Climbing the Academic Ladder: Doctoral Women Scientists in Academe: A Report to the Office of Science and Technology Policy (1979)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

CHAPTER 4 ACADEMIC EMPLOYMENT The Emplovment_Patterns of Women Scientists In this chapter we will be concerned with recent trends in the employment of women scientists as faculty members with particular attention to the rank of assistant professor where major new developments would be expected to occur first. Since research universities rarely appointed women to their science faculties in the past, we do not yet expect to find many women in their senior faculties. If equal opportunity policies are observed, we do expect to find women proportionally represented among newly hired junior faculty members, and we expect to find them being paid and promoted at the same rate as men. As shown in Table 4. 1, women doctoral scientists are less likely than men to be employed in industry and are more heavily represented in higher education. Within higher education, they are more likely to teach than men (Table 4.2). They are also far more likely than men to be found in the lower ranks; roughly two-thirds of male faculty are associate or full professors while only one-third of women faculty are at that rank. The distribution of women faculty is more skewed in the top institutions (by R6D expenditures) than in the others (Table 4.3 and Figure 4.1). Our main concern in this section is to assess changes over the last few years in the traditional pattern. Because the proportions of women faculty vary widely among fields and their numbers are extremely small in some disciplines, any generalized analysis is of dubious utility. Set against the backdrop of a nearly steady-state academic economy and a sharply declining one in some fields, even a slight relative improvement in the status of women on science faculties can be regarded as a very welcome sign of progress. 57

TABLE 4. 1 Percent of Employed* Doctoral Scientists and Engineers by Employment Sector and Sex, 1977 Total 267,206 Wen 242,913 Women 24,293 Number Employed Educational Institutions 56.22 55.1 67.4% 4-Year Colleges/Univ. 53.9 53.2 61.3 2-Year Colleges 1.6 1.4 3.5 Elem/Secnd Schools 0.6 0.5 2.5 Business & Industry 26.4 27.9 11.6 Federal Government 8.4 8.8 4.9 Other Government 1.9 1.8 3.2 Hospitals & Clinics 2.9 2.5 7.3 Nonprofit Organizations 2.9 2.8 4.0 Other Employers 0.8 0.7 1.0 Employer Not Reported 0.5 0.4 0.7 * Excludes postdoctoral appointees. Source: Survey of Doctorate Recipients, National Research Council. The statistics in this table are weighted estimates, derived from a sample survey of 65,000 Ph.D's in science and engineering. The estimates are subject to two types of error -- sampling and non-sampling, (e.g., non-response bias). A discussion of the survey is provided in Appendix D. 58

TABLE 4.2 Employment Sector and Primary Work Activity of Employed Doctoral Scientists and Engineers, Excluding Postdoctoral Appointees, by Sex, 1973-1977 Total Men Women 1973 1977 1973 1977 1973 1977 All Sectors & Activities 209,808 267,206 194,506 242,913 15,302 24,293 Educational Inst. 57.8% 56.22 56.8% 55.1% 71.4% 67.4% Research 12.6 13.1 12.5 13.1 15.1 13.9 Teaching 36.4 32.2 35.7 31.4 45.8 39.9 Administration 6.2 7.2 6.2 7.3 5.4 6.6 Other 2.6 3.7 2.4 3.3 5.1 7.0 Federal Government 9.3 8.4 9.5 8.8 6.0 4.9 Research 4.8 4.0 4.9 4.2 3.0 2.5 Administration 3.4 3.2 3.6 3.4 1.5 1.5 Other 1.1 1.2 1.1 1.2 1.4 1.0 Business & Industry 25.2 26.4 26.5 27.9 8.4 11.6 Research 11.2 11.7 11.9 12.5 3.0 3.7 Administration 9.4 8.9 10.0 9.7 1.1 1.4 All Other 4.5 5.8 4.6 5.7 4.3 6.5 All Other Employers 7.7 9.0 7.2 8.3 14.2 16.1 Research 2.8 2.7 2.7 2.6 4.0 3.7 Administration 2.5 2.7 2.5 2.6 3.0 3.7 All Other 2.4 3.6 2.1 3.0 7.2 8.7 Source: Survey of Doctorate Recipients, National Research Council. The statistics in this table are weighted estimates, derived from a sample survey of 65,000 Ph.D.'s in science and engineering. The estimates are subject to two types of error, sampling and non-sampling, (e.g., nonresponse bias). A discussion of the survey is provided in Appendix D. 59

TABLE 4.3 Percent Distribution of Doctoral Scientists and Engineers in Academe* by Faculty Rank, R&D Expenditures of Employment Institution*, and Sex, 1977 Top 25 Inst. by R&D Men Women Second 25 Inst. by R&D Men Women Other Institutions Men Women No . Empl oyed , incl Postdocs. 17,664 1,979 14,116 1,384 92,155 12,392 Professor 44.7% 9.8% 42.4% 13.4% 37.0% 16.0% Assoc. Prof. 19.5 14.6 24.7 18.4 30.3 24.3 Asst. Prof. 15.0 31.0 20.0 36.0 23.4 38.4 Inst./Lect. 0.9 6.8 1.4 4.9 2.0 5.7 Other/No Report 9.5 18.2 5.4 12.8 3.7 7.9 Postdocs. 10.4 19.6 6.1 14.5 3.5 7.7 * Includedare two-year and four-year colleges, universities, and medical schools, + See Appendix B-l for a description of the ranking of institutions by federal R&D expenditures. NOTE: Percents may not add to 100.0% due to rounding. Source: Survey of Doctorate Recipients, National Research Council. The statistics in this table are weighted estimates, derived from a sample survey of 65,000 Ph.D's in science and engineering. The estimates are subject to two types of error -- sampling and non-sampling, (e.g., non-response bias). A discussion of the survey is provided in Appendix D. 60

FIGURE 4.1 Faculty Rank Distribution of Doctoral Scientists and Engineers by R&D Expenditures of Institution* and Sex, 1977. 40 — Top 25 Institutions 30 20 10 f — ww////. I J _ ~ . — , Second 25 Institutions " 40 — rra 30 — i — c 8 20 — rrr 1 s. wmm 1 10 — ^i rl n$_ 1 40 — Other Institutions 30 20 10 n | | Men E%fl Women W//////MW////M — — ] Professor Associate Assistant Instructor/ Other/ Postdoctoral Professor Professor Lecturer No Report 'See Appendix B-1 for a description of the ranking of institutions by federal R&D expenditures. SOURCE: Survey of Doctorate Recipients, National Research Council.

Rank vs. Tenure_as Definitions of Position Most academic demographic data (American Association of University Professors, National Center for Education Statistics, status reports on various professions) use rank as a convenient definition of position, but rank does not necessarily indicate anything about job security. Traditionally, the ranks of associate and full professor have been taken to be tenured, and assistant professorships were considered "tenure-track." These definitions are now misleading. At present, many assistant professorships are short-term appointments, for one to three years, and clearly identified (at least to the incumbent) as not leading to consideration for tenure. An analysis of 164 faculty position announcements, chosen at random from those received by Higher Education Resource Services in August and September 1978, showed less than half to be tenure-track. Seventy-three positions or 45 percent were identified as tenure-track or were described as "senior appointment"; 52 (32 percent) were described as non- renewable term, one- or two-year term with one possible renewal, or non-tenure track, with the remainder not clearly categorized but suggestive of imperraanence with such descriptions as "continuation contingent on funding." Announcements came from both departments and affirmative action offices, were also widely advertised in professional publications and are believed to constitute a valid national sample. Associate professor rank may or may not carry tenure; typically only about three fourths of such appointments do. The conferral of tenure is the only long-term employment guarantee an institution makes to a faculty member. Having expressed this caveat, because rank is such a widely used measure of position, however, we have used it, as well as tenure status, to compare men and women faculty. Changes in_gex_Distributjon of Faculty Positions Table 4.4 shows changes in the distribution of women faculty by rank and field over the last four years at institutions grouped by RSD expenditures. At the 25 top- ranked institutions, there appears to be a small increase in the proportion of women full professors, although the change is not statistically significant, and an increase from 138 to 194 in their actual numbers for all fields combined. The increases in the life sciences are positive, and may represent an upgrading of in-house candidates. In engineering, mathematics, and physical sciences (EMP fields), women were only one percent of the full professors in 1977, despite an apparent increase in their numbers. At 62

the associate level, the estimated proportions of women show an increase from 6 to 8 percent across all science fields, although due to small numbers, this is not statistically significant. At the assistant level, though, the trend was clearly up. The second 25 institutions displayed a similar pattern, with however a statistically significant overall increase at the full professor rank, due to appointments of women in the life sciences. At assistant rank, the proportions again notably increased between 1973 and 1977. The "other" institutions displayed no overall growth in the percentages of women among full professors, although women are still more highly represented here than in the top 50 institutions. There was a marked increase in the proportions of women among assistant professors. Overall, in 1977 the picture was one of higher proportions of women senior faculty at the "other" institutions, with some gains, however, being made at the first 50. In the sciences as a group, there has been moderate growth with total faculty positions (assistant professor rank and above) increasing by 21.5 percent in four years, from 1973 to 1977, for an average growth rate of 5.4 percent per year (Table 4.5). When institutions are grouped by R&D expenditures, however, only 5.5 percent of the total growth has occurred in the top 25 institutions, 2.9 percent in the second 25, and 91.6 percent in all others. Women account for 21.3 percent of the overall increase in faculty positions, a slightly higher figure than their average share of recent doctorates (see Table 2.1). Women on science faculties increased about three times faster than total faculty growth between 1973 and 1977. The more detailed dimensions and the uneven character of the "academic depression" are summarized in Tables 4.5 and 4.6. In the top 25 universities, nearly 70 percent of the entire faculty increase is in the social sciences; mathematics, medical sciences, and biosciences have all declined substantially in this group (Table 4.6). Within the BMP fields there are notable differences which deserve comment. Table 4.6 indicates that, in the top 25 institutions, the proportions of women on mathematics, physics, and chemistry faculties have not changed materially, whereas departments of earth sciences and engineering in this top category have made four-fold increases; however, because of the small numbers of women faculty then and now in the EMP fields, the differences are not statistically significant. If we examine engineering, mathematics, and the physical sciences at institutions of differing rank, as in Table 4.7, we note a lower-than-average growth rate in all 63

TABLE 4.4a Number and Percent of Women Doctoral Scientists and Engineers in Faculty Positions, by R & D Expenditures of Institution, Field, and Sex, 1973 and 1977. Top 25 Institutions Professor Associate Assistant All Science/Eng. 1973 1977 1973 1977 1973 1977 Fids. Total No. 7,460 8,085 3,464 3,741 3,206 3,264 No. Women 138 194 205 289 368 614 % Women 1.8 2.4 5.9 7.7 11.5 18.8 (+0.4) (10.5) (11.0) (+1.2) (12.9) (12.0) Engr. ,Math,Phys. Sc. Total No. 3,104 3,671 1,426 1,439 1,170 1,141 No. Women 22 28 12 29 50 v 92 % Women 0.7 0.8 0.8 2.0 4.3 8.1 Life Sciences (10.4) (10.4) (10.6) (11 . 1 ) (11.6) (12.3) Total No. 2,295 2,089 1,180 1,196 1,105 868 No . Women 50 88 100 104 153 216 % Women 2.2 4.2 8.5 8.7 13.8 24.9 (10.7) (+1.1) (+1.9) (+2.1) (+2.4) (+3.9) Behav. & Social Sc. Total No. 2,061 2,325 858 1,106 931 1,255 No. Women 66 78 93 156 165 306 % Women 3.2 3.4 10.8 14.1 17.7 2i.4 (+1.1) (+1.1) (+2.7) (+3.0) (+3.4) (+3.8) *See appendix B-l for a description of ranking of institutions by federal R&D expenditures. Note: Estimated sampling errors associated with the percent statistics are shown in parentheses. Source: Survey of Doctorate Recipients, National Research Council. The statistics in this table are weighted estimates derived from a sample survey of 65,000 Ph.D's in science and engineering. The estimates are subject to two types of error -- sampling and non-sampling, (e.g., nonresponse bias). A discussion of the survey is provided in Appendix J.

