Women Agricultural 8,766 86 1.0 12,093 297 2.4 Medical 30,916 4,131 11.8 41,403 7,218 14.8 Biological 4,371 381 8.0 6,724 1,078 13.8 Total 44,053 4,598 9.5 60,220 8,593 12.5 TABLE 4–5 Numbers of Social and Behavioral Scientists Working Full Time, by Sex, Subfield, and Year of Survey   1973 1979 # Men # Women % Women # Men # Women % Women Psychology 16,048 3,060 16.0 24,690 7,005 22.1 Anthropology 1,226 246 16.7 2,193 756 25.6 Economics 7,359 376 4.9 9,261 781 7.8 Sociology 4,018 753 15.8 5,649 1,592 22.0 Other 9,434 778 7.6 14,058 1,820 11.5 Total 38,085 5,213 12.0 55,851 11,954 17.6 field of psychology and 36 percent in sociology. Women were least represented in the more mathematical field of economics, where they grew from 5 percent of the full-time labor force in 1973 to 15 percent in 1995. THE AGE STRUCTURE IN SCIENCE AND ENGINEERING The later entry of women into S&E is reflected in their younger professional age. Professional age is important for understanding career outcomes since years working in the profession affect the positions that scientists and engineers hold. For example, tenure and promotion in academia are related to the time a person has been in rank. If the average female faculty member is younger than the average male, proportionately fewer women would be full professors, all else being equal. Administrative positions and appointment to gatekeeping roles (e.g., editor of a journal) are also associated with age, with older scientists being more likely to

OCR for page 63
From Scarcity to Visibility: Gender Differences in the Careers of Doctoral Scientists and Engineers 1989 1995 # Men # Women % Women # Men # Women % Women 15,569 1,354 8.0 15,688 2,053 11.6 54,025 14,699 21.4 63,815 22,452 26.0 11,490 4,955 30.1 5,595 5,380 49.0 81,084 21,008 20.6 85,098 29,885 26.0 1989 1995 # Men # Women % Women # Men # Women % Women 34,961 16,818 32.5 36,317 22,833 38.6 2,914 1,441 33.1 3,419 2,283 40.0 11,907 1,654 12.2 11,659 2,063 15.0 6,575 2,717 29.2 6,255 3,536 36.1 18,501 4,251 18.7 14,633 4,218 22.4 74,858 26,881 26.4 72,283 34,933 32.6 hold these influential positions. Zuckerman and Merton (1972) discuss the many ways in which age, aging, and the age structure affect the scientific career. Age structure is determined by the size of new cohorts of Ph.D.s relative to the size of past cohorts and the rate at which each cohort leaves S&E through retirement, death, and migration to other occupations. If each new cohort were the same size, the age structure would be nearly uniform, with approximately the same number of people at each age until retirement. However, the number of new Ph.D.s grew rapidly after World War II and in response to Sputnik, with slower growth in the 1970s and 1980s. During these periods of slower growth, the proportion of women in each new cohort grew most rapidly. These trends resulted in different age structures for men and women. Population pyramids are the standard method for examining the age structure in a population. A population pyramid compares the age distri-

OCR for page 63
From Scarcity to Visibility: Gender Differences in the Careers of Doctoral Scientists and Engineers TABLE 4–6 Mean Years Since Ph.D., by Field, Sex, and Year of Survey   Combined­ Fields Engineering Mathematics Men Women Men Women Men Women 1973 11.4 10.0 8.8 7.4 9.9 10.3 1979 12.5 9.3 11.3 6.4 11.6 9.6 1989 15.4 10.4 14.9 7.1 15.3 11.7 1995 16.8 11.4 15.1 7.2 15.8 11.6   Physical Sciences Life Sciences Social/Behavioral Sciences Men Women Men Women Men Women 1973 13.0 11.4 12.2 10.5 10.8 9.1 1979 14.4 11.4 12.7 10.0 11.4 8.3 1989 16.9 11.5 14.9 10.4 14.8 10.2 1995 18.3 11.3 16.6 11.3 17.2 11.9 butions for two groups by showing the percent of individuals within each group who is a given age. For our purposes, we consider professional age, which is defined as the number of years since the receipt of the doctorate. Panels A and B of Figure 4–3 are population pyramids for male and female scientists and engineers in the S&E labor force in 1973 and 1995. Each figure contains two histograms. The histogram on the left shows the percent of men in each three-year age group; the histogram on the right shows corresponding data for women. Note that the sum of the bars in each histogram will equal 100. Comparing the plots for men and women in 1973, we see that a greater proportion of women is younger, with a median professional age of seven compared to a median age of nine for men. The shape of the distribution resulted from the proportionately smaller number of women receiving Ph.D.s before 1966 (corresponding to professional age 6), the greater inflow of women after 1965, and the substantial attrition of women who entered S&E before World War II. In 1995, shown in Panel B, the decreasing size of new cohorts of men resulted in an increase in the median professional age of men from 9 to 16. For women, the median age increased only from 7 to 9, since the sizes of new cohorts of women were larger relative to the total number of women in science and engineering. The percent of men at each age 1 through 27 is nearly uniform, while the distribution for women has a triangular shape corresponding to the in

