A recent study by Bakken et al.1 reported a gender difference in clinical research self-efficacy among postgraduate physician trainees. Women tend to report lower ratings on their perceived abilities to perform tasks and activities necessary to conduct clinical research. Self-efficacy has been shown to be one of the major factors in career choice, persistence,2–7 and advancement,8 and, therefore, is an important variable to consider in the career development of physician–scientists, especially women, whose proportions drop remarkably as they advance through the academic ranks.9–11 Self-efficacy is acquired and modified through personal performance accomplishments, social persuasion, physiological states and reactions, and vicarious learning experiences.12 Role models can enhance self-efficacy by acknowledging and rewarding performance accomplishments, encouraging challenging activities, being sensitive to emotions and feelings, and demonstrating positive attitudes and ways to overcome obstacles and barriers in career development. Role models, therefore, can be an important factor in influencing motivation and sustainability in academic careers.8,13,14
The University of Wisconsin's Clinical Investigator Preparatory Program (CIPP) staff members originally collected these data for educational purposes only; however, after analyzing the data, I felt it important to share the findings with my colleagues. In this report, therefore, I focus on characteristics of role models as a potential source of the gender difference in clinical research self-efficacy reported previously. The specific purpose of this study was to determine whether gender differences also exist in the role models—or experts—physician–scientists envision when they assess their own abilities to perform clinical research tasks and activities. I modified the assessment questionnaire described in Bakken et al.1 to solicit this image from physician–scientist trainees over a two-year period to determine whether these images provide any insight into gender differences in clinical research self-efficacy. I hypothesized that men's and women's images of the expert would differ and reflect qualities of expert role models valued by each gender.
The CIPP staff administered the previously developed self-efficacy questionnaire to 251 health care professionals who attended a three-day Short Course in Clinical Research in 2002 or 2003 or who entered the two-year CIPP (K30 HL04100) at the University of Wisconsin-Madison between February 2002 and February 2004.
The self-efficacy questionnaire consisted of 35 items related to abilities to perform clinical research tasks and activities in six general areas: biostatistics and study design, bioethics, scientific writing, leadership, teaching and presentation, and other research-related activities.1 The questionnaire also contained three questions that I added in November 2002 to determine the characteristics of the trainees’ envisioned expert: What is the expert's gender?, What is the expert's academic role?, and What qualities does this expert have? The envisioned expert's academic role was obtained by asking the participants to select one of seven categories: faculty mentor, peer, teacher/instructor, faculty in department (not a mentor), faculty of another department (not mentor), nonfaculty scientist or researcher, and other. Qualities of envisioned experts were obtained by asking participants to check each of 11 qualities that applied to their envisioned expert. These qualities were selected based on qualities typically valued by tenure-granting committees and funding agencies and those reported in the literature.15,16 An “other” category was also listed in this section of the questionnaire so participants could indicate qualities that were not present in the list. I instructed participants to use a person that they considered to be an expert researcher as the standard for the expert's abilities when they completed the questionnaire. I computed frequency distributions for each response and compared variables by gender using chi-square analysis and Fisher exact test. Statistically significant differences between men and women were determined with p < .05.
I excluded 17 of the 124 respondents from this evaluation because they did not indicate their gender or academic title and 12 respondents because they did not answer the three new questions. Of the 95 remaining respondents, four were medical students, 53 were postgraduate trainees, eight were clinical instructors, and 30 were assistant professors. Fifty-four percent were women and 46% were men. The distribution of men and women by title was similar (see Table 1).
Seventy-one percent of the female respondents reported their envisioned experts to be male, compared to 95% of male respondents. The most frequently reported role of the envisioned expert was that of a mentor who was a faculty member in the respondent's own department (72% for women, 60% for men) or a nonmentor faculty member in the respondent's own department (16% women, 21% men). The envisioned expert was much less frequently reported to serve other roles (see Table 2).
