CAREER DIVERSITY AS A GATEWAY

One barrier to recruiting students into doctoral programs is that many potential applicants do not know which career paths benefit from a PhD. Among those who do have an idea, many view becoming a professor as the primary career path.Footnote ⁴ This perception is not surprising given that students’ interactions with PhD holders usually are limited to their classroom experiences with professors. Doctoral programs often operate as if the pathway to professor is the most common or noble of positions to secure (Kyvik and Olsen Reference Kyvik and Olsen2012). Yet, approximately 50% of graduate students on track to receive a social science doctorate choose, accept, and succeed in positions outside of the academy that require the skills and knowledge obtained through PhD training.Footnote ⁵ Thus, increasing undergraduate awareness of varied nonacademic careers that deploy skills gained from doctoral programs may correct baseline expectations and attract more students into those programs.

Emphasizing career diversity is particularly relevant for two reasons. First, doctoral training is useful for a wide variety of careers in the modern knowledge economy, which is characterized by “production and services based on knowledge-intensive activities” that rely on intellectual capabilities and the capacity to adapt (Powell and Snellman Reference Powell and Snellman2004). This market rewards the type of human capital that PhD programs produce (Pedersen Reference Pedersen2014). Among other skills transferred, the tenets of comparative analysis, research design, strategic anticipation, and statistical analysis at the core of social science doctoral programs permit nonacademic professionals to study and understand attitudes, behavior, risks, institutional incentives, expected policy outcomes, and program effectiveness.

Second, the academic job market has contracted whereas the number of doctoral-degree seekers has expanded (Diaz and McGrath Reference Diaz and McGrath2023; McGrath and Diaz Reference McGrath and Diaz2023; Neumann and Tan Reference Neumann and Tan2011). Consequently, a substantial and increasing proportion of PhD graduates in the social sciences pursue nonacademic careers (Diaz and McGrath Reference Diaz and McGrath2023; McGrath and Diaz Reference McGrath and Diaz2023; Neumann and Tan Reference Neumann and Tan2011). This growing reality is producing changes in doctoral programs aimed at broadening the relevance of skill sets transferred to students (Cassuto and Weisbuch Reference Cassuto and Weisbuch2021; Diaz and McGrath Reference Diaz and McGrath2023; Grafton and Grossman Reference Grafton and Grossman2011; Kyvik and Olsen Reference Kyvik and Olsen2012; Lazer Reference Lazer2021). Nevertheless, most programs lag in these efforts (Cassuto and Weisbuch Reference Cassuto and Weisbuch2021; Hancock Reference Hancock2023; Seo and Yeo Reference Seo and Yeo2020). Taking career diversity seriously can both enhance labor-market success for PhD holders and increase the attractiveness of PhD programs (Yi et al. Reference Yi, Shin, Buchok and Han2023).

College students have a predilection for discussing sociopolitical change, which naturally draws them toward political science (Mueller Reference Mueller2023). If those same students consider the academy as the principal career for PhD holders, it is not surprising that many who enjoy studying and discussing sociopolitical issues select into trajectories toward law, policy making, nongovernmental work, and other nondoctoral career paths. Furthermore, even if they are aware that those holding a PhD in political science can pursue nonacademic careers, they may erroneously believe that such pathways are less likely to be satisfying. Although many graduates who pursue academic paths find satisfaction in those efforts, job satisfaction can be equally as high—or higher—among PhD holders who pursue nonacademic, research-centered career paths (Cassuto and Weisbuch Reference Cassuto and Weisbuch2021; Hancock Reference Hancock2023).

A BRIEF INTRODUCTION TO THE REU

A powerful tool to increase diversity in political science is undergraduate research programs; as such, the NSF funds REU summer programs. These initiatives tend to target students from small, regional, and teaching-centered colleges with fewer opportunities for research experiences, fewer faculty mentors, and often more diverse backgrounds. REUs open opportunities for first-generation college students and students of color, who are more likely to have familial and financial responsibilities outside of the classroom that limit extracurricular research opportunities (Carter et al. Reference Carter, Aponte, Murib, Baptist, Dickinson, Sadhwani and Le2022; Denny et al. Reference Denny, Amechi, Sanchez, Acosta, De Pina, Frausto and Meza2023; Druckman Reference Druckman2015). The NSF promotes these recruitment features to realize its objective of “increas[ing] the participation of the full spectrum of the nation’s diverse talent in STEM.”Footnote ⁶

