The influx of women into academic medicine over the past three decades has not been accompanied by equality for male and female faculty in terms of rank attainment, leadership roles, salaries, and treatment by colleagues or superiors. Although the percentage of female faculty in academic medicine has doubled over the last three decades, increasing from 13% in 19671 to 27% in 1999,2 women have not achieved parity with men in academic ranks or tracks. Only 6% of academic departments (170) were headed by women in 1998, and there were only six women deans at U.S. colleges of medicine in 1999.2 In addition to disparities observed in rank and leadership,3,4,5,6 women physicians receive lower salaries than their male colleagues.4,7,8,9 Possible explanations for these disparities have included lower productivity of women,5,8,10 conflict between familial and professional roles for women,11,12 and academic climates that support gender discrimination and sexual harassment.13,14,15
Most research on gender differences in academic medicine has used U.S. national level data, looking across and within specific disciplines. However, these studies have been hampered by a lack of data on academic track (i.e., tenure versus nontenure) and reliance on self-reported salaries. Our study reported findings of gender disparities at a single institution, the University of Arizona College of Medicine, Tucson. Focusing on one institution permitted the use of reliable institutional data on salary, rank, track, and productivity.
In 2000, at the request of a group of female faculty members, the dean of the College of Medicine created a committee of 19 women and three men, The Dean's Committee on Women Faculty, to assess the extent to which differences existed between the treatments of male and female medical faculty members. In 1999, the president of the University of Arizona had commissioned The Millennium Project, a study of faculty and staff perceptions of gender differences in all colleges of the university. The female faculty members in the College of Medicine, who had been meeting regularly, recognized that several issues faced by the medical faculty, in particular the provision of clinical service, were unique and deserved consideration in addition to the issues provided by the Millennium Project. The GRACE (Generating Respect for All in a Climate of Academic Excellence) Project was designed to complement the Millennium Project and provide a model for other colleges interested in an in-depth analysis of gender disparities.
The aims of the GRACE Project were to document gender disparities, test hypotheses for why they existed, and suggest ways to eliminate them. Areas for exploration included compensation, academic rank and track, levels of productivity, attainment of leadership positions, access to institutional resources and support, and perceptions of campus climate. We hypothesized that gender differences exist in salary, even after accounting for gender differences in rank, track, academic degree, specialties, and leadership positions. The dearth of women in higher ranks was postulated as being due to women's being less career-oriented than men or having greater difficulty balancing work and home responsibilities. Gender disparities in leadership positions might result from a lack of leadership aspiration or skills among women, an unwillingness to take on time-consuming leadership roles, or the fact that fewer women were in the pool of potential leaders (associate and full professors).
Given the variety of information required to test the hypotheses, we used two approaches. First, we obtained quantitative data on rank, track, leadership positions, and salary from the College of Medicine Appointed Personnel database for 1999–2000. Information was collected for 418 faculty members who met the following criteria as of July 1, 1999: 1) primary appointment in the College of Medicine; 2) on the tenure, clinical suffix, or research track; 3) rank of assistant, associate, or full professor; 3) ≥ 50% time; and 4) Tucson-based (some faculty from the College of Medicine practice in Phoenix).* Salary data was missing for five of the 418 faculty, reducing the total number of faculty to 413. We supplemented this information with data on productivity from two sources. The lifetime number of peer-reviewed publications was counted from current curricula vitae (CVs) provided by each department for 87.4% of faculty members (n = 361). We obtained clinical revenues generated during 1999–2000 for all faculty members who provided clinical services (n = 241) from University Physicians Inc., the faculty practice plan.
Second, faculty members were surveyed using a structured, online questionnaire to collect information about their ideas and experiences with leadership, promotion, professional and personal career issues, institutional support and space, resource procurement, treatment by colleagues, and leadership. We collected demographic data to characterize respondents. The questionnaire was designed by the Dean's Committee on Women Faculty in conjunction with the College of Medicine Division of Academic Resources to test the hypotheses about gender disparities. In November 2000, a letter from the dean was sent to all of the 418 eligible faculty members in the database, requesting their participation in the online questionnaire. We gave instructions for accessing the questionnaire, including a password to ensure that only eligible faculty participated. Over the next several weeks, we sent three reminders either by campus mail or by e-mail; we completed the survey in January 2001. To encourage participation, we donated a $100 restaurant gift certificate to one of every 100 respondents. All responses were confidential.
