Medical school faculty surveys have explored work-family balance (1, 2) and faculty satisfaction (3, 4). Few surveys have focused on gender (1, 3) or generational differences (5). Fox et al. (1) found a gender split in factors affecting academic advancement among part-time faculty at the University of Illinois College of Medicine. Women selected part-time status for child care, while men chose part-time to moonlight. Full-time men were more likely to be on the tenure track than any other group (1). In a study performed at the University of Wisconsin Medical School, similar percentages of women and men had seriously considered leaving academic medicine; however, women cited conflicting family responsibilities (51%), while men mostly cited uncompetitive salaries (59%) (6). In a survey of 337 pediatric faculty and house staff, 83% reported feeling stressed from efforts to balance work and family. Independent predictors of stress included increasing age, dependent children, less support from colleagues and supervisors, and female gender (2).

Generational differences have been implicated as contributors to difficult academic climates. However, recent data show few significant differences in reports of work-life balance, work hours, and attitudes toward patient care across generations and genders (5). Additionally, studies have shown that differences in traditional gender roles may be less pronounced in younger generations. For example, younger women and men physicians who were married with children did not differ in their rates of career change for marriage (7), while family and life style priorities were important to medical students of both genders in determining their career choice (8).

Our study explored professional faculty development while also aiming to elucidate any existing gender or rank differences. It was conducted in the Department of Psychiatry and Behavioral Sciences at the University of California (UC) Davis, by an established peer mentoring group, the Society for Women in Academic Psychiatry (SWAP) (9).

This study was approved by the institutional review board at UC Davis and consisted of a 10-question survey which was anonymous and administered online (www.surveymonkey.com). Participation was entirely voluntary. All department faculty members received a link to the survey followed by two e-mail reminders. Responses were collected during October and November 2007.

The survey instrument was adapted from the questionnaire validated by the Office of Faculty Development at the National Center for Leadership in Academic Medicine, with the authors’ permission (10). Several items were added (e.g., developing an educational portfolio, having a flexible work environment, and sharing any other area of interest not included). The first question asked participants to rate the importance of learning a number of skills, as based on their interests and needs, on a five-point Likert scale (1=cannot decide, 2=not important, 3=somewhat important, 4=important, 5=very important) and specified 29 items to be rated (see Appendix 1). Item 30 referred to other areas of interest not included.

The second question asked respondents to choose the three most important items listed in question 1 and rank them from most to least important, as modeled after another survey (11). Questions 3–10 addressed demographic variables, including academic rank, degree(s), academic series, gender, number of children, employment status (full- or part-time), as well as total years in academia and as UC Davis faculty. Due to the small department size, questions regarding faculty members’ age and ethnicity were omitted so that individual respondents would not be identified.

Statistical analyses were conducted using the SAS Institute Version 9.1 (SAS/STAT® 9.1, Cary, NC, 2002–2003). For each of the items in question 1, ratings were converted to scores from 1 (not important/cannot decide) to 4 (very important), with higher scores indicating higher levels of perceived importance. Item 30 did not receive scores. Internal consistency of scale responses was excellent, with Cronbach alphas for all 29 items greater than 0.90. Descriptive statistics were used to characterize the study sample, both overall and broken down by faculty gender and rank (Table 1). Academic ranks were grouped as junior (e.g., instructor and assistant professor) and senior (e.g., associate and full professor).

Exploratory principal component analysis was applied to responses to question 1. After extracting the components and applying an orthogonal rotation, six domains were retained, with Eigen values greater than 1 and including three or more items with loading above 0.50. Combined, the six domains accounted for 74% of the total variance. Domain scores were created for each faculty member by averaging that member’s scores on items clustered within each of the six domains. Means and standard deviations were calculated for each domain, both overall and stratified by rank and gender (Table 2).

Of the 75 faculty (25 women, 50 men) in the department, 41 completed the survey, with an overall response rate of 55%. The sample composition did not differ from the department as a whole in ranks or degrees. Women had a higher response rate (68%, n=17) than men (48%, n=24), but the difference was not significant (p=0.14). In academic ranks, seven respondents (17%) were professors, 10 were associate professors (24%), 20 were assistant professors (49%), and four were instructors (10%). Fifteen members (37%) were Ph.D.s/Psy.D.s; 20 were M.D.s/D.O.s (49%); five were M.D./Ph.D.s (12%); and one was a licensed clinical social worker. Twenty-nine faculty (73%) had children, with a significant gender difference: 20 men (87%) but only nine women (53%) were parents (p=0.03).