TABLE 4.4b Number and Percent of Women Doctoral Scientists and Engineers in Faculty Positions, by R & D Expenditures of Institution, Field, and Sex, 1973 and 1977. Second 25 Institutions Professor Associate Assistant All Science/Eng.Flds. 1973 1977 1973 1977 1973 1977 Total No. 5,919 6,168 3,631 3,746 3,122 3,319 No. Women 102 185 199 255 265 498 % Women 1.7 3.0 5.4 6.8 8.5 15.0 Engr. , Math, Phys. Sc. (+0.4) (+0.6) (+0.9) (+1-2) (+1.3) (+1.8) Total No. 2,479 2,773 1,494 1,478 1,509 1,190 No. Women 13 27 21 23 52 79 % Women 0.5 1.0 1.4 1.6 3.4 6.6 Life Sciences (+0.4) (+_0.5) (+0.8) (+1.0) (+1.3) (+2.1) Total No. 2,177 1,975 1,320 1,238 988 1,152 No. Women 37 87 83 110 103 135 % Women 1.7 4.4 6.3 8.9 10.4 11.7 Behav. & Social Sc (+0.7) (+1.2) (+1.6) (+2.2) (+2.3) (+2.6) Total No. 1,263 1,415 867 1,030 625 977 No. Women 52 71 95 122 110 284 % Women 4.1 5.0 11.0 11.8 17.6 29.1 (+1-6) (+1.7) (+2.8) (+3.1) (+4.1) (+4.3) *See Appendix B-l for a description of ranking of institutions by federal R&D expenditures Note: Estimated sampling errors associated with the percent statistics are shown in parentheses. Source: Survey of Doctorate Recipients, National Research Council. The statistics in this table are weighted estimates derived from a sample survey of 65,000 Ph. D's in science and engineering. The estimates are subject to two types of error -- sampling and non-sampling, (e.g., nonresponse bias). A discussion of the survey is provided in Appendix D. 65

TABLE 4.4c Number and Percent of Women Doctoral Scientists and Engineers in Faculty Positions, by R & D Expenditures of Institution, Field, and Sex, 1973 and 1977. Other Institutions Professor Associate Assistant 1973 1977 All Science/Eng. Fids 1973 1977 1973 1977 Total No. 28,610 36,103 23,930 30,927 21,789 26,325 No. Women 1,563 1,988 1,868 3,013 2,714 4,759 % Women 5.5 5.5 7.8 9.7 12.5 18.1 Engr. , Math, Phys. Sc (+0.3) (+0.3) (+0.4 ) (+0.5) (+0.6) (+.0.7) Total No. 11,517 14,923 10,045 12,303 8,383 8,397 No. Women 338 429 386 516 468 733 % Women 2.9 2.9 3.8 4.2 5.6 8.7 Life Sciences (+0.4) (+0.4) (+0.5) (+0.5) (+0.7) (+0.9) Total No. 8,877 10,984 6,909 8,865 6,375 8,346 No. Women 591 668 687 1,034 895 1,629 % Women 6.7 6.1 9.9 11.7 14.0 19.5 Behav. & Social Sc (+0.6) (+0.6) (+0.9) (+0.9) (+1.1) (+1.1) Total No. 8,216 10,196 6,976 9,759 7,031 9,582 No . Women 634 891 795 1,463 1,351 2,397 % Women 7.7 8.7 11.4 15.0 19.2 25.0 (+0-8) (+0.8) (+1.0) (+1.1) (+1.3) (+.1.4) *See Appendix B-l for a description of ranking of institutions by federal R&D expenditures. Note: Estimated sampling errors associated with the percent statistics are shown in parentheses. Source: Survey of Doctorate Recipients, National Research Council. The statistics in this table are weighted estimates derived from a sample survey of 65,000 Ph. D's in science and engineering. The estimates are subject to two types of error -- sampling and non-sampling, (e.g., nonresponse bias). A discussion of the survey is provided in Appendix D. 66

TABLE 4.5 Increase in Doctoral Scientists and Engineers in Faculty* Positions by R&D Expenditures of Institution+ and Sex, 1973-1977 Total Number Women as % Science** of of 1973-1977 Faculty Women Increase 4-Yr Growth 21,825 (21.5%) 4,655 21.3% Total 1st 50 Inst. 1977 28,727 2,109 1973 26,894 1,277 4-Yr Growth 1,833 ( 6.8%) 832 45.4% Top 25 Inst. 1977 15,352 1,129 1973 14,153 711 4-Yr Growth 1,199 (8.5%) 418 34.9% Second 25 Inst. 1977 13,375 980 1973 12,741 566 4-Yr Growth 634 (5.0%) 414 65.3% Other Inst. 1977 94,503 9,983 1973 74,511 6,160 4-Yr Growth 19,992 (26.8%) 3,823 19.1% Total All Inst.++ 1977 123,230 12,092 1973 101,405 7,437 * Faculty includes professor, associate professor, and assistant professor ranks. + See Appedix B-l for a description of ranking of institutions by federal R&D expenditures. ** Fields included are mathematics, computer sciences, physics/astronomy, chemistry, earth sciences, engineering, agricultural sciences, medical sciences, biological sciences, psychology, anc• social sciences. ++ Includes two-year and four-year colleges, universities, and medical schools. Source: Survey of Doctorate Recipients, National Research Council. The statistics in this table are weighted estimates, derived from a sample survey of 65,000 Ph.D's in science and engineering. The statistics are subject to two types of error -- sampling and non-sampling, (e.g., non-response bias). A discussion of the sur- vey is provided in Appendix D. 67

TABLE 4.6 Increase in Doctoral Scientists and Engineers in Faculty* Positions at 25 Leading Institutions'1- by Field and Sex, 1973-1977 Men and Women Women 1973 Faculty Net Gain, 1973-1977 1973 Faculty Net Gain, 1973-1977 Mathematics** 1,394 -219 (-16%) 42 [-10 (-24%)] Physics /Astronomy 1,208 53 ( 4%) 20 [12 ( 60%3 Chemistry 668 68 ( 10%) 11 L 3 ( 27%3 Earth Sciences 619 94 ( 15%) 6 27 (450%) Engineering 1,811 393 ( 22%) 5 [20 (400%)] Agricultural Sciences 822 221 ( 27%) 14 [ 9 ( 64%0 Medical Sciences 806 -187 (-23%) 66 40 ( 61%) Biological Sciences 2,967 -400 (-13%) 223 76 ( 34%) Psychology 1,299 46 ( 4%) 154 85 ( 55%) Social Sciences 2,559 834 ( 33%) 170 140 ( 82%) *Faculty includes professor, associate professor, and assistant professor ranks. +The top 25 institutions by R&D expenditures in FY 1976 are included. See Appendix B-l for a listing of the institutions. **The apparent decrease in mathematics faculty may be in part due to a redefini- tion of departments, separating applied mathematics, statistics, or computer sciences from pure mathematics. []Brackets indicate that the apparent net gain or loss between 1973 and 1977 is not statistically significant. Source: Survey of Doctorate Recipients, National Research Council. The statistics in this table are weighted estimates, derived from a sample survey of 65,000 Ph.D's in science and engineering. The statistics are subject to two types of error -- sampling and non-sampling (e.g., non-response bias). A discussion of the sur- vey is provided in Appendix D. 68

institutions, and a quite steady state in the top 50. Compared to the previous two decades of rapid growth in these departments, this is indeed a depression. Women have fared a good deal better in these departments than men in percentage gains, but they remain a very small proportion of all faculty. In these 50 universities in all EMP fields combined, there was a total of 261 women of faculty rank in 1977— or an average of about one woman for every department. In the remaining institutions (the great majority of colleges and universities) the number of women in EMP fields has increased by more than one-third, but, again, because the total is small compared to male faculty, their proportion of all faculty positions remains low. The table also points up interesting differences in the responses of various types of institutions to demands for equal opportunity. The top 50 universities, because of their high visibility and their very low proportion of women faculty in the past, have been special targets in the affirmative action debate. Because of their very large share of federal R&D funds, some of them are also currently targets of special pre-award compliance reviews. To our knowledge, only a half dozen compliance reviews have been conducted. An analysis of one case casts some light on the character of the reviews and the possible conclusions that may be drawn from them. In connection with its recently completed compliance review. Harvard University has published a comparison of projected and actual hiring (Harvard Gazette. September 29, 1978), abstracted here in Table 4.8. Internally established goals for women were not met at tenured and ladder (tenure-track) ranks but were slightly exceeded at "other instructional" ranks. It is interesting to note that the much greater difficulty of locating minority faculty than women faculty (in relation to the Ph.D. pool of each) is only minimally reflected either in Harvard's goals or actual appointments. In the Faculty of Arts and Sciences, which had a total of 352 members in 1978, the goal for tenured minority faculty was 21, with 20 actually in place, compared to a goal of 14 women at which time 12 held tenured positions. At ladder rank, there were 16 minority faculty members in 1978 in relation to a goal of 15, but 49 women as opposed to a projected 56. The tenure distribution of minority faculty (20 tenured and 16 ladder) resembles that of the total faculty (352 tenured and 220 ladder); for women the order is reversed at 12 tenured and 49 ladder. Of these 12 tenured women, two were in biology; there were no tenured women faculty in any other science department. The compliance review was favorable, and Harvard will continue to receive public funding. It is important to note that, as a rule, science departments are the major beneficiaries of federal funds in universities, and are therefore especially vulnerable to possible loss of such funds. 69