We analyzed both the salary data and the survey data using standard statistical software. Salary analyses were based on total remuneration, including base salary and incentives. Salaries were log-transformed to remove the effects of outliers, such as exceptional individuals whose salaries were substantially higher than those of their colleagues of comparable rank, track, and department. We analyzed questionnaire responses using standard statistical techniques, such as contingency tables, bivariate means tables, and multivariate analysis of variance (ANOVA).
A multivariate ordinary-least-squares regression model was used to estimate the effect of gender on salary, adjusting for the following possible confounders: rank and years at that rank, track (tenure, clinical, or research track), degree (MD or PhD), administrative responsibilities (section/division chief, department head), and type of “specialty” (basic science, generalist, nonsurgical specialty, surgical specialty). Basic science faculty were from the five basic science departments (biochemistry, cell biology and anatomy, microbiology and immunology, physiology, and pharmacology). Generalists included those in the department of family and community medicine, the Arizona Prevention Center, general medicine faculty, and ambulatory pediatrics faculty. Surgical specialties included the five highest-paying departments: surgery, orthopedic surgery, ob/gyn, anesthesiology, and ophthalmology. Nonsurgical specialties included all other groups: non-generalists (such as infectious disease) in pediatrics and medicine, and the departments of radiology, psychiatry, neurology, pathology, and radiation oncology.
A total of 198 faculty members completed the questionnaire. Table 1 shows the characteristics of the faculty members at the College of Medicine who were included in our salary, rank, and track analyses and the characteristics of our respondents. Table 1 also compares respondents with the larger population of faculty members (those in the salary, rank, and track analyses). Our respondents were significantly more likely to be female, be from basic science departments, be on the tenure track, and have a PhD.
Table 2 shows adjusted mean annual salaries for men and women at the College of Medicine and disaggregates them by type of department, rank, track, and specialty. On average, women earned $12,777 (11%) less than what men earned after adjusting for other factors. (The unadjusted difference was $36,768, 29% less than men's salaries.) Women's salaries were also significantly lower for most of the categories listed, and this disparity increased with rank, so that female full professors earned $23,764 less on average than men at the same rank. The differences remained statistically significant after adjusting for the number of peer-reviewed publications (p < .0001), with women earning $13,074 less than men for the College of Medicine as a whole. The observed differences also cannot be explained by differences in clinical activity since the gender difference remained significant (p < .01) after adjusting for clinical revenues. Finally, the salary disparities extended to most departments, including basic science departments (see Figure 1).
Rank, Track, and Time to Promotion
The distributions of male and female faculty members differed significantly according to rank (p < .00005) and track (p < .0005, see Figure 2). Women were predominantly in non–tenure-track positions at the assistant professor level (49%). In contrast, more than half of the men (55%) were promoted (associate or full professors) and tenured.
Based on questionnaire responses, women at the College of Medicine were significantly more likely to have considered changing academic tracks (58% versus 29%, p < .0001). This was particularly true on the tenure track, where almost half the women had considered changing tracks (46% versus 9% of men, p < .00001). However, women were no more likely than men to have actually changed tracks (19% versus 14% for men) or to have delayed the tenure clock (16% versus 10% for men).
Attaining the rank of associate professor took longer for women than for men. For the 97 associate or full professors who indicated on the questionnaire the numbers of years they had been assistant professors, the average time to promotion had been 6.5 years for women, versus 5.2 years for men (p < .05). There was no significant gender difference in lengths of time as associate professor.
A common explanation for the differences in rank between male and female faculty members is that women may be less serious about their careers than men and thus less productive. However, questionnaire responses indicated no significant gender difference in the reported importance of career advancement (p = .79) or the extent to which work and personal life conflicted (p = .78). Women who worked full-time were no more likely than men to want to work part-time (38% versus 27% for full-time men, p < .13). However, a higher proportion of men were willing to move to take a better job (71% versus 51% of women, p < .05).
