Applause should go to a good first act in which medical schools have improved significantly the numbers of women students and faculty. Women now average 29% of faculty and 48% of students in U.S. medical schools in comparison with much smaller numbers 20 years ago.1 Over the past decade, academic medical centers have invested considerable effort and resources to attract and retain women, and these investments have paid off. Close to 50% of instructor-level faculty are now women. A number of schools have paid increasing attention to helping women faculty build their careers. Initiatives have included appointment and support of Association of American Medical Colleges (AAMC) Women Liaison Officers, undertaking salary and promotion equity studies, and offering local skill-development programs.1
Medical schools must now direct attention to the second act. Women still lag behind men in leadership roles within academic medicine; women hold less than 10% of senior administrative positions in medical schools. As of the writing of this paper, there were only nine women deans of the 126 U.S. medical schools (only 7% of the full deanships).2 Women average about 1.7 department chairs per medical school (8%). At least 20 of 126 medical schools had no women chairs in 2001; most of these schools have never had a woman chair. Finally, women also hold only about three assistant, associate, and senior associate dean positions combined per medical school, less than 10% for most schools.
Studies document that the scarcity of women in senior leadership positions is not connected to their time in careers, their levels of grant funding and publication, the hours worked, or their specialties.1,2 The reasons for this scarcity are multifaceted. Yedidia and Bickel3 identified three main barriers to women's advancement into leadership roles in academic medicine: traditional gender roles, sexism, and the scarcity of effective mentors. These barriers mirror those identified in corporations4–6 and in government.7 Research shows that many women are routed into staff positions (e.g., human resources, legal counsel) that prevent them from getting the line experience (i.e., profit and loss associated with chair positions) needed to move into senior leadership roles.4,5 Moreover, letters of recommendation for appointment and promotion actions tend to favor men over women faculty.8
Medical schools have paid scant attention to preparing women faculty members for advancement into the administrative leadership ranks (e.g., division chief, department chair, associate and senior associate dean, dean, vice president, president). National leadership programs are one mechanism through which men traditionally have participated as part of their career progression. The AAMC, disciplinary societies, and universities sponsor many such programs.9 There is also an increasing number of internal leadership programs aimed at enhancement of leadership development of both men and women within a given medical school, university, or university system.10,11 Research, however, has indicated substantial benefits of single gender or ethnicity leadership development programs.12,13 To help increase the number of women in administrative roles, starting in 1988 the AAMC Office of Women in Medicine began offering three-day faculty development programs targeted to early-career and then to midcareer women faculty.
Within the category of national leadership programs, the Hedwig van Ameringen Executive Leadership in Academic Medicine (ELAM) program provides a unique gender-specific approach. ELAM started in 1995 at MCP Hahnemann University School of Medicine (now Drexel University College of Medicine) in Philadelphia. ELAM provides women faculty members in medical and dental schools in the United States and Canada with leadership skill development, mentoring, and networking aimed at helping them to advance into and succeed in formal and informal senior leadership positions. Most of the approximately 45 participants each year hold associate or full professor rank. Each year a new cohort meets in September for a week, convenes again at the AAMC annual meeting, and then meets in April for a final week. The curriculum focuses on building knowledge and skills in seven domains: paradigms of corporate, government, and academic leadership; financial management; strategic planning and organizational transformation; emerging issues in academic medicine; communication; personal dimensions of leadership; and career advancement strategies. Between sessions, participants complete and submit several individual and group assignments. The applicant's dean or a senior designee of the dean must agree to mentor her and to take part in the final two days of the program. Upon completing the ELAM program, many alumnae choose to continue their association with members of their own and other classes through membership in a separate organization, the Society for Executive Leadership in Academic Medicine (SELAM).
From its inception and with support from various foundations, ELAM has included a robust evaluation agenda to measure its impact on participants and institutions, and has published preliminary findings with the first cohort.14 This report is the first in a projected series documenting the impact on several cohorts. In this paper, we describe the changes in ELAM participants’ perceived leadership capacities over an 18-month period, measured through a time-series evaluation design using a pre/post instrument.14,15 The data reject the null hypothesis that there is no difference in ELAM participants’ perceived leadership capacities with the leadership program intervention over this period.