A breakdown of the frequency of the envisioned expert's qualities reported by male and female respondents is shown in Table 3. Men and women differed little in their three most frequently reported qualities of the envisioned expert. Ninety percent of the women and 86% of men most frequently identified the quality “multiple publications,” followed by “scientific knowledge” (86% for men and women, respectively). Women were somewhat more likely to identify “supportiveness” as an important trait (86% women, 77% men). Men and women differed significantly in identifying two qualities: 78% of women selected “communication skills” compared to 59% of men (χ2 = 4.166, df = 1, p = .042), and 78% of women indicated “excellent problem solver” compared to 59% of men (χ2 = 4.166, df = 1, p = .042).
To determine whether this difference in qualities of envisioned experts was accounted for by the fact that female junior investigators more frequently reported female experts, I compared qualities selected by women who reported female experts (no. = 14) and qualities selected by women who reported male (no. = 35) experts. As shown in Table 4, both women and men who reported their envisioned expert to be male reported their qualities in similar proportions and none of these differences were significantly different at p < .05. Women who reported female envisioned experts, however, more frequently reported “communication skills” and “excellent problem-solving abilities” than did men who reported male-envisioned experts. These differences were statistically significant by Fisher exact test at p < .10.
In summary, when rating their perceived abilities on knowledge and skills related to clinical research, women's and men's envisioned expert was most likely to be a male faculty member in the respondent's own department who served as a mentor. Respondents envisioned this mentor to exhibit expertise in clinical research primarily through multiple publications, scientific knowledge, and support. However, women who envisioned female experts most frequently reported the qualities “communication skills” and “excellent problem solving abilities.” Because role modeling can positively influence self-efficacy, these results suggest that an envisioned expert's gender and role model qualities may have been important factors that influenced a previous observation of gender differences in clinical research self-efficacy1 and propose additional variables that need to be considered when studying the career development of physician–scientists.
The purpose of this study was to determine whether gender differences exist in the role models physician–scientists-in-training or in their early years of career development envision when they self-assess their abilities to perform clinical research tasks and activities. Role models are important for providing vicarious learning experiences that shape individuals’ self-concepts and effectively bolster self-efficacy. In this study, when a respondent's envisioned expert was male, the reported qualities of that expert did not significantly differ between male and female respondents. This finding contradicted my hypothesis that these qualities would differ based on men's and women's different values. Upon further analysis of the data, however, the findings revealed that the hypothesis is supported when women have female role models. These are important discoveries in terms of understanding the gender differences in clinical research self-efficacy observed by Bakken et al.1 and addressing the unique differences of women as they develop their academic medical careers.
In male-dominated cultures, such as academic medicine, women are likely to emulate men's values, rules, and norms in order to succeed in their careers.17–19 The findings of this study revealed that women who envisioned male experts identified qualities in that expert in similar proportions as men with male-envisioned experts; however, women with female-envisioned experts did not identify qualities in that expert in similar proportions to men who envisioned male experts. These findings suggest that male values (e.g., multiple publications) are modeled for women as standards for success or expertise—standards that may seem difficult to achieve or are less important for women, who typically struggle with their roles as caregiver and partner and place greater value on relationships.15,17,20 Brown et al.21, p. 540 found that the “relation of the research training environment to research self-efficacy beliefs is larger among women than among men.” Role conflict is a major environmental factor that has been linked to job persistence, performance, and success.22 Positive (or same-gender) role models may operate by reducing role conflict and increasing discipline self-efficacy.8
In this study, women with female role models reported problem-solving and communication skills as the most important qualities of their envisioned expert. Studies in the social sciences repeatedly demonstrate that women place more value on relationships than men do.15, 17, 20 Women show preference for collaboration over competition, encouragement over challenge, integration over separation, and group affiliation over individual achievement23—values that are opposed to those of traditional academic environments that typically emphasize and reward individual accomplishments. Women's more frequent emphasis on problem-solving and communication skills may mean that women look to other women to communicate more effectively and help them balance the difficulties of juggling multiple roles between work and family.24–27
“Support” was an expert's quality frequently reported by both men and women. I expected that women would select this quality more frequently than men would and suspect that “support” has different meanings for women and men. Women with their value of relationships are likely to view support in social terms, such as through encouragement and collaboration.15 Men, however, are more likely to view support as enhancing their independence and competitive edge, such as laboratory resources, statistical support, or secretarial assistance.28 Further studies are needed to define this quality for men and women.