Initial comparisons of the numbers of peer-reviewed publications for faculty suggested that men were more productive than women (see Table 3), regardless of whether the estimates came from faculty members' CVs or from self-reporting on the questionnaire. However, this was an artifact of gender differences in rank, track, and time in rank. When stratified by rank and track, gender differences disappeared, except among full professors on the tenure track. In this category, men published more, but they had been at that rank twice as long as the women had been (11.7 versus 6.4 years, p < .05). When adjusted for years in rank, the difference in mean numbers of publications for female and male full professors became insignificant (63.1 versus 78.0, p < .49). Also, equal proportions of women and men had received teaching awards (30.2% and 33.3%, respectively).
Clinical revenues by gender, adjusted for specialty and rank, are shown in Table 4. The difference in clinical revenues was significant for only two of the comparisons: full professors who were generalists (in which case the sole woman in this category made twice the average revenues for the men) and surgical specialists at the level of assistant professor, where men generated significantly more clinical revenues than women (p < .05).
Respondents' leadership experiences are presented in Table 5. Contrary to our hypothesis, there was no significant gender difference in self-assessed leadership ability or aspirations to hold leadership positions. Nevertheless, women were significantly less likely than men to have been asked to serve in leadership roles. They were also less likely to report ever having served as committee, section/division, or department chairs. When analyses were limited to associate or full professors, women were still significantly less likely to have been asked to serve or to have served as section heads or department chairs. More than one fourth of the men who responded to the questionnaire had been invited to serve as department heads, as compared with only 6% of women.
Given the differences in leadership opportunities, it was not surprising that the women felt they had significantly less power and influence in their departments. Half as many women as men reported having decision-making authority over promotion of colleagues (27% versus 48%, p < .005) or over non–grant-related resources (22% versus 47%, p < .001). Also, the women were significantly less likely to offer advice to department chairs (2.7 versus 3.2 on a 1-to-5 scale from never to always, p < .001) and to feel that they effectively influenced departmental decisions (2.5 versus 2.8 on a 1-to-6 scale from never effective to always effective, p < .01).
Institutional Support and Access to Resources
There was no significant gender difference in access to or difficulty obtaining resources, office space, or staff support. However, differences did exist with regard to research space. Of those who responded to the questionnaire, 60.8% of the women and 43.4% of the men reported sharing their research space with other faculty members. Among full professors, 72.7% of the women reported sharing research space with other faculty, as compared with only 40.0% of the men. After adjusting for rank, women reported exerting significantly more effort to obtain non–grant-supported research space (4.42 versus 3.87 for men, on a 1–5 scale, 5 = “a great deal of effort,” p < .01).
Treatment and Climate Issues
Women were significantly more likely than men to cite safety concerns, with regard to working at certain times or in certain places (see Table 6). They were also much more likely to feel that their departments treated men and women differently. Almost one third of the women reported being discriminated against, as compared with only 5% of the men (p < .00001). Only two of the 22 women who reported being discriminated against had sought recourse for the discrimination. A majority (68%) of these 22 women but only 15% of the men who reported being discriminated against felt the College did not respond appropriately to such charges. Finally, women were significantly less likely to feel that they “fit in” at the College of Medicine (72% versus 85%, p < .03).
Our investigation into the status of male and female faculty members in a College of Medicine indicated substantial gender differences in the rewards and opportunities of academic medicine. There were significant gender disparities in salary, with women earning 89% of what men earned. A plurality of women were non–tenure-eligible assistant professors, while the majority of men were tenured associate or full professors. In general, women were as productive as men, indicated by both publications and clinical revenues, despite having less research space and less influence in their departments. Although the women aspired to leadership positions and felt they had leadership skills, few had been asked to lead. Almost one third of the women reported experiencing discrimination. These differences support the widespread perception that academic medicine is not gender-blind and that the road to success is more challenging for women than for men.
Similar studies have been conducted at other medical institutions.16,17,18,19,20 Our study provides a more comprehensive picture of gender differences in academic medicine by examining a wide range of aspects of academic life beyond salary,16 promotion rates,17 or faculty perceptions of the academic environment,18,19,20 which have often been studied in isolation. Further, we incorporated a number of measurement techniques into our study, including institutional databases, questionnaires, and interviews, to present a fuller picture of gender disparities at the University of Arizona College of Medicine.
