Advancing Minorities and Women to the PhD in Physics and Astronomy

Keivan G. Stassun
Vanderbilt University and Fisk University, Nashville, Tennessee

Doctoral pathways graph

Figure 1: Comparisons between underrepresented minorities (URMs) and White/Asian students, based on different permutations of the educational pathway to the PhD. An equal sign indicates degrees earned from the same institution. The fourth and sixth comparisons from the left show the “traditional” paths to the PhD, in which the student earns the bachelors degree from institution A, and either receives both the masters degree and the PhD from institution B or else forgoes the masters degree entirely. The fifth comparison from the left is shown the case for earning the bachelors degree at institution A, a “terminal” masters degree at institution B, and PhD from institution C. Minorities are much more likely to take this latter path than non-minorities. Based on analysis of 80,739 PhDs earned in science and engineering fields, 1998 to 2002. Adapted from Reference 3. Copyright 2009, K. Stassun.

Introduction

The under-representation of minorities in the space sciences is an order-of-magnitude problem, and is one of the major challenges facing the nation’s science, technology, engineering, and mathematics (STEM) workforce as a whole.1 Minority-serving institutions are important producers of domestic minority talent in the sciences. Roughly one-third of all STEM baccalaureate degrees earned by African-Americans are earned at Historically Black Colleges and Universities (HBCUs), and the top 15 producers of Black baccalaureates in physics are all HBCUs. Just 20 HBCUs were responsible for producing fully 55% of all Black physics baccalaureates in the U.S. for 1998 to 2007.2 Institutional partnerships with HBCUs are thus a promising avenue for broadening participation in the physical sciences.3 At the same time, recent research on the educational pathways of minority students in STEM disciplines indicates that these students are roughly twice as likely as their non-minority counterparts to seek a master’s degree en route to the doctorate.4 These facts motivate programmatic approaches aimed at deliberately preparing underrepresented minority students for success as they traverse the critical Masters-to-PhD transition.

Here we describe a program developed in partnership between Vanderbilt University, a PhD-granting R-1 university, and Fisk University, a research active HBCU, both in Nashville, Tennessee. The Fisk-Vanderbilt Masters-to-PhD Bridge Program is for students who seek additional coursework or research experience before beginning PhD-level work. Students are not evaluated on the basis of GRE but rather on alternative metrics that are predictive of long-term success. The program provides a continuous path — a bridge — to the PhD that we have found is particularly effective for students whose baccalaureate degrees are from small, minority-serving institutions, and who may for a variety of reasons seek a master’s degree en route to the PhD. The program is flexible and tailored to the goals of each student. Courses are selected to address any gaps in undergraduate preparation, and research experiences are designed to pave the way for PhD-level work in the chosen area of study. While at Fisk, students enjoy regular interaction with Vanderbilt faculty including access to Vanderbilt courses and, of critical importance, thesis research performed under the joint supervision of Vanderbilt and Fisk faculty. In all cases, we deliberately develop research-based mentoring relationships between students and faculty that will foster a successful transition to the PhD.

GRE QuANT chart

Figure 2: GRE Quantitative score distributions from 2006-2007 for US citizens whose intended graduate majors were in STEM (this is the most recent publically available data). The tick is the median, and the top and bottom of each marker represents the 75th and 25th percentiles within each group; labels indicate the total number of test takers. The left axis is labeled with the old GRE scale and percentile; the right axis shows the corresponding scaled scores for the new exam. The blue horizontal line represents a typical “minimum acceptable” GRE score for admission to physics PhD programs. Adapted from Reference 7. Copyright 2013 APS. 
 

The Importance of Masters-to-PhD Transitions for Underrepresented Minorities

In the decade between 1990 and 2000, the total number of master’s degree recipients increased by 42%. During this same time period, the number of women earning master’s degrees increased by 56%, African Americans increased by 132%, American Indians by 101%, and Hispanics by 146%. A recent studyvprovides critical new insight into the role of the master’s degree as underrepresented minority students proceed to the doctorate in STEM disciplines. Data from the NSED was used to examine institutional pathways to the doctorate, and transitions from masters’ to doctoral programs by race and gender, for a sample of more than 80,000 PhDs.