What do these findings mean in terms of the relationships between role models and self-efficacy? If role models do not recognize and address the unique needs or attributes of women who are developing a clinical research career, but instead impose a traditional value system that creates role conflicts and serves to erode self perceptions of abilities to perform tasks and activities needed to develop a research career (i.e., research self-efficacy), then it only makes sense that women may not choose or persist in this career pathway.
With the limited presence of female physician–scientists in academic medicine, how can we foster role modeling and environments that are favorable to women's career development and success? Role conflict can be reduced through role models (both male and female) who successfully combine family and work, informal interactions with trainees, and increased opportunities for job sharing, part-time work, and on-site child care.8 Additionally, alternative mentoring models that consist of multiple, peer, or collective mentoring14 may prove effective at serving the professional and personal needs of female physician–scientists. These types of role models and mentoring strategies are likely to be helpful not only to women, but also to young men who are taking more responsibilities for family and household duties.22
Creating more effective role models for women will require increasing the number of women in academic medicine. Not only are women less represented among academic faculty, but female clinician–scientists are also less likely than are male clinician–scientists to be mentored or have access to senior faculty.29 Increasing the number of women in work environments, particularly academic medicine, is not only likely to increase women's access to role models and mentors, but doing so may transform the culture into one more favorable to women's career development and success.9,13 Indeed, the results of this study, although limited by sample size, would suggest that increasing the availability of female role models and mentors would support the qualities that women value in expert clinical researchers—problem-solving and communication skills that are important in fostering positive role models and the types of environments that promote strong self-efficacy.
Although the sample size of this study was small and limited to one predominately white campus, it highlights the importance of environmental variables, such as role models, that can influence self-efficacy beliefs and career development of women. Plans are in place to extend these studies to larger and more diverse samples; however, it will be important to incorporate the knowledge gleaned from this study into future research. I encourage others to do the same as attempts are made to understand the complex nature of career development for women in academic medicine and create a culture that welcomes and encourages the ideas and success of a diverse faculty of both men and women representing a variety of racial and ethnic backgrounds.
The author wishes to thank Dr. Molly Carnes and Dr. Curt Olson for their excellent comments and editorial suggestions in preparing this manuscript. This work would not have been possible without the help and support of the Clinical Investigator Preparatory Program staff, Michelle Sloniker, Terry Gorman, and Joanna Liebl, who routinely collect evaluation data and coordinate educational activities. This research is supported in part by grants 5 K30 HL04100-04 (National Heart, Lung and Blood Institute) and 5 M01 RR03186-18 (National Center for Research Resources) from the National Institutes of Health.
1Bakken L, Sheridan J, Carnes M. Gender differences among physician-scientists in self-assessed abilities to perform clinical research. Acad Med. 2003;78:1281–6.
2Lent R, Brown S, Hackett G. Toward a unifying social cognitive theory of career and academic interest, choice, and performance. J Vocat Behav. 1994;45:79–122.
3Lent R, Brown S, Hackett G. Contextual Support and barriers to career choice: a social cognitive analysis. J Couns Psychol. 2000;47:36–49.
4Zimmerman B. Self-efficacy and educational development. In: Banduras A (ed). Self-Efficacy in Changing Societies. Cambridge, UK: Cambridge University Press, 1995.