The project follows a classic pre/post, within-group, self-report design. On the first day of ELAM, all participants complete a “pre” survey instrument. Approximately 18 months after the beginning of ELAM (or about 11 months after the final session), all participants are sent a “post” questionnaire. In this study, we used the data collected from 79 participants in three ELAM classes: 1997–98, 1998–99, and 2000–01 (Classes 3, 4 and 6; data from Class 5 were lost because of damage to the archived material.). The response rates to the “post” data collection ranged from 69% to 76% for each of ELAM's classes.
The instrument was created after the conclusion of the first ELAM class and piloted with members of the first and second classes. The draft questions were developed using content analysis of the objectives of each individual unit in the ELAM curriculum as a guide. The ELAM Advisory Committee and faculty also validated the instrument, critiquing the items for their relevance and accuracy in describing the program's curriculum and purposes. Both pre and post instruments elicit self-perception about leadership in the seven domains addressed in the ELAM program curriculum. Approximately 75% of the items use a seven-point semantic differential scale, with end points being 1 = strongly disagree and 7 = strongly agree. An example of this type of item is: “I can explain traditional and evolving organizational and governance models at academic health centers.” The remaining 25% of the items ask respondents to indicate their confidence in executing various leadership activities on a scale of 0 (least confident) to 100 (most confident). Items include “managing a budget” and “readiness to move to a leadership position.” During data analysis, the scores on these confidence items were divided by14.3 to approximate the seven-point semantic differential scale.
The questions represent ten leadership constructs: knowledge of leadership, management, and organizational theory; environmental scanning; financial management; communication; networking and coalition building; conflict management; general leadership skills; assessment of strengths and weaknesses; acceptance of demands of leadership; and career advancement sophistication. The post instrument repeats almost all the items from the pre instrument, while also incorporating several additional questions about the usefulness of the ELAM program and the demographic characteristics of the participant and the institutional environment in which she works.
We computed average composite indices for each construct using the arithmetic means of all individual questions related to that construct. The tables show the average composite scores and give examples of some of the individual items that make up each leadership construct.16
The number of responses for the individual questions ranged from 73–79 because some of the participants failed to answer all the items on both questionnaires. Accordingly, the number of responses for the average composite indices of the ten leadership constructs ranged from 68–75.
We analyzed the pre/post differences using the paired comparison t test with a two-tailed test of significance. Since 77 matched-pair t tests were computed at the .01 level, the cumulative alpha error suggests that one rejection of the null hypothesis may have occurred by chance. To be conservative, there was no application of the finite population correction factor. Results are described as statistically significant only if the p value was less than .001.
The results show increases in perceived leadership capabilities across all ten leadership constructs, between the time that participants began the ELAM leadership program and about 18 months later. These improvements were demonstrated for almost all variables within each construct.
Table 1 lists the composite score difference and examples of questions from the three constructs of perceived knowledge of organization, management, and leadership theory; environmental scanning; and financial management. ELAM participants exhibited large gains on knowledge of organization, management, and leadership theory. All seven questions within this construct showed large, statistically significant differences (ranging from 1.4–2.1 on a seven-point scale).
On the construct of perceived financial management capabilities, ELAM participants also showed large gains, with all questions showing substantial and statistically significant results. ELAM participants also showed substantial improvements on the construct of perceived ability to scan the environment for forces that might affect their institution, with all four questions showing significant differences.
Table 2 shows the composite score difference and examples of questions from the three constructs of communication, conflict management, and networking and coalition building. Self-perceptions of gains were statistically significant, though more modest in magnitude than for the constructs in Table 1. In the construct of perceived conflict management capacity, there were moderate to large results for changes in all three questions. In the networking construct, ELAM participants perceived gains both in being mentored and in mentoring others. In the communication construct, the lower perceived improvement is primarily because respondents initially reported high levels of skill.
Table 3 lists results of perceived gains on a large construct, general leadership skills. ELAM participants exhibited consistent increases, with all being statistically significant. The largest improvement was in perceived use of strategic planning. There also were substantial increases in understanding how to work with people of different work styles and in leadership participation. Two other questions of a different sort (not shown in the table) asked “in a rough estimate, how many of your work ideas have brought tangible results inside your unit (or outside your unit) during the past 12 months?” There were small increases for inside the unit (from 3.53–3.72) and outside the unit (from 2.72–2.89), but these were not statistically significant.
