Biomedical Science Ph.D. Career Interest Patterns by Race/Ethnicity and Gender

Abstract

Increasing biomedical workforce diversity remains a persistent challenge. Recent reports have shown that biomedical sciences (BMS) graduate students become less interested in faculty careers as training progresses; however, it is unclear whether or how the career preferences of women and underrepresented minority (URM) scientists change in manners distinct from their better-represented peers. We report results from a survey of 1500 recent American BMS Ph.D. graduates (including 276 URMs) that examined career preferences over the course of their graduate training experiences. On average, scientists from all social backgrounds showed significantly decreased interest in faculty careers at research universities, and significantly increased interest in non-research careers at Ph.D. completion relative to entry. However, group differences emerged in overall levels of interest (at Ph.D. entry and completion), and the magnitude of change in interest in these careers. Multiple logistic regression showed that when controlling for career pathway interest at Ph.D. entry, first-author publication rate, faculty support, research self-efficacy, and graduate training experiences, differences in career pathway interest between social identity groups persisted. All groups were less likely than men from well-represented (WR) racial/ethnic backgrounds to report high interest in faculty careers at research-intensive universities (URM men: OR 0.60, 95% CI: 0.36-0.98, p = 0.04; WR women: OR: 0.64, 95% CI: 0.47-0.89, p = 0.008; URM women: OR: 0.46, 95% CI: 0.30-0.71, p<0.001), and URM women were more likely than all other groups to report high interest in non-research careers (OR: 1.93, 95% CI: 1.28-2.90, p = 0.002). The persistence of disparities in the career interests of Ph.D. recipients suggests that a supply-side (or "pipeline") framing of biomedical workforce diversity challenges may limit the effectiveness of efforts to attract and retain the best and most diverse workforce. We propose incorporation of an ecological perspective of career development when considering strategies to enhance the biomedical workforce and professoriate through diversity.

Figure 1. Distinct career interest profiles among Ph.D. biomedical scientists by social identity.

(A) Bar graph showing mean response for sample of 1500 American biomedical scientists who received Ph.Ds. between 2007–2012 when asked to rate their level of interest in each of the following career paths at Ph.D. entry (black), Ph.D. completion (grey), on a 5-point scale (where 1 represents “no interest” and 5 represents “strong interest”): faculty at a research-intensive university; faculty at a teaching intensive university; a research career outside of academia (e.g. industry, pharmaceutical, biotech, government, start-up, etc.); and a non-research career (consulting, policy, science writing, patent law, business, etc.). (B) Pie chart showing the social identities of the respondents. Males from well-represented racial/ethnic backgrounds (WRM) are shown in blue and represent 25% of the sample; males from underrepresented minority backgrounds (URMM) are shown in red and represent 5.8% of the sample; females from well well-represented racial backgrounds (WRF) are shown in green and represent 53.9% of the sample; females from URM backgrounds (URMF) are shown in purple and represent 12.6% of the sample; and respondents declining to state racial/ethnic background or with an alternative gender identification are shown in grey and represent 2.7% of the sample. (C) Bar chart showing mean interest in the four career paths at Ph.D. entry, Ph.D. completion across social identity. Group means were compared at each time point and statistical significance was determined using Bonferroni corrected ANOVA. (D) Plot showing the average, individual level paired-difference between career pathway interest at Ph.D. completion versus Ph.D. entry across social identity groups. Statistical significance was determined using Bonferroni corrected ANOVA.

Figure 2. Disparate Career Interest Profiles Across Social Identity.

(A) Plot of adjusted odds-ratio (circle) and 95% confidence interval, with males from well-represented racial/ethnic backgrounds (WRM) as the reference group, showing likelihood of expressing high interest (i.e. 4 or 5 on the 5-point interest scale) in four career paths at Ph.D. completion: (i) faculty at a research-intensive university, (ii) faculty at teaching-intensive university, (iii) a non-academic research career, and (iv) a non-research career. Odds ratios are adjusted for personal dispositions (level of interest in career path at Ph.D. entry, intention to pursue faculty career at Ph.D. entry, and confidence in ability as an independent researcher), objective measures (rate of first-author publications, h-index, time-to-Ph.D., Ph.D. institution type), and graduate training experiences (socialization measures, advisor interactions, and career development experiences).

Figure 3. Ecological Conceptualization of Factors Impacting Career Development of Biomedical Scientists.

As disparities in career interest at Ph.D. completion by social identity persist when controlling for productivity, self-efficacy, advisor interactions, and other determinants of career choice, we propose that workforce development and diversity efforts utilize an ecological framework that takes into account the multiple agents that can influence the career decision-making process. These include individual level variation, research group and advisor, department and institutional training environment, pressures exerted by funding agencies, and the broader dynamics in the research enterprise.