As shown in Figure 1, the study identified six primary pathways to the PhD. Statistical analysis reveals that pathways are significantly different for underrepresented minorities (χ2=49.1, df=18, p 0.001). The two major differences are that White/Asian students are more likely to forgo earning the master’s degree altogether (“No MS, BA≠PhD” in Figure 1), and underrepresented minority students are much more likely to earn all three degrees at three different institutions (BS≠MS≠PhD). Underrepresented minority students are thus more likely to use the master’s degree as a stepping-stone toward success at the PhD level. Unfortunately, very often the transition from master’s degree to PhD is one that students must navigate on their own.

The Fisk-Vanderbilt Masters-to-PhD Program

Admission begins with application to the Fisk MA program in physics, which includes undergraduate transcripts, letters of recommendation, and a personal statement. The applicant indicates on the application that they wish to be considered for the Bridge program and submits an additional Bridge program information form.

Officially speaking, admission to the Bridge program does not constitute pre-admission to the Vanderbilt PhD program, nor does it carry with it a formal guarantee of admission to Vanderbilt in the future. We did not want to create the appearance of a “back door” into the PhD program, and we did not want to encourage passivity in the students admitted or in the faculty mentors responsible for preparing them. But this does not mean that the program makes no promises. On the contrary, Bridge students are guaranteed support and mentorship in a number of concrete forms, described below, and receive an explicit commitment that they will get the personalized attention, guidance, and one-on-one mentoring relationships that will allow them to develop — and to demonstrate — their full scientific talent and potential. This philosophy is more than a platitude; the program has been formulated with oversight by the appropriate Deans at both universities, who hold the program’s directors accountable for its success.

Identifying and Evaluating Students with the “Right Stuff”

The continued use of standardized tests — in particular the GRE — as a filter for determining who gets in to graduate school is a major factor in the ongoing, massive underrepresentation of minorities and women in STEM PhD programs.

As shown in Figure 2, GRE scores are not blind to the demographics of test-takers,. Indeed, the correlation of GRE score with gender and ethnicity are among the strongest correlations in the exam (along with socio-economic status). Consequently, adopting a cutoff GRE score (a score of 700 on the quantitative portion is typical in STEM PhD programs) leads to only ~30% of all women in the physical sciences, and only ~5% of all African Americans in the physical sciences, “making the cut” for PhD admissions.

An interview protocol with a scoring rubric designed to measure the seven facets of “grit” demonstrated by Angela Duckworth and others to be strong — and unbiased — predictors of student potential, is a much more robust and fair approach to identifying which students actually have “the right stuff” to succeed to the PhD and beyond.

Facilitating a Successful Transition to the PhD

The vehicle by which successful transitions to the Vanderbilt PhD program are realized is through carefully orchestrated student-faculty mentoring relationships focused on research. We have found that the extent to which a student is successful in developing one-on-one research-based relationships with faculty mentors — mentors who may very well become the student’s PhD advisor — is the single most reliable predictor of the student’s eventual admission into the Vanderbilt PhD program. Faculty mentors not only provide key guidance on course selection and research topics, they also become the student’s most important advocates in the PhD admissions process. The fact is that a student who is well known to the faculty of the admitting department is more likely to have their potential for success evaluated on the basis of direct faculty interaction, and not simply on how the student appears “on paper.”