Table 4.7 Changes in Size and Sex Composition of Doctoral Faculty* in EMP** Fields by R&D Expenditures of Institution+, 1973-1977 Total Faculty Women Faculty Number % Top 25 Institutions 1977 1973 6,089 5,700 136 84 2.2% 1.5 4-yr. Growth 389 +6.8% 52 +62.0% Second 25 Institutions 1977 1973 4-yr. Growth 5,399 5,490 -191 -3.5% 125 86 39 +45.0% 2.3 1.6 Other Institutions 1977 1973 4-yr. Growth 34,740 30,008 4,732 +15.8% 1,642 1,192 450 +37.8% 4.7 4.0 *Faculty includes full professors, associate professors, and assistant professors at two-year and four-year colleges, universities, and medical schools. +See Appendix B-l for a description of ranking of institutions by R&D expenditures. **EMP fields are mathematics, physics/astronomy, chemistry, earth sciences, and engineering. Computer sciences are omitted because they were not re- ported separately for 1973. Source: Survey of Doctorate Recipients, National Research Council. The statistics in this table are weighted estimates, derived from a sample survey of 65,000 Ph.D.'s in science and engineering. The statistics are subject to two types of error -- sampling and non- sampling, (e.g., non-response bias). A discussion of the survey is provided in Appendix B. 70

TABLE 4.8 Affirmative Action Faculty Statistics,Past and Present Faculty of Arts and Sciences Harvard University Total Minorities Women Tenured Faculty 7/72 (Projected 6/76) 12/76 (Projected 6/78) 6/78 361 (386) 372 (395) 352 15 (17) 19 (21) 20 (12) 12 (14) 12 Ladder Rank 7/73 (Projected 6/76) 12/76 (Projected 6/78) 6/78 194 (229) 214 (228) 220 10 (18) 15 (15) 16 21 (37) 47 (56; 49 Other Instructional 7/73 (Projected 6/76) 12/76 (Projected 6/78) 6/78 144 98 (89) 75 21 22 (22) 16 37 (36; 37 (28) 36 NOTE: 6/76 projections from 1973 Affirmative Action Plan. 6/78 projections from 1976 Affirmative Action Plan Source: Harvard Gazette, September 29, 1978. 71

In any case, in terms of recent improvement, the record of the top 50 universities is better than that of the vastly larger group of "other" institutions, which employ about three times as many people although they spend a great deal less federal money. Given the difficulty of identifying trends from such small numbers, and bearing in mind that within the "top 25" category we are looking at 25 different departments in each field which all have their own more or less unique problems of age, rank, and specialty distributions, it is possible to draw some conclusions regarding employment of women by these faculties. Chemistry departments remain almost entirely male and show no significant change. In mathematics there is also no significant change, apparently due to the decrease in total positions; however, the fields of medical sciences and biological sciences experienced comparable declines in total positions, yet increased their numbers of women faculty during the same period. Chemistry and mathematics have consistently produced many more women doctorates than other BMP areas (Table 2.1). The proportion of women faculty at the top 25 institutions in either field has not exceeded three percent, although women were six and seven percent of the doctoral labor force. The disparities between supply and utilization are much larger than in the other science fields. In earth sciences and engineering, both producing few women doctorates until very recently, they are well represented, in relation to availability. In the remaining science fields—biological, medical, and social sciences (Tables 4.9 and 4.10)—the proportions of women on faculties have traditionally been higher than in the BMP fields although here, too, they were considerably below their representation in the doctoral labor force (Table 2.8). Agricultural sciences remain an exception in this group, with a history very similar to that of chemistry of near-total prior exclusion of women faculty, but better results in catching up. In all of these fields, involving much larger total numbers of faculty, women have made strong gains. Even in medical sciences in the top institutional category and in biosciences in both upper groups, where substantial cuts were made in total faculty between 1973 and 1977, appointments of women increased significantly. An additional element to be considered in hiring is illustrated in Table 4.11 which shows the extent to which leading departments in six fields hire their own and each other's Ph.D's. The fields are selected to illustrate a range in the degree to which the disciplines are experimental in nature since this property is correlated with fractions of postdoctoral and off-ladder appointments by field. For the departments shown in Table 4.11, the percentages of men and women hired who are their own (or comparable) graduates are quite similar, except for 72

TABLE 4.9 Changes in Size and Sex Composition of Doctoral Faculty* in the Life Sciences,by R&D Expenditures of Institution-!-, 1973-1977 Total Faculty Women Number Faculty % Top 25 Inst. 1977 4,229 428 10.1% 1973 4,595 303 6.6 4-Yr Growth -366 125 * -8.0% 41.3% Second 25 Inst. 1977 4,389 348 7.9 1973 4,496 223 5.0 4-Yr Growth -107 125 % -2.4% 56.1% Other Inst. 1977 28,521 3,399 11.9 1973 22,200 2,173 9.8 4-Yr Growth 6,321 1,226 % 28.5% 56.4% * Faculty includes full-professors,.associate.professors, and assistant profes- sors at two-year and four-year colleges, universities, and medical schools. + See Appendix B-l for a description of ranking of institutions by federal R&D expenditures. Source: Survey of Doctorate Recipients, National Research Council. The statistics in this table are weighted estimates, derived from a sample survey of 65,000 Ph. D's in science and engineering. The estimates are subject to two types of error - sampling and non-sampling, (e.g., non-response bias). A discussion of the survey is provided in Appendix D. 73

TABLE 4.10 Changes in Size and Sex Composition of Doctoral Faculty* Psychology and Social Sciences,by R&D Expenditures of Institutions+, 1973-1977 In Total Faculty Women Faculty Number % Top 25 Inst. 1977 4,738 549 11.6% 1973 3,858 324 8.4 4-Yr Growth 880 225 % 22.8% 69.4% Second 25 Inst. 1977 3,447 496 14.4 1973 2,755 257 9.3 4-Yr Growth 692 239 93.0% % 25.1% Other Inst. 1977 29,993 4,881 16.3 1973 22,303 2,795 12.5 4-Yr Growth 7,690 2,086 % 34.5% 74.6% * Faculty includes full professors, associate professors, and assistant professors at two-year and four-year colleges, universities, and medical schools. + See Appendix B-l for a description of ranking of institutions by federal R&D expenditures. Source: Survey of Doctorate Recipeints, National Research Council. The statistics in this table are weighted estimates, derived from a sample survey of 65,000 Ph.D.'s in science and engineering. The estimates are subject to two types of error - sampling and non-sampling, (e.g., non-response bias). A discussion of the survey is provided in Appendix D.

c M »« >t 01 lO r— UO Lf> • • • cn ,— iO CVJ O 3 00 r— O CVi &s • ,— C in rjn r~-^ ^^ ^^ u CO • • o s: M Ol iO ro ^^ GO CVi *** ^" A ro X >J cu in r— CO r— CD cn E LO O O • iO in oo M . Ol 3 |M ^— c cn in ^^ o o cu c c f^ JC CM g-5 +3 S- •r- O ^4 u c OO to IO rO i. LJ- >, cu O^ • • • "^ Ol i0 -a :•: A to oo * at Ol CJ CL. in •^ c s. r— +J JO •r- C <D C0 re cn O iZ 1- ^j (U 0 c cu in ro S CM T3 o cn E _ CO JC C cn c o ro • • • 4J +J to • OJ O VO r-^ VO (O Ol o 3 •— c CU- •r— • r- 1 •r« u O .0 >^ c cn to O c PmB ^^ r— in CU C E cu cn CM Ot CTi JC U •r~ •r— X u s: •^ O ••— D r— •( s: ao r^. *J oo a. E co ^~ JC I/I c to •r" c 01 9 4J •r- ^O JC I/I • re 1 •r- •4-J cu V) +J C0 C c f^ 3-S cu S- I/I - o ^ >. O1 LO !_ o o !-(_ o • e CO -M s- +J o IO co 00 O C0 c o cn i/) 0) Q> Q JC C • JC C C0 CL 4- l— Q- to O •^ E C0 CL o CO c r-~ t^ •r- to • O cn x JC cu rj^ ot r^f P-^ r— Q. r— C D C •r- {_5 s: ^j- 06 . Ol r— •fm O •r- T3 rrt •* rr fx^ ro -t-J u "r— C S ^^ JC ^ g; c un CL CU Q- O ro VO E Q. w •u 45 o tO CL f^ s_ T3 u I/I 4- in <C C o .,— O •^J CO IO to •o t/) r— 1 C CD c s^ I/I u >t •>- ,j ^j cu CVJ CT-. o OJ a* I/I re E a? $- "*^i CL E • c o ro > o -a •o o CM CM Ljf) •r- a. •o Ol i- s- cu Q£ 3 oo 4.J> c I/I 3 S- -O l o to •o Ol in oi ••— ,>-) ^•N^ t_ c ac. > P>« r* c CVJ &^ to A 01 it- O •S Ol CM 00 CM C D 01 r— o s- to s: i. re CL a. z: CM CO E oo A ^f 4_J O) o c uo t~^. CM E I/I o to Oi in 4- ••- in •r— in CL••— £ o 4-> ai 4J >~j O 1 tf" re ro -I-> >> ^ c 4-J o •z. OJ ;>, o S_ E O > 14- •!• to CL A o s s- •u CL Q> C0 S- +-> 3 «/» c o to +-> 4- IO O E c CM &« c •o c 01 00 in CO c Ol • -M p: CVJ re ro -a +-> oi o 6 fm • • • p« •r- •r~ CO JC in CO csj CL U CL > 4J JM • c^ 3 to r^- CJ X cu O •r- (J (— e •^ f"~ ^s^. Ol • I/I O S- CO 4- Q_ in CVJ •r" I/I Ol Ol •r-) O u C ^J- 5^ 00 C 1 4JI a: 0_ •r- 0> o , — o ro 03 1 3 C 0 VO s: CM ^ Q1 " 10 O 00 to VT) S- X I/I +J C0 •!- +) 10 to "— O •r- U- T3 3 4- S- O to cu cu m 1 Q_ s_ -u i- in C Ol • CL I/I o ro to 3 E U •r- C0 C0 •*-> CO •r- s? c• — • to CU ^^ •o I/I +-> U ^- cn a. c <: 01 cu O 4- O o cn cu Oi l— o a: i. Q in +J T3 I c S- E •r— r_ Ol to l— • -^" +-> cu O. Q. M- E CC O-0-t- Ol 4J >>-t-> in • •r- O) Ol i. 4-> CO CL C0 ro Ol 03-0 to to i. C0 V ) JC 3 Ol Ol O O JC CO cn 0 U c •* 01 -o Ol JC 3J-~- -1 JC < 4-> O •r- to to 01 -a > cn oo 3 o -T: u. u. in i — O ,— s_ •r- oi ro ^ E C -a 1 i. 01 JC U 3 O) JC •!- 2i0 •r- 4J> c cu C i0 •r- -I-> c -t-J c I/) S (- .0 •r- -t-> O T3 4- • 4- C + 1 |Q C c U) ^- cu •o -o a: ~D CO cC r~ Ol >. oo E CO I — CO 1 QJ E 1 O -o 4-> • • UJ - 4-> >>!-» ^l ^"* ^> 1 * Ol to ^ •^ Ol I • t- o cn O r— O $- In Oi ,__ 3 u S" C3 to r— r— r— lO O u U t- c-t • 0. Q. Q. Ol a. a. O C c h- JC 01 E C E •!- E QJ u_ o O- Q LU -r- UJ IE UJ O Jf 4- * co * 75