OCR for page 63
From Scarcity to Visibility: Gender Differences in the Careers of Doctoral Scientists and Engineers FIGURE 4–3 Distribution of professional ages in the science and engineering labor force, by sex and year. NOTE: Professional age is defined as the number of years since the receipt of the doctorate.

OCR for page 63
From Scarcity to Visibility: Gender Differences in the Careers of Doctoral Scientists and Engineers FIGURE 4–11 Reasons for part-time employment, by sex, and year of survey. NOTE: Respondents can choose more than one response. FIGURE 4–12 Percent who cite family reasons for working part time, by sex and year of survey. NOTE: Values are moving averages across 5-year periods.

OCR for page 63
From Scarcity to Visibility: Gender Differences in the Careers of Doctoral Scientists and Engineers Factors Associated with Labor Force Participation6 Unfortunately, the SDR only asked respondents the reasons for part-time employment. To consider factors affecting other types of less than full employment, we must use statistical models to determine the degree to which characteristics of an individual, such as marital status and Ph.D. origins, are associated with labor force status after controlling for variables such as broad field and Ph.D. cohort. The effects of these control variables are not discussed since they were considered earlier in the chapter. Marriage and Family Past research has often found only limited effects of marriage and family on the careers of women in science and engineering. However, most of this research studied only women who were fully participating in S&E. But, for example, to say that the success of women who are full professors at research universities is not affected by family obligations does not imply that marriage and family did not affect their chances of staying in the labor force and thus having a chance to become a full professor. Conversely, women who do not become faculty at research universities may have had their career choices limited by familial status. Indeed, our analyses show that there is a strong association between marriage, children, and whether a female doctorate is less than fully employed, while there are only small effects for men. These results are now considered. Overall, marriage and family are the most important factors differentiating the labor force participation of male and female scientists and engineers. Figure 4–13 plots the percent of male and female scientists and engineers in 1995 who are predicted by our model to work full time either in or out of S&E according to familial status.7 Four familial statuses are considered: single without children (black bar), married without children (dark gray bar), married with child one or more between the ages of 7 and 18 living at home (light gray bar), and married with one child or more younger than 7 living at home (white bar). Other statuses, such as di- 6   See Appendix Table B-12 for additional data. Analyses are based on multinomial logit analyses of the five categories: full-time employment in S&E, full-time employment outside of S&E, part-time employment, unemployed and seeking work, and unemployed and not seeking work. Technical details are given in Chapter 2. 7   Predictions are for an individual who is average on all of the variables in the model. See Chapter 2 for details.

OCR for page 63
From Scarcity to Visibility: Gender Differences in the Careers of Doctoral Scientists and Engineers FIGURE 4–13 Predicted percent with full-time employment in 1995, by sex and familial status. voiced with children, had too few cases to evaluate. Among men, those who are single are least likely to be working full time, with small increases for married men and those with children. In contrast, single women are most likely to be working full time. While being married slightly increases a man’s chances of full-time work, being married without children decreases the predicted proportion of women working full time by 5 percentage points. Having older children at home decreases the proportion by 8 more points, while being married with young children decreases the proportion with full-time employment by 22 points. It is interesting to note that as a consequence of the opposite effects of marriage and children for men and women, an identical 94 percent of single men and single women are expected to be working full time. That is, differences between men and women in labor force participation are eliminated if we compare single men to single women. Figure 4–14 shows that for women there were some changes in the effects of marriage and family over time.8 The percent of women with young children who are predicted to work full-time increased by nearly 10 points between 1979 and 1989, with only a tiny increase in 1995. The predicted percent working among those with older children at home increased by over 3 points between each survey. Figure 4–15, plots differ- 8   Data on marriage and family are not available for 1973.