In the summers of 2023 and 2024, we ran an eight-week residential REU program at Vanderbilt University.Footnote ⁷ The program paired students with faculty experts in conflict, political behavior, and democratic institutions. We narrowed a large applicant pool based on the demonstrated capacity for social scientific thinking, the added value of this program, and indications of whether their personal experiences would enhance diversity in social sciences. Across the combined summers of 2023 and 2024, we selected 20 students in a manner that prioritized regional and partnering institutions and created balance across our program’s intellectual areas. We accepted seven additional students via the national Leadership Alliance program.Footnote ⁸

Participants’ time was structured around three main activities designed to increase interest, skill development, and career preparation. First, participants were paired with a faculty mentor to guide their individual research projects with a ready-to-use dataset. Meetings were structured around a research plan, with scaffolded assignments to advance projects culminating in a paper and poster presentation. Second, students attended weekly seminars on research design, statistical inference, and basic statistical analysis. They had further opportunities to practice reading, critiquing, and responding to social science research in weekly reading groups, in which we facilitated a discussion between them and the faculty author of an article. Third, other activities, including career panels and a writing seminar, focused on professional development. Throughout the eight weeks, our program addressed theorized barriers to applying to and succeeding in doctoral program—including beliefs that these programs are too expensive, too difficult, or unobtainable due to feelings of self-doubt or imposter syndrome (Carter et al. Reference Carter, Aponte, Murib, Baptist, Dickinson, Sadhwani and Le2022; Dickinson, Jackson, and Williams Reference Dickinson, Jackson and Williams2020; Tormos-Aponte and Velez-Serrano Reference Tormos-Aponte and Velez-Serrano2020).

A FOCUS ON CAREER DIVERSITY

Many undergraduate students are uncertain about the potential to translate a doctoral degree in political science into a nonacademic career path.Footnote ⁹ This lack of information and these feelings of uncertainty are potential barriers to entry (Dickinson, Jackson, and Williams Reference Dickinson, Jackson and Williams2020), especially for students from groups that traditionally have been excluded from research careers in the social sciences. In designing our program, we anticipated that programming around career diversity would broaden student awareness of nonacademic career paths for PhD holders (Expectation 1). We further expected these students to become more inclined to pursue a PhD than they were before they participated in the program (Expectation 2).

We intentionally designed our program to address and counter inaccurate assumptions about career diversity. First, we held a series of career panels featuring individuals with doctorates in political science who work in research-oriented positions. Some of these panels included faculty members from teaching- and research-centered colleges, while others featured nonacademic researchers working in non-profit organizations, government divisions, and survey research firms. In addition to showcasing career diversity, we recruited panelists who varied along other dimensions. These panelists described the paths into their current positions and how their skills and experiences from graduate school contributed to their daily tasks. Students and faculty members then engaged with panelists during Q&A sessions.

Second, we held a weekly seminar on writing and the profession. Half of these sessions dissected different forms of writing about research results, including articles, briefings, op-eds, press releases, and narrative storytelling. Discussions and activities connected the different styles of research reporting to a variety of career options. The other half of the sessions workshopped students’ draft materials for PhD programs, set expectations for what programs are like, and discussed why one might want to enter a doctoral program, with explicit connections to diverse career goals. Third, we hosted two panels at the beginning and the end of the program with current doctoral students who discussed their different experiences and intended career paths (e.g., law professor, data scientist, government service, and non-profit).

DATA AND METHODS

We administered surveys through Qualtrics before and after the program,Footnote ¹⁰ noting three limitations. First, the number of observations was small; of the 27 program participants across 2023 and 2024, 25 completed the presurvey and 23 completed the postsurvey (Barnes et al. Reference Barnes, Ritter, Thrower, Tripp and Zechmeister2025). Second, REU applicants were already interested in research-centered careers; therefore, we expected small effects. Third, while we could assess preprogram versus postprogram responses, we lacked a true control group. Instead, we made comparisons to a sample of similar students drawn from participants in a political science research lab at our university.