mathematics; however, the percentages hired as faculty are markedly different, varying with the availability of off- ladder positions. Table 4.12 shows that in those fields where the option of off-ladder or research appointments is open, women are hired preferentially in non-faculty positions. Where this option does not exist, women become faculty members if they are hired at all. Table 4.13 shows that the predominance of women over men at the instructor/lecturer rank increased at the top 25 institutions from 1973 to 1977, remained the same at the second 25, and decreased only at the "other" institutions. Off-Ladder Positions As a general rule, departments hire professional staff in variously designated positions which are outside the tenure stream. Such positions may be a holding pattern for people they would like to keep until a more promising appointment becomes available. More frequently, however, they are marginal jobs, fluctuating with enrollments, unexpected leaves of regular faculty, and other exigencies. Some are used as semi-permanent ways to teach service courses or supervise laboratory instruction. Many such positions are part-time or part-year, often obviating the need to pay even prorated benefits. They provide a very economical way for science departments to get the chores done. In virtually all the cases with which we are familiar, people in such off-ladder positions are not permitted to apply for outside research support, thus eliminating any chances they might have to establish independent research records and improve their prospects. Women are heavily over-represented in positions of this sort. The considerable fluctuations over time and among fields with otherwise similar characteristics that are apparent in Table 4.12 testify to the marginal nature of this academic labor supply. Persons characterized as "instructor/lecturer" and "other/no report" are not included in faculty ranks but are departmental employees; postdoctoral appointments also fall into this category but have already been discussed in Chapter 3, so that the data under consideration here concern only the first two classifications. Table 4.14 shows the numbers and percentages of women, for all science fields, who held such off-ladder positions in 1973 and 1977. Not only the absolute numbers but the proportions of women in off-ladder positions increased between 1973 and 1977 in all except two categories ("other/no report" in the top 25 institutions and "instuctor/lecturer" in the "other" institutions). It should be noted, however, that the total numbers of both men and women in off-ladder positions are small compared to total faculty. 76

TABLE 4.12 Number and Percent of Doctoral Scientists and Engineers in Academe* at Rank of Instructor/Lecturer, by Field and Sex, 1973-1977 1973 1977 Field Men No. Women % No. % Men No. * Women No. % Mathematics 125 1.2 48 7.5 833 8.2 90 1 0.9 Phys i cs/As tronomy 173 2.5 26 9.6 173 1.9 35 1 2.4 Chemistry 222 2.3 92 11.9 264 2.3 100 8.4 Earth Sciences 23 0.5 8 5.3 38 0.7 9 3.6 Engineering 41 0.3 ** 4.C 154 1.1 ** 5.0 Agricultural Sci. 25 0.4 ** ** 57 0.7 4 2.8 Medical Sci. 122 2.6 32 5.1 169 2.5 59 5.0 Biological Sci. 338 1.6 174 5.2 443 1.8 267 5.5 Psychology 81 0.4 125 6.4 228 2.0 162 5.2 Social Sciences 153 0.9 82 4.1 341 1.5 176 4.7 *Included are two-year and four-year colleges, universities, and medical schools. **Estimates based on fewer than three sample individuals are not shown. Source: Survey of Doctorate ecipients, National Research Council. TABLE 4.13 Number and Percent of Doctoral Scientists and Engineers in Academe* at Rank of Instructor/Lecturer, by R&D Expenditures of Employment Institution"*" and Sex, 1973-1977 R&D Expenditures 1973 1977 of Employment Institution Men No. % Women No. % Men No. % Women No. % Top 25 Institutions Second 25 Institutions Other Institutions 217 1.3 151 1.1 935 1.3 81 5.9 4.6 6.1 154 0.9 192 1.4 1,877 2.0 134 68 704 6.8 38 470 4.9 5.7 *Included are two-year and four-year colleges, universities, and medical schools +See Appendix B-l for a description of the ranking of institutions by R&D expenditures **Estimates based on fewer than three sample individuals are not shown. Source: Survey of Doctorate Recipients, National Research Council. 77

TABLE 4.14 Number and Percent of Women Doctoral Scientists and Engineers in Selected Positions in Academic Institutions* by R&D Expenditures of Institutionro*ro, 1973-1977 1973 1977 No. % No. Top 25 Institutions Total Employed in Academe 1370 7.6 1979 10.1 Faculty** 711 5.0 1129 7.4 Instr./Lectr. 81 27.2 134 46.5 Postdoctorals 209 11.7 387 17.4 Other/Rank Not Reported 369 21.5 329 18.5 Second 25 Institutions Total Employed in Academe 825 5.7 1384 8.9 Faculty 566 4.4 980 7.3 Instr./Lectr. 38 20.1 68 26.2 Postdoctorals 124 14.1 200 18.9 Other/Rank Not Reported 97 14.0 136 16.9 Other Institutions Total Employed in Academe 7643 9.4 12392 11.9 Faculty 6160 8.3 9983 10.6 Instr./Lectr. 470 33.5 704 27.3 Postdoctorals 443 17.6 951 22.5 Other/Rank Not Reported 570 22.5 754 23.3 *Includes two-year and four-year colleges, universities, and medical schools. +See Appendix B-l for a description of ranking of institutions by R&D expenditures. **Includes full professors, associate professors, and assistant professors. Source: Survey of Doctorate Recipients, National Research Council. The statistics in this table are weighted estimates, derived from a sample survey of 65,000 Ph.D.'s in science and engineering. The statistics are subject to two types of error -- sampling and non- sampling (e.g., non-response bias). A discussion of the survey is provided in Appendix B. 78

Positions of this sort do at least provide some employment for women (and a much smaller fraction of men) who cannot find more promising appointments. Traditionally they have been viewed as opportunities to continue some professional activity for women scientists who were ineligible for regular positions because of nepotism rules or overt sex bias, or who temporarily preferred less demanding commitments for personal reasons. If positions of this type were restructured to encourage some individual initiatives, access to research facilities and funding, and broader professional contacts, their impact on women scientists would be a good deal less detrimental. Programs to further such professional development for people in off- ladder positions need not be costly; in many cases they would involve changes in attitudes and institutional policies rather than expenditures. However, a modest funding program to provide for research costs, including overhead, would encourage more rapid changes in attitudes. The National Science Foundation's Women in Science program would be an appropriate location for such a project, perhaps on an experimental basis. Promotion and Tenure Whether or not women faculty are promoted to the upper ranks and to tenure—the two are not necessarily equivalent—provides a particularly sensitive test of the commitment of universities to equality of opportunity. A great many charges have been levelled at academic institutions in the past few years concerning their readiness to give women "revolving-door" appointments— short-term assistant professorships with no prospects of consideration for tenure and therefore none of the security necessary to become established and to initiate and carry out research. Reports of studies in progress at several major research universities have indicated internal concern that the rates of promotion to tenure for female junior faculty are well below those for men.» Simple counting of positions by rank and sex (Table 4.15) does not lend support to such charges. Women have indeed done very well as a proportion of new hires, i.e. assistant professors, although the available pool is not being drained. Their rate of increase at this rank suggests that some women who have held long-term postdoctoral or other off-ladder positions are being appointed to faculty posts. This inference is supported by numerous case histories.2 The increases at associate and full professor ranks represent promotions and presumably include some inter-institutional movement which may or may not differ systematically for the two sexes. Again, women seem to be doing well in promotions, especially in the top 50 institutions where, however, their actual numbers are very 79

small. In the more numerous "other institution" category, they are advancing much less rapidly perhaps in part due to a transfer of the best women at these institutions to the more prestigious ones. Still, at the associate professor level they are advancing in proportions comparable to their presence as assistant professors. These data tell us little that might resolve the "revolving door" or flow-through problem. Promotions made between 1973 and 1977 went to individuals who had been hired about 1970, on the average, with occasional exceptions for "fast track" individuals. Monitoring comparative rates of promotion for male and female assistant professors over the next five to seven years will shed some light on the problem but will not give conclusive answers. People who have held only short-term appointments in top universities may move down to similar posts in lower-ranking institutions, perhaps displacing individuals who leave the system altogether. Various other permutations are possible, but eventually many such people become ineligible for further academic positions through length of prior service, or they are simply viewed as undesirable through having been terminated too often, for whatever reason. Whether this is happening disproportionately often to women is a question that should be investigated now by longitudinal studies. It is possible, however, that salary analyses may illuminate this problem to some extent. This possibility is discussed below (P. 97). Before presenting data comparing men and women with respect to tenure, it should be pointed out that an additional factor needs to be considered. This is indicated by a recent tabulation by marital status of the tenure standing of 1971-1975 Ph.D.'s in the biomedical and behavioral sciences who were employed in academic institutions in 1976 (NRC, 1977, Vol. 2, p. 138). The results showed that in both fields, married women were somewhat less likely than single men or women to hold tenured or tenure-track positions, but the differences were small. Large differences were noted, however, between these three groups and married men who were far more likely to nold such appointments. Thus, 65 percent of the married men in the biological sciences held tenured or tenure-track positions (compared with 13 percent of the unmarried men, 38 percent of the married women and 44 percent of the single women). In the behavioral sciences, 83 percent of the married men were in tenured or tenure-track appointments (as opposed to 70 percent of the single men, 69 percent of the married women, and 72 percent of the single women) . In this respect again, married men present the picture of high achievers. It should be recalled that married men make up more than half of all Ph.D. recipients. 80

TABLE 4.15 Changes in Doctoral Science and Engineering Faculty by Rank, R&D Expenditures of Employment Institution*, and Sex, 1973-1977 Top 25 Institutions Professor Men Women (%) Assoc. Men Professor Women (%) Asst. Professor Men Women (%) 1977 7891 194 (2.5%) 3452 289 (8.4%) 2650 614 (2?. 2%) 1973 7322 138 (1.9%) 3259 205 (6.3%) 2838 368 (13.0%) 4-Yr Growth 569 56 193 84 -188 +246 Women as % of total increase 9.1 30% 100% 1977 1973 4-Yr Growth Women as % of total increase 5983 5817 166 Second 25 Institutions 255 (7.3%) 185 (3.1%) 102 (1.8%) 83 33% 3491 3482 9 199 (5.7%) 56 86% 2821 2857 -36 498 (17.7%) 265 ( 9.3%) +233 100% 1977 34115 1973 27047 4-Yr Growth 7068 Women as % of total increase Other Institutions 1988 (5.8%) 1563 (5.8%) 425 5.7% 27914 22062 5852 3013 (10.8%) 1868 1145 16.4% ( 8.5%) 21566 19075 2491 4759 2714 2045 45.1% (22.1%) (14.2%) * See Appendix B-l for a description of ranking of institutions by R&D expenditures. Source: Survey of Doctorate Recipients, National Research Council. The statistics in this table are weighted estimates derived from a sample survey of 65,000 Ph.D's in science and engineering. The estimates are subject to two types of error - sampling and non- sampling, (e.g., non-response bias). A discussion of the survey is provided in Appendix D. 81