OCR for page 63
From Scarcity to Visibility: Gender Differences in the Careers of Doctoral Scientists and Engineers FIGURE 4–14 Predicted percent of women with full employment, by year of survey. FIGURE 4–15 Differences in predicted labor force status between single women and married women with young children, by year of survey.

OCR for page 63
From Scarcity to Visibility: Gender Differences in the Careers of Doctoral Scientists and Engineers ences between single women and married women with small children in the predicted proportion in each category of labor force participation. The biggest difference is in the proportion working full time. In 1979 women with small children were over 30 points less likely to be working full time compared to single women. This difference dropped to 22 points by 1995. The next two categories, part time work and not seeking work, account for the lesser full time work of women with young children. Women with children are nearly 15 points more likely to be working part time, with a slight drop in 1995. The next most likely work status is not seeking work. In 1979 women with young children were nearly 15 points more likely to be in this category, a difference that was reduced to around 10 points by 1995. There is very little effect of having young children for not working and seeking work. Similar analyses for men (not shown) found only the small effects of marriage and family. Overall, family has a significant effect on labor force status, but the effect appears to be gradually declining. Baccalaureate Origins The effects of baccalaureate origins on labor force participation were examined by including the Carnegie type of the baccalaureate institution. For men, we found no effect, but for women there was a small, negative association between receiving the bachelor’s degree from an exclusively undergraduate institution (e.g., a small liberal arts college) and working full time. Women with degrees from baccalaureate-only institutions were 2 percentage points less likely to be working full time in S&E occupations and roughly 2 points more likely to be working part time, after controlling for other factors. These effects were slightly smaller in 1995. It is possible that this represents differences in critical socialization to the scientific career that would be obtained with an undergraduate degree from an institution with research activities at the graduate level. On the other hand, it may reflect someone with lower career aspirations. We also examined whether receiving an undergraduate degree from a women’s college or university affected labor force participation. While some past research has suggested that attending a women’s college greatly increased a woman’s chances of being highly successful (Tidball and Kistiakowsky 1976), we found no effect on labor force participation. Elapsed Time from Baccalaureate to Ph.D. Logit analyses show that the time elapsed between the baccalaureate and the Ph.D. affected the labor force participation of scientists and engineers. Our measure of elapsed time is the difference between the date of

OCR for page 63
From Scarcity to Visibility: Gender Differences in the Careers of Doctoral Scientists and Engineers the undergraduate degree and the date of the doctorate, thus combining time enrolled in a graduate program and interruptions such as predoctoral employment or raising a family. Past research suggests that individuals with elapsed times of more than 10 years have interrupted their education (Tuckman, Coyle, and Bae 1990). Accordingly, elapsed time was entered into the analyses as a binary variable indicating whether it took more than 10 years to complete the doctorate—that is, indicating whether it is likely that the education was interrupted. Using data from 1995, Figure 4–16 plots the changes in the proportion of scientists and engineers predicted to be in each labor force status if the education was interrupted. The effects are in the same direction for men and women, but are larger for women. Individuals with interruptions are 5 to 7 percentage points less likely to be working full time in S&E, which is offset by a greater likelihood of working outside of S&E or working part time. Since interruptions between the bachelor’s degree and the Ph.D. may involve work outside of S&E (e.g., a social worker who goes back for a Ph.D. in psychology), it is possible that these men and women returned to their original employment after the completion of the degree. It is also possible that those who take longer to complete the degree are either less FIGURE 4–16 Changes in labor force status if elapsed time between the baccalaureate and Ph.D. was more than 10 years, by sex for 1995.