Our findings of gender differences in rank are similar to those of other analyses. These differences are often attributed to a pipeline effect,1,3,4,5,6 the notion that an insufficient number of women have been in academia long enough to have reached the rank of full professor. Yet, analyses of gender differences in rank attainment have not supported this hypothesis.1,3,5 An alternate explanation for the paucity of women at higher ranks is that they are less committed to their careers because of family and other personal responsibilities. In our study, the women were no more likely than the men to report conflicts between family responsibilities and work. However, other studies have shown that women are differentially affected by and concerned with family responsibilities.11,12,21 It is possible that there were gender differences in interpreting our questionnaire, with more men than women taking for granted the full support of their spouses in caring for children. Finally, gender differences in productivity have been postulated as another factor leading to women's slow progression through academic ranks. Our study found substantial unadjusted gender differences in numbers of peer-reviewed publications and clinical revenues, but once adjustments were made for rank, track, and years in rank, these differences disappeared. Although some widely cited studies showed that male faculty members publish more papers than female faculty members,1,5,8,10 these studies did not adjust for differences in track, rank, and years in each rank. The only other study that adjusted for confounding influences on productivity found no gender difference in numbers of publications or grants.4
The rewards of the system also appeared to differ by gender. Substantial gender differences were evident in salary, after adjusting for other predictors of compensation. Salary disparities have been documented in many specialties and both academic and practice environments.4,7,8,9 Our observation that significant gender differences in salary also existed for assistant professors suggests that the problem has not been solved.
Our study also showed gender differences in leadership opportunities. At the time this research was conducted, there was no female department head of any of the 19 departments in the College of Medicine, and only 10% of section or division chairs (four) were women. National data have revealed a similar absence of women in leadership positions.1,2 The paucity of women at higher ranks may contribute to the underrepresentation of women in high-level leadership positions, since the candidate pool of potential leaders remains small. Our study showed that a very high proportion of women in academic medicine believe they have leadership aspirations and skills. This suggests that leadership that is overwhelmingly male may not recognize leadership abilities in women or that women may abandon their aspirations without female role models.
Given these findings, it is not surprising that women were more likely to report differential treatment on the basis of gender at the University of Arizona College of Medicine. The literature on gender discrimination in academic medicine has indicated that subtle forms of gender discrimination are prevalent across the United States.13,14,15 Besides being illegal, gender discrimination and harassment can have detrimental effects on career advancement and satisfaction. Carr et al. reported that women who experienced gender discrimination reported lower levels of career satisfaction than did women who had not experienced discrimination.15
Our study had limitations. First, the respondents were not entirely representative of the larger faculty, and longitudinal data were not available to track changes over time. Further, the College of Medicine database did not contain accurate information about research grants, making it impossible to adjust for potential differences in research funding, which might account for some of the observed salary and rank disparities. Nevertheless, the fact that similar disparities have been observed at the national level suggests that these findings are not likely to be limited by the data.
Numerous solutions have been proposed to address gender disparities among academic medicine faculty.2,11,22,23 Regular monitoring of salaries, track assignments, start-up packages, and promotions would help to identify disparities and assess progress.22,24,25 A system for mentoring and advising junior faculty members early in their careers about promotion requirements will facilitate advancement.2,22,24,26 Institutional changes should be made to allow greater flexibility in tenure-track positions and in the definition of scholarly contributions, while maintaining a commitment to academic excellence. Promotion and search committees as well as influential individuals such as department heads and deans should be educated about subtle forms of discrimination. Perhaps the most critical element for ensuring change is assigning accountability.27 Progress in the recruitment, promotion, retention, and advancement of female faculty members to leadership positions should be rewarded. Departments with exemplary performances could be rewarded with additional research funds for faculty or additional faculty lines, while administrators who consistently impede the diversification of faculty must be penalized.25,26
In conclusion, our analysis suggests that male and female faculty members have substantially different experiences in academic medicine. In general, women's contributions are less likely to be recognized (either financially or through promotion) and their leadership abilities are more likely to be ignored. Changing these discriminatory patterns and behaviors will require persistent effort, open communication, continuous monitoring of progress, and a commitment to diversifying faculty and leadership. The alternative, in the words of Dr. Catherine DeAngelis, is to “waste half our genetic pool of intelligence, creativity and critical insights and experience. Medicine simply can't afford that loss.”24
1. Bickel J. Women in medical education: A status report. N Engl J Med. 1988;319:1579–84.
2. Bickel J. Women in academic medicine. J Am Med Wom Assoc. 2000;55:10–2.
3. Dial TH, Bickel J, Lewicki AM. Sex differences in rank attainment among radiology and internal medicine faculty. Acad Med. 1989;64:198–202.