All of our analyses were difference-of-means t-tests performed in R, using a threshold of p<0.05 (one-tailed) to assess statistical significance. Given the unidirectional nature of our expectations, we used a one-tailed approach. A small sample size has a comparatively greater likelihood of producing Type II errors (i.e., false negatives) (Knudson and Lindsey Reference Knudson and Lindsey2014); we addressed this challenge by noting when a result was close to the p<0.05 threshold.

In our sample of REU students, 75% identified as women, 20% participated in the federal work-study program, 44% received Pell Grants, 48% were first-generation college students, 52% attended minority-serving institutions, and 63% attended a public university. Students were able to select multiple ethnicities: 44% identified as white, 40% as Black, 32% as Latine, and 12% as Asian. The modal student was a rising senior (46%). Prior to REU participation, 40% reported having at least one mentor at their home institution who helped them to “independently perform research” (the average number of mentors was 1.7).

ANALYSIS: AWARENESS OF CAREER DIVERSITY AND CAREER INTENT

To evaluate Expectation 1 (career-diversity awareness), we asked students about career paths available to PhD holders with this statement: “I do not know much about what career paths are available to those who obtain a PhD.” Lower values on the 0–10 response scale indicated less uncertainty (i.e., higher levels of awareness). The pooled data (both years combined) showed a decrease in uncertainty around which career paths are available for PhD graduates: we find a preprogram mean of 5.76 versus a postprogram mean of 1.91 (significant at p<0.05, one-tailed).Footnote ¹¹

For comparison, in Fall 2024, we sent surveys to undergraduate students at our home institution who were current or recent participants in a political science lab that provides skill-building and mentored research experiences.Footnote ¹² The members of this lab do not receive additional programming related to career diversity. The mean of agreement that they lacked knowledge of career options for this group was 6.0, which was similar to the pretest mean for the summer REU participants.Footnote ¹³ Thus, compared to the preprogram mean and to this quasi-control group, the postprogram REU students reported significantly less uncertainty about career paths for PhD holders.

Although our treatment was a compound of all program elements, those that emphasized career diversity were salient and well received by the students. In the postsurvey, students rated this component, averaging 8.38 on a 0–10 scale (where 10 was “very positive”). Additionally, in a midsummer survey administered to the 2023 cohort, responses reflected interest in the panels, with multiple students requesting more time to engage with the panelists. We caution, however, that these responses may have been driven by factors unrelated to career diversity—for example, we may have selected panelists who had positive or fun interactions with the students.

To further investigate Expectation 1, we asked the 2024 postprogram cohort to offer their perspectives on whether the “emphasis on career diversity ha[d] a positive, negative, or no effect” on their “plans for postgraduate education and/or career path?” All six of those who answered this open-ended question stated that the effect was positive, including comments that awareness of multiple possible career paths “was amazing and very eye opening”; “helped me realize more of the options available”; was “important to know”; and “made me more excited about possible career paths and the doors that a PhD opens up.” Whereas some students felt the emphasis helped them to understand the “nuances of different career paths,” others noted that there was room for even greater clarity on the daily activities of those in alt-academic careers. For example, one student stated, “I know what a professor does. But when someone says ‘I work for a think-tank organization,’ what does that mean?”Footnote ¹⁴

To assess Expectation 2, we measured students’ inclinations toward pursuing a social science PhD. This interest began at a high level (presurvey mean = 8.48 on a 0–10 scale) and only shifted marginally by the end of the program (postsurvey mean = 8.78; differences were not statistically significant).Footnote ¹⁵ However, we did observe suggestive evidence of an increase in students’ likelihood of choosing to pursue a PhD as their first career-path option: within the pooled dataset, the mean value increased from 0.52 in the presurvey to 0.74 in the postsurvey. Yet, at p = 0.06 (one-tailed), this result did not meet our significance threshold.Footnote ¹⁶ Thus, we cannot conclude that an emphasis on career diversity has a standalone and significant impact in elevating already-high inclinations toward pursuing a PhD. Rather, we suggest a more modest conclusion: an emphasis on career diversity broadens participants’ perspectives, is well received, and may help buoy—maintain, if not boost—the likelihood of REU participants applying to PhD programs.