Tenure status is an important factor in assessing equity in academic appointments. The uneven proportions of tenured positions for men and women are shown in Table 4.16 and Figure 4.2. We have examined the sex distribution of tenure in associate professorships in more detail. Field totals in both numbers and percents for 1975 and 1977 are shown in Table 4.17. The trends for successive Ph.D. cohorts are given in Tables 4.18A and 4.18B. It is clear that overall, men are still somewhat more likely than women associate professors to be accorded tenure. The discrepancy increased in some fields, notably chemistry and biological sciences, from 1975 to 1977. The exceptions are the social sciences and those fields containing very few women—physics/astronomy, earth sciences, agricultural and medical sciences—for which there are also substantial fluctuations from year to year and cohort to cohort because of the small numbers of women. In the biological sciences, the overall difference remains but appears to be diminishing for the most recent cohorts. In psychology (except for the most recent cohort in 1975), mathematics and strikingly in chemistry, the male advantage with respect to tenure remains down to the latest cohort. The fact is that despite the very recent appearance of equality in the granting of tenure to men and women in some fields, it could well be several decades before the relative proportions of men and women scientists who are tenured faculty approach equality. As the numbers in these tables indicate, women are still a minority of the new Ph.D.'s in each field and in some disciplines, a very tiny minority. Siven the decline in college enrollments of the 1970's and the general tightening of opportunities for faculty advancement of both sexes, the granting of tenure to men and women faculty in equal proportion will not produce substantial changes in faculty composition for a long time to come. Performance of Male_and Female Faculty The construction of standards for assessing the relative merits of faculty is an enterprise in which countless committees have come to grief. We will attempt not to join their ranks, but wish to note several points relevant to quality comparisons. The desirable qualities of faculty members are usually assessed under the heading of teaching, research, and service to the institution. Women's capacity either as teachers or contributing members of the university community has not been questioned, but their research potential has. 82

TABLE 4.16 Tenure Status of Science and Engineering Faculty at Four-Year Colleges and Universities by Rank and Sex, 1977 Faculty Rank Number and Percent in Tenured Positions Men Women No. %a No. %a Professor Associate Professor Assistant Professor 49,275 95.8 29,784 81.6 3,458 12.6 2,314 92.0 2,638 71.4 593 10.0 a Percent is based on the number who reported tenure status. Source: Survey of Doctorate Recipients, National Research Council. The statistics in this table are weighted estimates, derived from a sample survey of 65,000 Ph.D.'s in science and engineering. The estimates are subject to two types of error -- sampling and nonsampling (e.g., nonresponse bias). A discussion of the survey is provided in Appendix D. Figure 4.2 Percent of Doctoral Science and Engineering Faculty Holding Tenured Positions, by Rank and Sex, 1977 100 80 I 60 CU u (U 40 20 b^•v^ Men I I Women Professor Associate Professor Assistant Professor 83

TABLE 4.17 Tenure Status of Associate Professors Holding Science and Engi- neering Ph.D's by Field of Employment and Sex in 1975 and 1977 * Percent in Tenured Number and Positions Field of Employment 1975 1977 All Science and Eng. Fields Men Women Men Women 25,136 77.8 2,097 71.3 28,755 81.9 2,250 72.9 Mathematics 2,573 81.9 134 74.9 3,098 88.2 164 82.0 Computer Sciences 273 — 290 68.6 15 100.0 74.8 Physics/Astron. 1,956 81.3 35 100.0 2,181 31 66.0 88.2 Chemistry 2,406 216 78.3 2,493 89.1 169 76.1 85.4 Earth Sciences 1,076 68.8 23 1,125 79.0 18 100.0 Engineering 3,054 100.0 3,176 80.5 80.4 — Agricultural Sci. 1,555 82.7 12 1,616 81.1 12 100.0 100.0 Medical Sciences 800 63.7 133 55.0 1,162 67.6 200 65.1 Bio. Sciences 4,737 80.1 528 72.9 5,218 80.9 612 65.9 Psychology 2,274 70.8 422 65.8 2,508 77.6 485 70.6 Social Sciences 4,432 74.7 594 73.3 5,888 82.6 844 79.4 Percent is based on the number who reported tenure status. Source: Survey of Doctorate Recipients, National Research Council The statistics in this table are weighted estimates, derived from a sample survey of 65,000 Ph.D's in science and engineering. The estimates are subject to two types of error,-- sampling and non- sampling, (e.g., ron-response bias). A discussion of the survey is provided in Appendix D. 84

TABLE 4.ISA 1975 Tenure Status of Associate Professors Holding Science and En- gineering Doctorates by Field of Employment, Sex and Ph.D. Cohorts Number and Ph.Ds * Percent in Tenured Ph.Ds Positions 1970-72 Field of 1950-59 1960-69 Ph.Ds Women Employment Men Women Men Women Men Total 2,290 91.7 294 73.1 18,101 82.6 1,302 74.0 3,814 60.3 347 60.0 Mathematics 149 85.6 — 1,925 87.9 103 78.6 415 62.9 19 52.8 Physics/Astron. 137 100.0 12 100.0 1,655 85.1 23 164 60.1 — 100.0 Chemistry 346 50 100.0 1,827 134 133 11 40.7 100. 0 84.9 75.3 61.0 Earth Sciences 85 100.0 — 831 72.3 23 100.0 160 53.9 — Engineering 227 100.0 — 2,444 86.5 — 317 51.5 — Agricultural Sci. 224 100.0 __- 1,046 84.9 12 100.0 285 67.1 — Medical Sciences 115 31 51.7 510 67.9 70 68.6 156 49.8 32 52.5 77.7 Bio. Sciences 622 92.1 95 75.4 3,416 81.9 299 70.7 526 51 61.1 58.6 Psychology 139 59.4 66 79.5 1,737 266 62.3 398 56.1 90 68.7 79.0 Social Sciences 246 40 56.3 2,710 82.4 372 1,260 144 61.0 100.0 84.4 64.5 Percent is based on the number who reported tenure status. Note: Computer Sciences are not shown here due to the small number of cases; cohorts prior to 1950 and after 1974 have been excluded for the same reason. Source: Survey of Doctorate Recipients, National Research Council The statistics in this table are weighted estimates, derived from a sample survey of 65,000 Ph.D's in science and engineering. The estimates are subject to two types of error -- sampling and non- sampling, (e.g., non-response bias). A discussion of the survey is provided in Appendix D. 85

TABLE 4.18B 1977 Tenure Status of Associate Professors Holding Science and En- gineering Doctorates by Field of Employment, Sex and Ph.D. Cohort Number Ph.Ds and Percent in 1960-69 Tenured Ph.Ds Positions 1970-72 Field of Employment 1950-59 Men Ph.Ds Women Women Men Women Men Total 1,992 92.4 278 86.1 17,110 86.6 1,816 75.8 8,704 72.6 746 65.4 Mathematics 139 100.0 7 100.0 1,907 92.7 111 88.1 1,002 79.1 39 65.0 Physics/Astron. 167 100.0 7 100.0 1,568 90.7 24 60.0 391 ___ 74.9 Chemistry 205 100.0 30 1,822 90.7 119 76.8 379 76.0 20 100.0 54.1 Earth Sciences 38 — 731 84.7 18 100.0 356 68.1 — 100.0 Engineering 128 — 2,229 88.8 779 63.9 100.0 ___ — Agricultural Sci. 191 882 83.6 12 100.0 543 72.7 100.0 — Medical Sciences 112 68.3 38 100.0 648 66.7 71 353 66.2 57 52.3 68.9 Bio. Sciences 625 91.9 84 65.1 3,288 84.0 352 64.9 1,182 69.0 121 65.1 Psychology 151 72.9 45 100.0 1,438 82.8 280 75.1 919 160 71.5 59.5 Social Sciences 236 100.0 67 100.0 2,597 89.4 389 2,800 76.2 349 72.9 87.0 Percent is based on the number who reported tenure status. Note: Computer Sciences are not shown here due to the small number of cases; cohorts prior to 1950 and after 1974 have been excluded for the same reason. Source: Survey of Doctorate Recipients, National Research Council The statistics in this table are weighted estimates, derived from a sample survey of 65,000 Ph.D's in science and engineering. The estimates are subject to two types of error - sampling and non- sampling (e.g., non-response bias). A discussion of the survey is provided in Appendix D. 86

Estimates of research potential are based on prior productivity, ordinarily measured by publication and citation counts, the latter considered a proxy measure of quality. Citation counts suffer from a variety of well- recognized methodological problems having to do with ordering of authors' names, citations to review articles, and frequency of references to erroneous results. Productivity measures using citation counts must therefore be evaluated with caution. Studies comparing the productivity of men and women faculty have yielded ambiguous and sometimes contradictory results. One major study of a national sample of faculty (Bayer 1970:15) found that 2 percent of the women and 11 percent of the men had published 21 or more articles, and that 63 percent of the women and 39 percent of the men had not published any research papers. These results are to be expected for a national sample, taking into account the relative distribution of male and female faculty among types of institutions and ranks documented in this chapter. Studies controlling for rank, field, and institution give quite different results, finding little or no difference in publication rates (e.g., Loeb and Ferber, Hargens) or higher publication rates for married women than for unmarried women and married or unmarried men (Simon, Clark, and Galway). For the specific case of science faculty, factors such as access to appropriate research facilities, division of time between undergraduate and graduate teaching responsibilities, and especially availability of graduate and other research assistants may be of far greater signficance to productivity than rank or other variables which have been controlled in the studies cited above. We have not found any studies that control for these factors or indeed consider them. As we have shown in this chapter, the distribution of women faculty in research departments is such that productivity comparisons between men and women of similar age and experience in the same field and institutions, and thus with comparable research opportunities, are virtually impossible. A lively illustration of this difficulty is found in Career Achievements of NDEA Fellows (NRC, 1977a) which reports lower publication and citation rates for women who were National Defense Education Act fellows than for men. In nearly all fields, the average number of publications and citations for men fellows exceeded those for women by 50 percent or more. It also reports, however, a mean salary differential of approximately 25 percent for the two groups, which is far too large to be explained by sex discrimination per se. Clearly, the male and female fellows, presumably 87

quite evenly matched at the outset, must have had different careers providing very different opportunities for research. It appears likely that comparisons of this type serve more as indicators of inequality in initial appointments and later promotions, with attendant differences of access to graduate students, facilities, and funding, than as measures of inherent sex differences in research capability. Certainly, the most carefully controlled studies of productivity do not suggest a significant advantage for either sex. Appointment and tenure decisions are not usually based on statistical comparisons but on individual judgments of a given person's performance. That performance may or may not be affected if the person is an isolated phenomenon, a token; none of the studies of comparative productivity have taken account of this possibility. The dangers of tokenism cut both ways, as Ranter (1977) has pointed out: the token person may, by the pressures of being an outsider, be forced into behaving in uncharacteristic and counterproductive ways but the majority group may then respond similarly. Until an occasional major research department can assemble at least a critical mass of women faculty—something of the order of one-third of its members, according to Ranter's hypothesis—we do not believe studies of comparative performance will have much validity. Faculty Salaries Salary differences between men and women are widespread in all occupations and at all educational levels. Academic salaries are no exception but the extent of the differences has been difficult to determine. The reason is this: breakdowns by field, institutional category, and years of service, etc., yield so few women in most cells, at least in the sciences, that multivariate regression analyses are often required. A detailed study by Darland et al. used multivariate regression analysis with over 25 predictor variables to investigate the extent to which male/female differences could be explained by relatively objective factors, such as highest degree earned, differences in performance, etc. The study found that women were typically paid about $1500 less than the predicted salary for a man with the same attributes (Darland et al., 1973). The analyses were based on a large-scale national survey conducted by the Carnegie Commission in 1969. In this section, we will examine more recent salaries of men and women faculty, and the extent to which salary differences may have narrowed between 1973 and 1977. 88