5Nauta M, Kahn J, Angell J, Cantarelli E. Identifying the antecedent in the relation between career interests and self-efficacy: is it one, the other, or both? J Couns Psychol. 2002;49:290–301.
6Flores L, O′Brien K. The career development of Mexican American adolescent women: a test of social cognitive career theory. J Couns Psychol. 2002;49:14–27.
7Luzzo D, McWhirter E. Sex and ethnic differences in the perception of educational and career-related barriers and levels of coping efficacy. J Couns Psychol. 2001;79:61–7.
8Nauta M, Epperson D, Kahn J. A multiple-groups analysis of predictors of higher level career aspirations among women in mathematics, science, and engineering majors. J Couns Psychol. 1998;45:483–96.
9Bickel J, Wara D, Atkinson B, et al. Increasing Women's leadership in academic medicine: report of the AAMC Project Implementation Committee. Acad Med. 2002;77:1043–61.
10Buckley L, Sanders K, Shih M, Kallar S, Hampton C. Obstacles to promotion? Values of women faculty about career success and recognition. Acad Med. 2000;75:283–8.
11Wright A, Schwindt L, Bassford T, et al. Gender differences in academic advancement: patterns, causes, and potential solutions in one U.S. college of medicine. Acad Med. 2003;78:500–8.
12Bandura A. Social Foundations of Thought and Action. Englewood Cliffs, NJ: Prentice-Hall, 1986.
13Stalker J. Athene in academe: women mentoring women in the academy. Int J Lifelong Educ. 1994;15:361–72.
14Chesler N, Chesler M. Gender-informed mentoring strategies for women engineering scholars: on establishing a caring community. J Eng Educ. 2002;91:49–55.
15Surrey J. The “self-in-relation”: a theory of women′s development. In: Jordan J, Kaplan A, Miller J, Stiver I, Surrey J (eds). Women's Growth in Connection: Writings from the Stone Center. 1st ed. New York: The Guilford Press, 1991:51–66.
16Mavis B, Katz M. Evaluation of a program supporting scholarly productivity for new investigators. Acad Med. 2003;78:757–65.
17Bierema L. Women, work and learning. New Dir Adult Contin Educ. 2001;92(Winter):53–62.
18Pratto F, Espinoza P. Gender, ethnicity, and power. J Soc Issues. 2001;57:763–80.
19Valian V. Why So Slow? 4 ed. Cambridge, MA: The MIT Press, 2000.
20Hayes E, Flannery D, Brooks A. Women as Learners. 1st ed. San Francisco: Jossey-Bass Publishers, 2000.
21Brown S, Lent R, Ryan N, McPartland E. Self-efficacy as an intervening mechanism between research training environments and scholarly productivity: a theoretical and methodological extension. Couns Psychol. 1996;24:535–44.
22Warde C, Allen W, Gelberg L. Physician role conflict and resulting career changes: gender and generational differences. J Gen Intern Med. 1996;11:729–35.
23Gilligan C. In a Different Voice: Psychological Theory and Women′s Development. Cambridge, MA: Harvard University Press, 1982.
24Bierema L. A synthesis of women's career development issues. New Dir Adult Contin Educ. 1998;80(Winter):95–103.
25Carr P, Ash A, Friedman R, et al. Relation of family responsibilities and gender to the productivity and career satisfaction of medical faculty. Ann Intern Med. 1998;129:532–8.
26Yedidia M, Bickel J. Why aren′t there more women leaders in academic medicine? The views of clinical department chairs. Acad Med. 2001;76:453–65.
27Carnes M. Balancing family and career: advice from the trenches. Ann Intern Med. 1996;125:618–20.
28Johnson J, Williams B, Jayadevappa R. Mentoring program for minority faculty at the University of Pennsylvania School of Medicine. Acad Med. 1999;74:376–8.
29Chew L, Watanabe J, Buchwald D, Lessler D. Junior faculty's perspectives on mentoring. Acad Med. 2003;78:652.