Table 1 summarizes mean scores of the top 10 responses to question 1 for the whole sample, in decreasing order of the overall mean score. The mean scores ranged from 3.51 (SD=0.87) to 2.13 (SD=0.76). Table 1 also shows the cumulative percentage of participants rating these items as important or very important. The following items were strongly endorsed by more than 75% of respondents, with an overall mean score higher than 3: finding meaning in one’s work; maintaining integrity and ethics; maintaining one’s own values; balancing personal and professional demands; maintaining academic vitality; flexible work environment; and preventing, recognizing, and handling burnout. Only one element was listed under item 30: “support from the department.”

Principal component analysis revealed six item clusters (domains), described as follows: personal identity (items 1–7; see Appendix 1), career management (items 10, 16–18), academic skills (items 19–21, 23), interpersonal management (items 13–15, 22), research (items 9, 26–28), and diversity and flexibility (items 11, 12, 29). Table 2 presents the mean domain scores for the whole sample and stratified by gender and rank. Personal identity was the only domain with mean scores above 3, showing that it was universally viewed as very important. Mean scores for the remaining five domains ranged between 2 and 3. Although there was substantial individual variability, there was no evident pattern of differences across either gender or rank.

Overall response frequencies to question 2 mirrored responses to question 1. Item 1, “Finding meaning in my work,” was ranked as the most important choice by 11 faculty members (27%) and in the top three by 15 (36%) respondents. Nineteen members (46%) also placed item 7, “Balancing personal and professional demands,” in their top three choices.

This is a study of 41 faculty members at UC Davis that explored the importance of key aspects of professional development. Faculty of both genders endorsed maintaining personal identity and work-life balance as very important. Not surprisingly, personal identity is a central factor, because being able to maintain one’s core values contributes to a sense of mastery and harmony. Misalignment of faculty and institutional values may contribute to faculty discontent (4). Similarly, Pololi et al. (10) found that all faculty members prioritized sustaining their vitality, balancing personal and professional lives, and retaining their own values.

Almost half of respondents placed balancing personal and professional demands in their top three priorities. This shows the high impact that work-life balance has on faculty satisfaction. Difficulties balancing work and family were the most likely predictor of faculty’s intent to leave academia, according to a recent study (4). Gender was not a significant predictor of intent to leave, nor was an association found with minority status, faculty role, or highest degree (4).

Overall, mean scores of individual items and domains were similar across gender and rank, showing that the faculty surveyed attach importance to the same needs, regardless of these demographic variables. This is in contrast with prior studies that revealed several gender differences in values, attitudes, and priorities of faculty. Our study is limited by the small sample size and the difference in men’s and women’s response rate, making these findings difficult to generalize. While we recognize that the sample sizes were small and we had low power for detecting small effect sizes, the differences observed between genders and ranks were modest. Therefore, we cannot rule out some gender/rank differences, but it is unlikely we would have missed differences as large as to have departmental policy implications (i.e., larger than 1 point).

Because SWAP conducted this study, selection bias may have affected the sample. Respondents who were generally more favorable to gender equity and diversity may have been more likely to complete the survey. Additionally, the number of women M.D.s in the UC Davis Department of Psychiatry is very low, and the women are mostly at junior ranks, whereas the number of women Ph.D.s exceeds the national average (9). Also, women respondents, as compared with men, were significantly less likely to be parents. About half of the women and 87% of men had children, and children’s age and care-giving responsibilities were not explored. Carr et al. (3) found that, among faculty with children, women had greater obstacles to academic careers, fewer publications, less institutional support, slower self-perceived career progress, and lower career satisfaction than men. No significant gender differences were found for faculty without children. In a landmark survey of University of California faculty, Mason and Goulden (12) showed that “married with children” is the success formula for men, while the opposite is true for women, for whom there is a serious “baby gap.”

This study did not evaluate the academic climate or level of needs fulfillment. Instead, it focused on perceived needs and interests of faculty in different career developmental stages. Association of survey responses with other demographic variables was not pursued because it could lead to identification of respondents. A larger sample size would allow review of findings in light of degree, academic series, years in academia and at UC Davis, employment status, and number and age of children.

UC Davis men and women academic psychiatrists in different career development stages have similar needs, although the paths leading to personal and professional fulfillment may vary according to gender. Maintaining personal identity and work-family balance are very important to faculty of both genders. Further research exploring gender disparities in faculty development will help maintain grassroots efforts and guide the creation or enrichment of institutional programs tailored to individual faculty needs. We hope that these initiatives will translate into recruitment and retention of diverse faculty.

The authors acknowledge support from the UC Davis Department of Psychiatry and Behavioral Sciences. We are also grateful to Laurel Beckett, Ph.D., for her valuable input in statistical analysis.

At the time of submission, the authors reported no competing interests.