In 1977 men's academic salaries overall exceeded women's by approximately 20 percent, with a slightly better record in public institutions and a slightly worse one in private ones, corresponding to other differences such as rank distribution. The magnitude of the effect is due in part to women's concentration at lower ranks, and is often ascribed to the fact that relatively more of them are in lower-paid fields. As we shall demonstrate, such deductions are not necessarily valid. In Tables 4.19A-C we examine median salaries for men and women faculty by rank for those science fields containing enough women to make the comparison at all, and we examine as well the trends between 1973 and 1977. At the full professor level, where women have suffered larger percentage and dollar differences than in other ranks, three fields (chemistry, medical sciences, and psychology) show larger discrepancies in 1977 than in 1973. In other fields there was some reduction of the difference in 1975 followed by larger gaps in 1977. The dollar differences for men and women full professors were at least $2,500 in every field and reached $6,200 in chemistry. The trends are not the same at other ranks. We still note an increasing salary advantage for men in chemistry and fairly large differentials in the medical sciences. Two fields— psychology and physics—show relatively small gaps at associate and assistant ranks by 1977. When we examine the data by Carnegie Classification of institutions* as in Table 4.20, we have almost no fields left for comparison in Research Universities I and II, as might be expected from rank distributions discussed earlier. Assistant professors' salaries should show the fastest response to change, and are the ones given here for the 1970-1974 and 1975-1976 Ph.D. cohorts in 1977. Improvement for the later cohort is generally not dramatic, except in chemistry in the category of "other institutions," and in psychology in Research Universities I where there is a surprising reversal of the difference, the only one we have noted. Psychology in "other" institutions, however, has lower salaries for the more recent women doctorates, and sociology shows no significant change. We have also examined similar data for all ranks, Ph.D. cohorts, and institutional categories. The data are not reproduced here because of their complexity, but they seem to hold no major surprises. In all cases women's salaries are lower than men's; in many areas the differentials are largest for the 1960-1969 Ph.D. cohort. It should be pointed out here, as well, that an extraneous factor is operating to affect the salaries of 89

OJ to £ S 0 8 01 o •r— u CO 01 o ro to i— •o i c co ai r^ s- cn CU S ra M_" S CO CM ro CTl ^_ •r- ,_! co" l>! ^i" CO co Q ( CM r" ( r™" O !*- O r— O CTl oo" r""i r*^ O co C o tn 0 0 O CO C D r— O CO o in CU CO r— CTl CM in i — CTI in O CO co in 1 — . E vo*~ " •~—*• r. ^ * •^ ^_^ •-*•— ^ **.~^ f-. O r— co in co VO 2 3 OJ CM CM CM CM CM o m ogr CD CTl o rT O *d- C CTv C D r— C CTl o co O CTl O in o co n i CTl CO i — CO ^t o vo r•^ in CO CO OJ vv cn co CO ^J- CO CM CTI r-^ O CO o co CM CO OJ CO CO *l CO eo CM CO 0 O ,_ cn CO CM (O CO ^~ l~ ( ^^ O VO O CM O O O CO O CM oo" £; C D ^1- O VO o *a- o in O ** O r— CU VO r— r — i — co r— CO VO OJ CO oo m in E M^_^ ^ -~_-- *^v* *• -•— - ••*»* *i^_^ r^ O ^> o in CO in CO CTl 3 CM CM 00 CM CM CM CD VO O VO CD r*^ O CTl O O o ^s- O ** O VO 0 0 C CO O r— O r— c tn r — CO CO tn CM CM CO co o CTl O to s! r^ co <* ^i- O CM vo r^ vo ** vo co L CM OJ- — CO CM CM- CM 0 CO CO CU O I. OL • ^5 «n 00 o ^ 4- in CO 1^ !^I CM CM CM f^ |^ r- 3 U. Q ^~ CM •~ r— r— to r— O VO O CT. O CO o co^ 0^? E O •— O CM o r-. O CO o co o in ^& •~~ f^ r^ CM — - vo m CTl OJ CO CO CO F '• ^_^ '•—-_' A f % — - •I*v* "-—-• r— O CM co ^^ O CM O 2 CM rmm CM CM CM CM r~l r^ O VO o in O CTl f""i VO O O o r- o co O OJ CD vo O CM o oj c OJ VO CTl i — CO VO r— r— VO .— CU •s. in oj CM CO !*- r— ^- r>. «a- co -r vo OJ CM OJ~— • CM CM OJ-— - VI to CU • co 0 •r— CU • U u CU 00 c •K CO •^ Ol ^-> u u r— >•> •r- C •r— ^ 00 ro cn u 4. - CU 4_> ^ O O 00 ( 3 £ E to ro 01 'o T 3 O CU •r— O 0 r~ ro F- • r— r* ^ ••— fW 0 •p* < a Q. 4~> CU -o o u • r - E rO -C CU ••— CO o U . LU 0 C3 Q- oo 3 ro 4- U CU T3 to C O •r- I/) OJ l/> Q) c oi •o c o ex Q1 1_ CU ot 4- o c +-> i_ CU (O i4- .c 5 •o CO s_ o TD O) 0 ro U o c CU . 0l -o CU CU C S- CU ro S- to ro Q. rO C Q ••- M • S- co CU •)-> .0 £ .3 •fO 3 CO CU O o S- .C to O 0l to i •— i— tO i— O 4- -5 CU O E T3 O CU to to 3 CU • ••-)••- X, T3 +-> S- to *r~~ rO CO r— c i- as CU CU to > •r- 01 C C CU CU 3 ro - S- .IT in O CU C- to cn CU -C CU S- -1-> r— C r— CO O O r— E U t0 3 O S- T3 i— ro •!- I CU > CT> >>••- I ~a S- t- C O 3 •r- 4- O 4- 4- tO O CU 4-> •r- ro S- S- CU ro T3 _O •=£§ CO O C U "o-TS •r- E CU E CU 4-> oi -c T3 CU Ul O ••- O) I/) CU I c o D- T3 O ro O O cnO " C in •!- x vo r— T- D.-0 4- E C O (O CU CO Q. >> Q- 01 I eC i. C 3 S- •*- CO O S- -O CU S- CO i— CU T3 O. •r- E 4- > ro O O CO S- CO O. fO CU Q. CO E +J*•(" S- >> i4- O QJ ^3£ CU 3 > O CO O CU +J -C C T3 O 4-> 3 CU 0 -•'->4- O CO _O O CU 3 -C 4-> CO C t-J rO O S- E CU v- ro ••- S- CO CU -I-1 rO to CO CO 3 CU CU CO U QC U CO T3 ••- -r- i— CU 4-> T3 ro 4-> CO C -C -r- et o cn+J •r- •r- HJ +J CU •t-> • S to-— to CU 01 ro S- JZ •r- rO r— -Q to to C CU CU r— Q. ro •r- -I-> U CU CO CC. ••- CU 4-> +-> ro C O •(-> to O O O •r- Q 4-> CO 4- T- O 4-> tO >t•t-> CU to > S- CU 3 -C 00 r— CU g O oo (U • CO tn c e o •r- CL S- CO CU CO CU S- C I •r- c Ol O c c CU •I T3 • c cn fO • CU (U—, o c: cn CU C 0 t— to CL E C ro •r- tO 90