The ELAM participants also were asked about three constructs important in strategic career planning (see Table 4). The ELAM participants exhibited moderate to large improvement on several questions about their career advancement sophistication. When asked questions about the construct of self-assessment of their strengths and weaknesses, results also indicated moderate to large improvements. When asked questions about the construct of acceptance of the demands of leadership positions, all but one of the eight responses were statistically significant. The response to the statement, “I enjoy working in an administrative position” showed an average initial rating of 5.83, among the highest rated items on the pre questionnaire, and a significant gain on the post questionnaire. Only the response to “preparedness to deal with conflicts between personal and professional roles” was not statistically significant (p < .016).
The goal of ELAM is to increase the number of women in leadership positions, not merely to increase their knowledge and skills in leadership. Thus, on the post questionnaire, participants were asked several additional questions about how useful the ELAM program had been in their career development (see Table 5). Although the scales varied from question to question, all were seven-point scales with 1 representing the equivalent of “very low” and 7 representing the equivalent of “very high.” The responses indicated that most thought the program had been of high quality and of substantial use in their career. The lowest results were on two questions asking about participants’ perception of whether the ELAM program had impacted their desired career projection. This may be because the post questionnaire was sent out only 18 months after participants had started (or just 11 months after participants completed) the program.
To document that the perceived usefulness of the ELAM program was associated positively with gains in perceived leadership skills, we performed multiple regression analyses (see Table 6). Each postmeasure composite score was predicted by the corresponding premeasure composite score and by the composite score of the perceived usefulness of the ELAM program. We checked the data to ascertain that they met the assumptions of normality and linearity, and there were only a few modest departures from normality. In all ten regressions, the perceived usefulness of the ELAM program was positively and statistically significantly associated with the postmeasures of the leadership constructs after controlling for the premeasure of the corresponding construct.
The consistency of improvements in the ELAM participants’ self-perceptions of leadership knowledge and skills is striking. All of the ten constructs, and 60 of the 65 individual measures, showed statistically significant improvements over the 18-month interval following the initial data collection at the beginning of the ELAM program. Significant improvements were reported even when the ELAM participants viewed themselves as already skilled prior to attending the ELAM program.
These gains in leadership knowledge and skills are consistent with the outcomes of the ELAM program in its eight years of assisting women to advance to and be successful in top leadership positions in academic health centers. Of the 126 full deans of U.S. medical schools, ten are women; four of those are alumnae of the ELAM program. ELAM graduates now comprise 30% of all the women full deans of the U.S. and Canadian allopathic and osteopathic medical and dental schools. Over 75% of the graduates of the first three classes have obtained significant senior positions, including department chair, center director, senior associate dean, and university vice president. In addition, graduates have moved into leadership positions in health-related fields, including health care foundation, hospital, or pharmaceutical president or vice president.
The greatest gains in self-perceived knowledge and skills occurred in three constructs in the surveys: knowledge of organization, management and leadership theory; financial management; and environmental scanning. The large perceived composite gains (ranging from 1.47–1.91 out of 7 points) for each construct are not surprising. First, the pre-ELAM self-reports for these three constructs were among the lowest of the ten constructs. Secondly, the ELAM curriculum emphasizes these topics, with special emphasis on the financial area in which women often feel less self-confident than do men or in which women are stereotypically viewed as being less knowledgeable. Approximately 25% of the curriculum is devoted to financial topics. Moreover, much of the curriculum focuses on enlarging the perspective and leadership knowledge of participants—through interactions with other participants from different schools and disciplines, interactions with a diverse group of faculty experts and leaders in academic medicine, attendance at the AAMC annual meeting, intersession assignments that require participants to learn more about the structure and function of key administrative offices in their organizations, and interaction with their deans during the course of ELAM and at the ELAM Forum.
The ELAM participants also reported substantial self-perceived gains (ranging from 1.18–1.52 out of 7 points) in four other constructs: conflict management, networking and coalition building, self-assessment of strengths and weaknesses, and career advancement sophistication. The results regarding career advancement sophistication are consistent with the ELAM program's emphasis on individual in-depth professional assessment and career planning; the ELAM program provides more such emphasis than any other program in academic medicine. Each participant takes the Center for Creative Leadership's Benchmarks® Instrument and the Myers-Briggs Type Indicator; has individual sessions with certified consultants; has two career consultations with leaders from academic medicine; and attends a day-long session on strategic career planning. Equal attention is paid to development of a close network of colleagues in a small learning community within their ELAM class, with all the class's participants, ELAM alumnae, faculty, and leaders in academic medicine, and leaders at the participants’ academic health centers. The increased perception of knowledge and skills in conflict management is probably related to the increased self-understanding from the above self-assessments, along with gains from a full day spent on developing skills in conflict management and a half-day on building a diverse community.