OCR for page 63
From Scarcity to Visibility: Gender Differences in the Careers of Doctoral Scientists and Engineers committed to their careers or alternatively that they are equally committed but not viewed by employers as being as as committed as their peers who complete the degree more quickly. YEARS OF WORK EXPERIENCE9 The net effects of workforce participation can be summarized by the number of years that a scientist or engineer has spent working since the receipt of the doctorate. The impact of the greater time out of the full-time labor force for women is shown in Figure 4–17 for 1979 and 1989. Comparable data for 1973 and 1995 are not available. The lines plot the difference between the mean years of professional experience for men and women by the number of years since the Ph.D. For example, a value of 1.5 means that an average woman worked 1.5 years less than an average man. In 1979, for every year since the degree the female scientist worked 0.12 years less than an average male scientist; in 1989 this loss was reduced to 0.09, with the largest improvements being noticed between the 9th and 15th years of the career. Overall, compared to male scientists, nearly 10 percent of the potential work experience of female scientists is being lost. FIGURE 4–17 Difference between men and women in mean years of work experience by years since the receipt of the Ph.D., by year of survey. NOTE: Values are 3-year moving averages. 9   See Appendix Tables B-13-B-14 for additional data.

OCR for page 63
From Scarcity to Visibility: Gender Differences in the Careers of Doctoral Scientists and Engineers A possible explanation for gender differences in professional work experience is that women are taking more time out of the workforce to care for their families. This explanation is consistent with many studies that have shown that within society as a whole women are more likely to be the primary caregivers (e.g., Moen 1985) and consequently to be absent from the paid labor force while they are working as caregivers. While we do not have information on the reasons for interruptions in the career, we do know an individual’s familial status at the time of the survey. Panels A and B of Figure 4–18 plot the difference between the mean years of work experience for male and female scientists according to their familial status, with the horizontal axis indicating years since the Ph.D. was received. In 1979, there was no difference in average work experience between single men and single women. For those who were married without children at home, the differences were small with some increase later in the career. But, for those who have children at home, the differences were much larger with an average woman with young children losing three-tenths of a year of work compared to men with young children for each year since the doctorate. Panel B shows that there were some decreases in the effects between 1979 and 1989, possibly reflecting increasing trends for women with families to remain in the labor force. Still, for married women and especially for those with young children, there are substantial losses in years of experience. SUMMARY AND CONCLUSIONS From 1973 to 1995 there were significant advances in the entry of women into S&E, but their advances did not completely overcome gender inequalities. While the representation of women in the S&E labor force is increasing in all fields, a greater proportion of women than men are not working full time. One way to summarize these differences, as well as to demonstrate the improvements since 1973, is to compare the percent of women working full time to the percent of men. Figure 4–19 plots the ratio of female and male rates of full-time employment. For example, if both 80 percent of men and 80 percent of women were working full time, the ratio would be at .8/.8=1. Values less than 1.0 indicate that women are less likely to be working full time. The figure shows that there were large improvements between 1973 and 1979, followed by gradual improvement through 1995. As shown earlier in the chapter, the net effect of the lower labor force participation of women is that as late as 1989 nearly 10 percent of the potential work activity of female scientists and engineers was lost as a result of their less than full employment. Among the variables we consider, by far the strongest factor that affects a female scientist’s labor force participation is familial status. While

OCR for page 63
From Scarcity to Visibility: Gender Differences in the Careers of Doctoral Scientists and Engineers FIGURE 4–18 Difference between mean work experience of men and women, by familial status, years since the Ph.D., and year of survey. NOTE: The vertical axis is the difference in the mean years for women and the mean years of experience for men.

OCR for page 63
From Scarcity to Visibility: Gender Differences in the Careers of Doctoral Scientists and Engineers FIGURE 4–19 Ratio of the percent of women who are working full time in S&E to the percent of men who are working full time, by field and year of survey. NOTE: There are too few women in engineering in 1973 to compute statistic. single women have rates of working full time that are identical to those of single men, being married and having children substantially reduces labor force participation for women but not for men. As summarized in Figure 4–20, married women are less likely to be working full time, women with older children are even less likely, and women with young children at home have rates of only around 70 percent. Implications for Later Chapters It is essential to understand the implications of our findings on labor force participation for the analyses of career outcomes in Chapters 5, 6, and 7. Those chapters examine the type of employer and work activity of scientists and engineers who are fully employed. Thus, even if the results in later chapters found no gender differences in career outcomes (which they do not find), there would still be important differences between male and female scientists and engineers in their success in moving into full-time employment. The less frequent full-time employment of women that is documented in the current chapter represents a significant loss of highly trained and talented scientific personnel, a loss that is not reflected in later chapters.

OCR for page 63
From Scarcity to Visibility: Gender Differences in the Careers of Doctoral Scientists and Engineers FIGURE 4–20 Predicted percent of women with full-time employment, by marital status and year of survey.