-o "aJ X 0) in o o o 0) cu •r— u to o 0) S- ro to u I0 c ro •o (U CTl co i— •o i c: co a> r>~ s- cn CO cn 0) 4-' CO CO ^ «, co cn CO 4- • • • • • • •r— CO ^— Cft f r— o CM ^3 r^^ o LO o cn f~> LO O LO O CM oor o C O CO O ID o oo fi r— - o cn O CM o cn cu LO l — cn- — CM i — CO CM LO t^. r— LO o cn r-. o ,_ ro"" rl o CO CM CM CM 2 CM CM CM CM CM CM CM O •=!- O O O LO O cn O LO o «d- O r— o oo O CM o r-- o cn O CM O CM O r— c CM OO CO CO i — Ln O ID co oo CO , — Ln ID 2: CM CO CM CM CM CM LO r— CM LO CM CO CM LO CM CM-— CM CM CM • CM CM- 4-' r- co O vo in iT) !£ LO CO CO a-' co ,_ CM Q o in o ^- o or o C r— O CO O CM o cn C t i i— - O «3- o «a- o »a- o ^3- o cn O LO cu CM r— Ln — i — CM CM CM i_n LO oo en r^^ r^^ LO E 4h c< -—- ^ --— - r -^_^ ff-— ^* P-IN. o cn cn oo t— cn cn cn O^ 3 S rro~ i CM r— * r— tn O r-- CD cn O Ln 0^r o 0^ 0^r O CM S- O LO o «a- O r- O CO O CM o .— O r- O c CO O i — *d- LO p— i — CM CM r-. i — CM CM CO co cu 01 O CO O CM cn CM CO i — o en O CO O LO QJ |^ CM CM- i — - — CM^ CM CM CM O E Q. QJ ro . *e cn •r - 4- p— . r— in CO in 1 — . L) 4- • • • • • • • O • r— LO CO LO in LO o LO 01 Q co ,a: O r— o Ln O O G5 I — - o cn O r— O ,— C o ,* o ^- o Ln O CO 0 0 O in O CM QJ CO-— cn <— •3- co LO LO CO E MX^X •i «— — * r, ^ • ^ ^_- •V -_-- ^ -_^ r— , o r— . fN^ LO CTt r — co 1 — cn 3 K *"" *— r~ T~ *"~ *— CO *^r o o Ln c^ r- O LO O r— CD CO O CO O CM O CO 0 r— O CO O «d- O LO 0 0 c 10 «* m cn *Ct- CM 0 0 LO i — LO *J- LO r-. cu 00 CM oo .— r-— CM oo LO 00 CM oo «^- cu • co (J •|M QJ c U O cu c * in •t— cu O ^^ u ^-~ >^ 4 c •r - tl CO rO cn 0 - oj 4_> LJ O 00 ( 3 £ ro CO -I-J r_ •i — E 0 ro CTl O r^ ^ 3 0 cu •r— •r» o 0 -C ro r* u 'a. e— l/l E •r- u •f— < +J >-j CD •a O o •r - E ro JC cu •r™ i/) o U - LiJ s: D- c_> CD a_ OO in 1 01 . c QJ QJ Q O 01 01 • c O QJ JC x: ^: 0- -O -»-' -t-^ c C o ro s- cu 0 0 S- O cn o 4- ro - c CL LO •r- X QJ LO , — •r- S- C Q-T3 ro •r- 4- E C o ro cu ro co 01 CL +J S- >, Q. ro cu QJ 1 <C f^ rQ > I E S- C . -z_ i/l O 01 t- -a 4«A cu s- cu C • r— QJ T3 QJ 01 a. -r- , r— E 4- > ro o ro o o • > o cn S- I/i •r- -C co CL +J 3 O ro cu c cr co Q. OO QJ cu £>,-r- •o 0 4-> c t- ro s- >> o ro o 4- O QJ Q. cu •*- 2 > l/» -O 4-> S- QJ 1 QJ QJ 3 S- ,— E . > O co ai 3 S- ^ s- cu r— 4- 0 • O QJ +•> -C ro >^ C -O U +•> E o • s- 3 QJ QJ 4-> co ro O " •r~j4— <-*— QJ r— C_J LO f^l O •D -r- ro CU 3 O QJ -M co -C +J 01 C r— •4-^ *r~ o ro o co LO cn s- E cu •r- C 3 S- C ro •r— S- co 1o •r-> QJ •r- cu ••-> ro co -C -o > 4-> 01 CO 3 -f-J 10 •i- S- cu cu co o c: o cr u co . c 3 Q. T3 •r- -r- cu QJ QJ i — CU 4-> T3 £ cu - s- ro -t-> co ^2 OO C -C -r- <C 4- cu in o cn -i-> QJ cni — •r- -r- ro ^^ > cu ro 4-> QJ -!-> • -t-> ro >— 3 ro 2 co - — - •r" -(^ r— -^3 r" co 5 O -r- cu cu ro 01 O > - S- .C -r- I/I -C •r- co ro l— J3 •o 4-J t_ roO +J p™ c ro c c cu cu 0) O QJ -r- CU i — • l/i •r - E >> •r- _a cn c 4— 1 4- Q. ro c o O S- O •r- 4-> -r- Q. -. T o s_ O S- OO O QJ 01 QJ QJ 4- cn 4- cu OH -r- QJ S- -Q -C C 1 •o S- 4-> E QJ •(-> -r- C cu o ro 3 4-> cn o -t-i 4- C ro c c: c ro T3 s- -^- cu r— co cu -a 0 3 QJ >> OJ 4-> 01 roO • 0 •r- O +-> o o c cn F^ S- r— x: o •r- ro • (O ro CL cn Q +J QJ o r- E -r- co QJ ro cu cu 4- •r- 0 4> co 2 o -t-> c ai O cu ro a> c c: O E QJ >) 4-> T- •r- •r- -r- +J QJ 01 0 i — cu E +-> ro > 01 CS. s- QJ 1 O s- cu E cu -0 r— •r- 3 -c c ro 2 ro r— *T3 CO 1 — •i— 01 0 3 C 01 e£ 4- -r - ro cu -u cu s. cu 4_> 3 s: o o -K •z. to 91

i — 3 14- O r— O LO CO VO CO 1 I*- co M- • • LO CO O CM CO Q O cu . c •r- ^J- r— r-~. a ° ^— *^ „-—-.. ^i n^ *^"^H **^™•V X""^ <^~** QJ ^ -t-> CO Q. ~O O <d- O r- O LO ocoor-oooocn s-co ore c: O LO CD r**~ O LO ocnovoor-^ovo ocu o • r**H ^ r*** o cu CO CM i — — r— CM <Tt CM ^" OO O i — Cn ^" M— -C: CD cn O r-. 00 P-. cn oo r- oo r— p- r— cue Lo•r^x O> 3S i — r— r^ t — r1— ^.^ r— -*_** r^ *^-* S- QJ CO r— •• ro S- n -r-i r— t^- rO 14- f= c -^^% ^^~^ ^•~^ x— ^ ^-> (^ n ro CL O ro QJ O CM o oo O •— OOOOCOO•— OOO 03C >>Q_ -a O CM CD OO O VO o oo ro cn O LO O CTl O O^ O CVJ tO T- QJ i <£ C oo ro o ro r— cvj r^ r*-* ^ LD > i QJ CU 00 CM CO r— i — i — CTlLOCOCMOOLO «t- ^S--•— u. s: &* OO r- 4-> -Q 4_ -a c E CO S- CO •o QJ 3 r— QJ TD c r— z: a. .,_ ro rO £ 4_ > > ra O O X ••- • co S- CO • t5^ IO 3 CO CO Q. oo l^~ LO CO 1(^ ^ CM O **• +-1 CTi — ro QJ i|_ • • • • • • • c: CU o Q. co ^^ •^ CM co 00 «3-^J-OCM cu o E>v- _a Q ~o s- -c: o +-> C ro U s_ >, i/) O QJ CO l|_ o CO CU o r-- o cn O r-- Oi — O^J- Or-. OCO CO Ir— TJ^S- 1 gj 0 0 CD P"-- o cn OVOOLOOOOLO QJ i — ro QJ 3 fo ( rt t~ cu CM CM O^ CM . — OOOJ LOr— VOO VOO i- i — VJ .>OCO O LI) c: f•*"- O VO to LO co vor— vo. — vor — a> f- -a * ^tl cu y~ CT* ^E r— r^ ^— i — i — r— i i — QJ >> (_) a> +-> -E U r— <^V ro OEs- c•OO-i-> o E -I-> ro 3 QJ o cu s- i — o «T-)VI- ' — * ^~- ^"** • — • H" "O O ro <__> ^ _Q o CU LO O vo O LO o cn ocnoi — at — o ^ cu co oj3 u s_ O OO O LO CD P"*- OLO OCM OCM OOO CD 4->« JZ4->COC c O C VO 1 — VO CM LO CO i — CM CMr— VOr- Oi — i — trtcocn 0ro O cu •-/") QJ cr>i — i — en VOCM r^-LO c •r->•,- -r- ro-^S-co •r— in z: VO CM VO r- VO r- o t2? •f» -C ro T- w co co 3 O 4-> CO O QJ QJ CO O • [ S- C S- Q- c£. O CO o Q. s_ QJ QJ QJ -o .,_ •r- CU CU>s_ r— CO4JT3 co 4•J 3 -Q T- ro 4-> CO cu C CO C CO C -C ••- «£ ra • &« 4- CO 3 i — 0 CT>-I-> i_ CO t_ oo r^ r— LO O vo O >ro•r-T-ro rO • • • • • • • _C rO •* 3 4_> CO 4-* • •r— -r- ^• CM r*** i — ^- O ^- -I-> j:co"Oro5to - — - ro OO CO O 3 CO CD > ~ CO CO ro JZ CO •!- " s_ .C T- >> ,— ••- ' — * ,— ^. • — * ^^* *~** *^~* CO 4-* r-~ ~O f/} ro 1 — -O £ O LO O vO 0 0 o** ovo oco CDr — ~o c r— c +j c cu o o O ^O o ^i- ovo OLO ocn oro i — oo•r- ecu CU vo.— or-~ rovo cnvo cu EU ojr— -co LO Osi a*^-' ^3 ro E *•-—- * A - _.- •** — * * — ' *-• "^ — ' *r- 4— ,r_ r^ r_j, £- o r-- o ^- ^ ^~ LO LO LO ** M- OS-Oc3.rOC:O ro cn 3 1 — ^~ ( — i — i — . — i — i — rO •r- •t->•r- O. U. r— •t^• S- 1 CO S- 0 S- CO o cn>iQj cucocucu C V|- 1 JD Q; -r- QJ t- ro ^— * .-— *. ^-— -, ' — * - — * ^~- * — • >- V- E c~ f^ i O CM o r— O CM CD i — o ro o cn o cn ~o 033 cu +J ••— c -— o r-^ o ro o ro ovo oo ocn ovo cu v*_oc ^_> cno CO C ro r-^ ro vo ( — p~- «d-O VOOO K3-00 LOr— -I-> <+- rOCCC 2 CO f~-i — LO^l- LOCM LO*d" r— CU +J CO o ' •• z: LO CM LO r— LO P— ^ r*^ *fr 0 S- -C O 0 C CD CT\ r— ro *O CT> o *r- ro • (/> r-* ra i — CO •r- O •t-> CO *n i VJ ro >) QJ (/> CO - — -— ^ i c oo co O 3 4_ .,— (_> <u r^ -*-' i — o 4-> c cn co OUQ_rjJroajC S- cn CO • </I C T- E +->>,+-> •r- •r- t— r- 0 •r- CO ECOrocUcOOr- C 0 0 COCUU>cOQ. QJ OO C J-TJCO•r-S-dj^ 0 * I/I ••— CO COrOET33^;cro cn 4-> o 0 r— >> T- S 0•r-CoOh-•r-CO C •r~ >^ 00 (0 Cn O <C 4-> -r- • cu +J L. O O OO CO --J- E ro CO 4,J •— •r- •— c >•) e U CO ra CD o i — ro • • cu cu •o o cu •r~ • 1 — <J O .E ro T- co u r— i — r" CO E •r- r- O •r- T3 CO (^ CO Q. •4-J >5 CU T3 O >, 0 CO 4-> 3 rO •r- E ro .c r~ cu ••- (/> o z: o o U_ UJ z: Q- <-J :: CQ Q. oo * -ZZ oo 92