4. Carr PL, Friedman RH, Moskowitz MA, Kazis LE. Comparing the status of women and men in academic medicine. Ann Intern Med. 1993;119:908–13.
5. Tesch BJ, Wood HM, Helwig AL, Nattinger AB. Promotion of women physicians in academic medicine: glass ceiling or sticky floor? JAMA. 1995;273:1022–5.
6. Nonnemaker L. Women physicians in academic medicine: new insights from cohort studies. N Engl J Med. 2000;342:399–405.
7. Baker LC. Differences in earnings between male and female physicians. N Engl J Med. 1996;334:960–4.
8. Kaplan SH, Sullivan LM, Dukes KA, Phillips CF, Kelch RP, Schaller JG. Sex differences in academic advancement: results of a national study of pediatricians. N Engl J Med. 1996;335:1282–9.
9. Ness RB, Ukoli F, Hunt S, et al. Salary equity among male and female internists in Pennsylvania. Ann Intern Med. 2000;33:104–10.
10. Levey BA, Gentile NO, Jolly HP, Beaty HN, Levey GS. Comparing research activities of women and men faculty in departments of internal medicine. Acad Med. 1990;65:102–6.
11. Fried LP, Francomano CA, MacDonald SM, et al. Career development for women in academic medicine. JAMA. 1996;276:898–905.
12. Carr PL, Ash AS, Friedman RH, et al. Relation of family responsibilities and gender to the productivity and career satisfaction of medical faculty. Ann Intern Med. 1998;129:532–8.
13. Lenhart S, Klein F, Falcao P, Phelan E, Smith K. Gender bias against and sexual harassment of AMWA members in Massachusetts. J Am Med Wom Assoc. 1991;46:121–5.
14. Frank E, McMurray JE, Linzer M, Elon L. Career satisfaction of U.S. women physicians. Arch Intern Med. 1999;159:1417–26.
15. Carr PL, Ash AS, Friedman RH, et al. Faculty perceptions of gender discrimination and sexual harassment in academic medicine. Ann Intern Med. 2000;132:889–96.
16. Holmes-Rovner M, Alexander E, O'Kelley B, et al. Compensation equity between men and women in academic medicine: methods and implications. Acad Med. 1994;69:131–7.
17. Nickerson KG, Bennett NM, Estes D, et al. The status of women at one academic medical center: breaking through the glass ceiling. JAMA. 1990;264:1813–7.
18. Hostler SL, Gressard RP. Perceptions of the gender fairness of the medical education environment. J Am Med Wom Assoc. 1993;48:51–4.
19. Foster SW, McMurray JE, Linzer M, et al. Results of a gender-climate and work-environment survey at a midwestern academic health center. Acad Med. 2000;75:653–60.
20. Buckley L, Sanders K, Shih M, et al. Obstacles to promotion? Values of women faculty about career success and recognition. Acad Med. 2000;75:283–8.
21. Levinson W, Tolle SW, Lewis C. Women in academic medicine: combining career and family. N Engl J Med. 1989;321:1511–7.
22. American College of Physicians. Promotion and tenure of women and minorities on medical school faculties. Acad Med. 1991;114:63–8.
23. Association of American Medical Colleges. Increasing women's leadership in academic medicine. Acad Med. 1996;71:800–11.
24. DeAngelis CD. Women in academic medicine: new insights, same sad news. N Engl J Med. 2000;342:426–7.
25. Massachusetts Institute of Technology. Special edition: a study on the status of women faculty in science at MIT. The MIT Faculty Newsletter 1999;XI.
26. Yedidia MJ, Bickel J. Why aren't there more women leaders in academic medicine? The views of clinical department chairs. Acad Med. 2001;76:453–65.
27. Morahan P, Voytko M, Abbuhl S, et al. Ensuring the success of women faculty at AMCs: lessons learned from the national centers of excellence in women's health. Acad Med. 2001;76:19–31.
*The clinical suffix track is a non–tenure-eligible track for clinician–teachers employed full-time at the College of Medicine. Research track is a non–tenure-eligible track for researcher–teachers. Phoenix-based faculty members were excluded because the University of Arizona does not supply salary support for them.