There were significant, although lesser, self-perceived gains (ranging from 0.84–0.97 out of 7 points) for three constructs: communication, general leadership skills, and acceptance of the demands of leadership. These results may be related to the fact that these constructs showed the highest initial pre-ELAM composite scores. Thus, the total possible gains were small; the post-ELAM composite scores for these three constructs ranged from 5.75–5.93 out of 7 points, among the top four post-ELAM scores for the ten constructs.
Only five questions out of the 65 in the survey did not show significant self-perceived improvement. Two questions involved interaction with the dean. When entering the ELAM program, participants already perceived considerable comfort working with their dean (6.63 out of 7 points); thus, there was very little room for improvement. However, it is not clear whether participants were considering their relationship with the full deans or associate deans. When asked how many times in the past 12 months participants had interacted one-on-one with their deans, the participants reported increases from an average of 9.95 times prior to ELAM, to 17.13 times 11 months after completing the ELAM program. Although this was a 72% increase, the difference was not significant because of the large standard deviation.
There were interesting responses to questions relating to perceptions of balance between personal and professional roles. ELAM participants did not perceive increased skill in dealing with conflicts in managing these roles. This may reflect the fact that these women already had reached some peace with the tradeoff required for balancing personal roles and leadership positions, because their pre-ELAM scores were already high (5.72 out of 7 points). The ELAM participants, however, did report significant increased skill and confidence in considering the personal and familial ramifications that might be required if they left their current roles for a higher-level leadership role. Similarly, the Center for Creative Leadership's Choices and Tradeoffs Facing High-Achieving Women study findings suggested “that women view life holistically and that lessons learned in one aspect of life spill over into another.”17,18
Our data have several limitations. The cohort response rates to the post questionnaire were 69% to 76%. Although this is an excellent response rate for this type of study, there is the possibility of some nonresponse bias, with participants who changed the least, or not at all, perhaps being less willing to reply to the post questionnaire. While there probably were some such cases, the most likely explanation is that the nonrespondents were just too busy with more pressing demands on their time. Because of resource limitations, we did not have an intensive follow-up process. In recent survey research with the ELAM alumnae, our evidence shows that time pressure appears to be a major reason for lack of timely completion of surveys, and we have instituted a more extensive follow-up design.
That participation in the ELAM leadership experience caused the perceived increase in leadership skills cannot be concluded with assurance given the obvious limitation of a one-group, pre/post design. Clearly, the ELAM participants partly self-selected themselves into the program on the basis of their administrative ambitions, which in turn may be partly a function of their perceived success and satisfaction with their initial low-level administrative experiences. Selection bias cannot be excluded since the participants, who were sufficiently interested in leadership to apply to ELAM, were not compared with a group who were not interested in leadership advancement. It seems unlikely, however, that these faculty members would improve on all ten of the measured constructs by natural maturation alone. Moreover, the immediate “halo” effect associated with program completion probably would dissipate by the follow-up survey conducted 11 months after “graduation” from the program. There was plenty of time for ELAM graduates to become disillusioned, particularly if they did not find ready opportunities to apply their new leadership capabilities and achieve some success in those roles. That most participants were also the beneficiaries of other leadership development interventions during the same time frame seems unlikely, given participants’ already large time commitment to ELAM during this period.
It is important to note that survey data can provide only one perspective on how the ELAM program impacts women and their leadership development. Other, more naturalistic research will be necessary to explore the context underlying the responses. There have been very few longitudinal quantitative or qualitative studies19 addressing leadership development in general (of men or women) and none with women in academic health. We have been interviewing ELAM alumnae at various points after completion of the program. Our initial results identified six major themes: understanding new leadership strategies; introducing and implementing ideas more effectively and confidently; increased confidence and knowledge in handling conflict situations; increased awareness and skill in building networks; increased knowledge and awareness of a wider range of career possibilities; and increased insight and confidence into “how the game is played” and how to play to succeed.14 There is an obvious meta-theme of increased knowledge and networks, leading to increased confidence. The whole arena of how women develop self-confidence in their leadership abilities and readiness to assume a place at the leadership table will be one focus of our ongoing qualitative research as we probe for the layers of meaning beneath the statistical results reported in this article.20–23
This work has been supported in part by the Jessie Ball duPont Fund, Connelly Foundation, and The Robert Wood Johnson Foundation. We thank Ray Francis, Sharon Kim, Victoria Odhner, and Elizabeth Verost for superb assistance with database design and implementation, data entry and validation, and statistical analysis. We also thank Jennifer Humphrey for her work with Dr. McDade in design and validation of the original survey instrument. We appreciate review of the manuscript and helpful suggestions provided by Hisashi Yamagata, Nancy Gary, and Janet Bickel.