TABLE 4.20 Median Annual Salaries of Full-Time Employed Assistant Professors Holding Science Doctorates by Sex, Carnegie Classification* of University, Field, and Ph.D. Cohort, 1977 Field* Men Women Diff. Men Women Diff. Mathematics $18,500 (337) $18,400 (27) 0.5% $17,300 (91) — Biology 19,800 (711) 18,800 (127) 5.5% 17,500 (100) — Psychology 18,900 (296) 18,300 (116) 3.2% 17,400 (123) $18,000 -3.4% (80) Sociology 19,500 (714) 19,300 (196) 1.0% 18,300 (499) 17,000 7.1% (33) Research Universities I I. Biology $20,000 (354) $18,000 (44) 10.0% $17,400 (129) Sociology 18.900 (375) 17.900 (57) 5.3% 17,600 (232) 17,500 0.6% (101) Other Institutions Mathematics $18,300 (798) $17,400 (87) 4.9% $16,100 (493) $15,700 2.5% (47) Physics 17,700 (434) 17,400 (28) 1.7?, 14,500 (140) Chemistry 18,100 (780) 15,900 (104) 14.4% 17,100 (221) 17,000 0.6% (35) Medical Sciences 20,700 (320) 20,300 (79) 1.9% 19,000 (143) ... Biology 18,400 (1.773) 17.500 (394) 4.9% 16,500 (506) 16,000 3.0% (155) Psychology 18,200 (1,080) 18,000 (420) 1.1% 16,800 (652) 16,400 2.3% (212) Sociology 18,400 (1,851) 17,800 (456) 3.3% 17,600 (1,609) 17,100 2.9% (425) Research Universities I 1970-74 Ph.D.'s 1975-76 Ph.D.'s *The Carnegie classification, like the AAU and R & D rankings, rates entire institutions of higher education by indices assumed to measure quality. It distinguishes doctorate-granting institutions in terms of federal financial support and number of Ph.D.'s awarded. The two categories rated highest. Research Universities I and Research Universities II, contain 52 and 40 uni- versities respectively (Carnegie, 1973, pp. 1-7). (Used with permission. Copyright© ^976 by the Carnegie Foundation for the Advancement of Teaching.) '••The fields shown are those in which the number of women assistant professors was sufficient to permit a breakout by classification of university. — Indicates fewer than 10 sample individuals reporting salary. Note: Academic salaries for 9-10 months have been adjusted to full-year equivalents. Source: Survey of Doctorate Recipients, National Research Council. The statistics in this table are weighted estimates, derived from a sample survey of 65,000 Ph.D's in science and engineering. The estimates are subject to two types of error - sampling and non-sampling, (e.g., non-response bias). A discussion of the survey is provided In Appendix D. 93

recent Ph.D.'s. When sex, marital status and employment sector were controlled in the NRC survey of 1971-1975 biomedical and behavioral scientists,1 it was found that within each field, the median salaries of single men and women and married women did not differ greatly or consistently but in both fields and in every employment sector, married men (who are most Ph.D.'s) earned more than the others. The differences are shown in Table 4.21. Within academic institutions, married men had a distinct salary advantage. Discussion of Findings In this chapter, we examined recent trends in the status of women on science faculties with respect to changes in numbers, rank, tenure, and salary, and have given some attention to changing distribution along types of institution. Complex inter-relationships among these factors make it difficult to define and evaluate real change. Equality in employment, promotion, and salary presupposes that individuals are comparable in ability, and in most universities and many colleges, ability is defined largely in terms of research productivity. The latter is critically dependent, in the case of experimental sciences, on the particular institutional environment. Studies comparing research productivity of men and women show little or no difference when properly controlled for field and institutional category (in addition to the more obvious variables of age, rank, etc.). but none have taken account of uniquely important factors such as accessibility of research funding, graduate students, or research assistants. The majority of women scientists are not in the institutions where most research is done, and not in positions with opportunities for independent research initiatives. Nonetheless, it is clear that the increase in women Ph.D.'s which began in the sixties has been followed by an increase in their presence among science faculties. The gains in total numbers of new academic positions for women scientists, and especially the gains in tenured positions, are modest, and indeed almost invisible when viewed as a fraction of the total of approximately 120,000 faculty positions. But in the top research universities, with a steady or declining population, there have been some quite real increases, and that fact has importance beyond the magnitude of the numbers. These are the institutions which are the pace-setters, and the finding that women account for 35 percent of the growth in their science faculties since 1973 is important. In view of the very limited numbers of available positions in most of these departments, and of their very high quality standards, the large fraction of women in these new positions is evidence of a different climate of opportunity. Perhaps the most important aspect 94

M M 0 0) P X! 0 P 0) O en <c M J rH O 0 "^ M 83 0) 3 •O n en C 0 ^ 1 S •rH p V| . 1 ,— -, r~1 • • 3 B cS £ w S- Hj 0 II i m -P o id 0) £ -P g rrt 3 -rl •H (1) M H >i ' 0 H to id -_ S rH H Id rH p. -P Q - • X 3 ? 0 fc t§ H .C CO CM en r* XI to rH 10 rH l -P id rH C M (^. Q) l O SB -p -p co o C 0 0 «*-! -rl 04 T> ° a in a 0) < •H id id rH rl <rJ o en 4J o •rl !S •O 10 0) O S PH C\j • o o o o VD VD > - * o cn CN rH O O o o m <n ^ *. n <N CN CN o o o o rH CN CN O CN CN O O O OOO \Q CO CN •W- O O O o o o CN m o O CO O CN rH CN O O O o o o CN O CN o> -P (0 0) 0) id •H rH •P tH DI -H U C rl * * * o o - * * rH CN O O O O CN ^r " - * O rH CN CN O O o o 10 en o o o o i-H CN 04 CO 9) O (U •rl O CO rd rl O •H CO I oo oo o o o o en m •O rH a> 0 id •rl rH 4J rl CT-H H C H a -ri id 6 S. O O o o o o o o 00 vO o in " •- -K » - * o en o oo CN rH CN •-< O O O O O O m o en H *• » * - * * •H VTl rH en CN CN H C 0 o O O O O - _rj o o O 0 TJ y vo en CN in 0) rl •• - * - - * U id H en en o M 0) CN rH rH CN a in 0) K rl o o o O O 4J id o o o O O o c *D ^* O CN en C 0 * * • - " * •H (Ti 00 O oo r-- • -P rH rH CN rH rH b id id z to , in in -P C CO Q •H O O O o o •P O O O O 0 to C H vO VD 0) 0) - * * - - * M •H o en o oo oo 0 CN i-H CN H rH ^ CO 0) E rH n> 3 M to o •H o ^> 0 0 O O CN id o o O O c (1) CN CN m CN id m - - * • - * M rH rH rH -P •o rH rH rH rH C rl id (1) s rH id C g| •H 5 ^ c T) § O o 0) •g C a) •H D D in A s in id 4-4 3 o -p 4J in id id cu ^ -p 4J cu en CO •rl E rl id Q) 0) Id 0> Q) Id rH 3 id en •H rH 4-l •H rH .P 10 M CM -rl M Cn -H rl C M |s«a O flJ *H fl3 § • • H S en S •H X •d cu D as 0 •X C0 95

to note is that we are witnessing the gradual reversal of a very old tradition in these institutions. These changes have not occurred uniformly in all fields. In earth sciences, where total numbers of women are still very small, they have increased dramatically; in engineering, with still smaller numbers, the increase is almost as large but does not attain statistical significance. The smallest changes, which are however also statistically insignificant, have occurred in chemistry and mathematics. In the latter two fields the proportions of women faculty at leading institutions are strikingly at variance with the relatively large doctoral pool. A very few, very recent appointments in leading chemistry departments (Rawls and Fox, 1978) may, however, be the beginning of a new trend. It has been suggested that the apparent persistent sex biases in chemistry and mathematics departments may result from different distributions of the sexes among subfields. For example, if women chemists disproportionately prefer the "softer" parts of their discipline or women mathematicians the less rigorous fields of theirs, then they may not in fact be eligible for the job openings which arise. The available facts do not support this interpretation. An examination of the distribution of men and women doctorates among the subfields of the two disciplines in recent years (Harmon, 1977, Appendix A) shows them to be essentially identical except that in chemistry, women tend to slightly favor the specialties traditionally regarded as more rigorous, and to be significantly less likely than men to specialize in organic chemistry (about one-fourth of the women and one-third of the men specialized in this in 1974). A future study would do well to examine in detail the situations in mathematics and chemistry which seem to have special problems. In the great majority of colleges and universities that are not among the leading research institutions, the numbers of women have increased at comparable rates. Here, as in the top institutions, the proportion of women assistant professors has nearly doubled. Their rate of promotion to associate professorships corresponds roughly to the average proportion of women among assistant professors over the four-year period, while in the research universities that rate is very much higher. The lag in the granting of tenure to women, however, has persisted throughout the period under examination. It becomes most evident in the "other" category of institutions where the fraction of women promoted to full professorships in all the sciences is actually slightly below their already low presence in that group. A slower advancement of women at the "other" institutions may be, in part, due to a 96

transfer of the best women at these institutions to the more prestigious ones. Differences in the salaries of men and women faculty have followed the general trend toward equalization, though change in this factor is bound to be less rapid when women constitute large fractions of the new entrants to each rank. A notable exception occurs in chemistry, where the salary differential has increased for all ranks, and in medical sciences for full and associate professors. As already noted in Chapter 3, salary discrimination is especially serious because it is usually covert and cumulative. It is also hard to justify even on economic grounds: given the small numbers of women in most fields, the actual cost of correcting blatant inequalities would be small. Among full professors of chemistry, where the difference in median salaries of men and women is far greater than in any other category, the added salary costs (excluding benefits) for men between 1973 and 1977 were about $50 million; the difference between the actual salaries paid to women professors in 1977 and the potential cost if their median salaries had been equal to men's would have been about $1.2 million. Spread over four years and dozens of institutions, that cost is not prohibitive. More detailed studies of salary equity are beyond the constraints of this report. The data of the Survey of Doctorate Recipients, however, could well be the subject of a more extended analysis offering the possibility of relating salaries to length of service. Such a study is likely to shed some light on the flow-through or "revolving door" assistant professorship; if disproportionately high fractions of women assistant professors (compared to their hiring and promotion rates) are found to have entry level salaries, that would be an indication of abnormally high turnover. What of affirmative action policies? There is no doubt, in our judgment, that the existence of equal opportunity laws, regardless of their actual enforcement, is primarily responsible for the relative increase in the numbers of women faculty and also for the relative improvement in salaries, at least at the entry level. Current efforts to simplify and strengthen affirmative action enforcement have our full support. Particular attention needs to be paid to both tenure and salary equalization; internal initiatives with departments and institutions are preferable by far to legal action. The costs of the latter for institutions are likely to be higher than the cost of the necessary salary adjustments themselves. For aggrieved individuals, the price of legal action is often measured in irreparable damage to professional prospects as well as in dollars they cannot afford. 97

NOTES Private communications to Committee members. Among institutions that have expressed concern over this problem are Yale, Cornell, Indiana University, Princeton, the University of Minnesota, the University of California-all campuses, Stanford, and Purdue. Private communication. Higher Education Resource Services. This is not directly comparable to the ranking of institutions by R&D expenditures which we have used elsewhere, but serves as a reasonable approximation for this purpose. See the explanation of the classification in Appendix B-2. The salary tabulation was specially prepared for this report. 98