1. Bickel J, Clark V, Yamagata H, Lawson RM. Women in U. S. Academic Medicine Statistics 2002. Washington, DC: Association of American Medical Colleges, 2003.
2. Bickel J, Wara D, Atkinson BF, et al. Increasing women's leadership in academic medicine: report of the AAMC Project Implementation Committee. Acad Med. 2002;77:1043–61.
3. 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.
4. Catalyst. Perspective: Report on 2002 Catalyst Census of Women Corporate Officers and Top Earners of the Fortune 500. New York: Catalyst, December 2002.
5. Morrison AM, White RP, Van Velsor E, Center for Creative Leadership. Breaking the Glass Ceiling: Can Women Reach the Top of America's Largest Corporations? Reading, MA: Addison-Wesley Publishing Company, 1992.
6. Rowe MP. Barriers to equality: the power of subtle discrimination to maintain unequal opportunity. Employee Responsibilities and Rights Journal. 1990;3:153–63.
7. Yamagata H, Yeh KS, Stewman S, Dodge H. Sex segregation and glass ceilings: a comparative statics model of women's career opportunities in the Federal government over a quarter of a century. AJS. November 1997;566–632.
8. Trix F, Psenka C. Exploring the color of glass: letters of recommendation for female and male medical faculty. Discourse Soc. 2003;14:191–220.
10. Morahan PS, Kasperbauer D, McDade SA, et al. Training future leaders of academic medicine: internal programs at three academic health centers. Acad Med. 1998;73:1159–68.
11. McDade SA, Lewis PH (eds). New Directions for Higher Education: Developing Administrative Excellence: Creating a Culture of Leadership. No. 89. San Francisco: Jossey-Bass, Fall 1994.
12. Ohlott PJ, Hughes-James MW. Single-gender and single-race leadership development programs: concerns and benefits. Leadersh Action. 1997;17:8–12.
13. Ohlott PJ. Change and leadership development: the experience of executive women. Leadersh Action. 1999;19:8–12.
14. Richman RC, Morahan PS, Cohen DW, McDade SA. Advancing women and closing the leadership gap: the Executive Leadership in Academic Medicine (ELAM) program experience. J Womens Health Gend Based Med. 2001;10:271–7.
15. Weiss CH. Evaluation. 2nd ed. Upper Saddle River, NJ: Prentice Hall, 1998:194–9.
16. Campbell DT, Stanley, JC. Experimental and Quasi-Experimental Designs for Research. Boston: Houghton Mifflin Co., 1966:13-24.
17. Ohlott PJ. Personal and work lives: overlapping spheres. Choices and tradeoffs facing high-achieving women. Cent Creat Leadersh. 2000 Winter:1–2,8.
18. Hewlett SA. Executive women and the myth of having it all. Harv Bus Rev. April 2002:66–73.
19. McDade SA, Dean DR. Relationship of faculty experiences of college presidents to their leadership development. Paper presented at the annual meeting of the Association for the Study of Higher Education, Sacramento, CA, November, 2002:5.
20. Catalyst. Creating Women's Networks: A How-to Guide for Women and Companies. San Francisco: Jossey-Bass, 1998.
21. Catalyst. Advancing Women in Business—The Catalyst Guide: Best Practices from the Corporate Leaders. San Francisco: Jossey-Bass, 1998.
22. Wellington S, Spence B, Catalyst Inc. Be Your Own Mentor: Strategies from Top Women on the Secrets of Success. New York: Random House, 2001.
23. Ely RJ, Meyerson DE. Theories of gender in organizations: a new approach to organizational analysis and change. Res Organ Behav. 2000;22:103–51.