It is thus the explicit goal of the Bridge program that its students will be well known by the Vanderbilt faculty by the time that they are ready to apply to the Vanderbilt PhD program of their choice. Indeed, fostering individual research-based mentoring relationships between Fisk students and Vanderbilt faculty is at the very heart of the Bridge program, and is the guiding principle for all other programmatic design considerations. To that end, the Bridge program includes the following key elements, requirements, and benefits:

  • Participation in supervised research, at Fisk or Vanderbilt (or both), during at least the second academic year of the program, and participation in supervised research at Vanderbilt (or at an affiliated research site) during at least each summer of the program. Students are required to produce a publication-quality master’s thesis.
  • Assignment of both a Fisk advisor and a Vanderbilt advisor. Joint mentoring allows tracking of student progress and helps to ensure student readiness for PhD-level work.
  • Scheduling of at least two meetings per year with the Bridge program steering committee to review progress and receive guidance, in addition to the day-to-day interactions with primary faculty advisers. This helps keep key personnel abreast of student progress, helps to keep each Bridge student on the PhD program’s “radar screen”, and helps PhD program directors in planning the needs of each year’s incoming PhD class.
  • Requirement of at least B grades in all graduate courses, with at least one of these courses being a core PhD course taken at Vanderbilt. This allows the student to demonstrate competency in a core PhD course, which is essential to demonstrating promise for PhD study. Typically, Bridge students take several core PhD courses at Vanderbilt. Together with a judicious selection of courses taken in fulfillment of the MA degree at Fisk, many Bridge students complete most of the course requirements for the PhD by the time they apply to the Vanderbilt doctoral program.

Underlying Programmatic “Theory”

Recognizing and nurturing unrealized potential in students: In formulating our admissions strategy, we have abandoned the usual mindset of filtering applicants on the basis of proven ability to one of identifying applicants with unrealized potential that can be honed and nurtured. Recognizing that potential takes a number of forms, and often plays out differently for each student. One student’s undergraduate transcript might show a low GPA that, on closer inspection, is the result of a slow start but a clear upward trajectory. Another may have an excellent GPA but missing upper-level courses in the major because they were simply not available at the undergraduate institution. Still another may simply have made a strong positive impression on a faculty recruiter during a poster presentation at a national conference. At the same time, we have formed strong, positive relationships with colleagues at numerous minority-serving institutions. As we get to know these undergraduate programs better, we are able to make more informed evaluations about specific strengths and weaknesses of incoming students. A report studying strategies for building effective partnerships with minority-serving institutions3 found that undergraduate mentors at these institutions take a very active role in advising their students, and will actively steer their students away from graduate programs that they do not trust will nurture their students’ success.

Tracking the second derivative of student performance: We constantly monitor student performance and intervene as soon as we detect an inflection in trajectory. For example, we track the courses that Bridge students enroll in as part of the advising process, and then actively monitor their progress by asking their instructors to promptly notify us at the first signs of concern. One-on-one tutoring is provided, as needed, by advanced graduate students or postdocs, and course-load adjustments are made mid-stream if it is determined that remedial instruction is required before re-enrolling in the course. These mid-stream adjustments typically involve the student taking an incomplete in the course, to be completed in a subsequent semester, and instead either first taking a lower level course or participating in a directed study course custom-designed to fill preparation gaps to ensure eventual success in the required graduate course. At all times, full time enrollment status is maintained to ensure satisfactory progress and eligibility for financial support.

Outcomes

Since its inception in 2004, the Fisk-Vanderbilt Masters-to-PhD Bridge program has attracted nearly 120 students, 85% of them underrepresented minorities, 45% women. Of these, 82% have either already transitioned to the Vanderbilt PhD program, to another PhD program of their choice, or are making satisfactory progress toward that goal. In addition, our students have been awarded the nation’s top graduate research fellowships from NSF (GRF and IGERT) and NASA.

The program’s key design considerations can be summarized as follows:

  • Focus on retention. Direct programmatic efforts toward fostering one-on-one mentoring relationships between students and potential PhD advisers, through enrollment in core PhD courses and through research assistantships in PhD faculty labs. When faculty know a student personally, and can vouch for their performance in coursework and in the laboratory, they can effectively and persuasively advocate for the student based on a holistic evaluation of the student’s ability.
  • Focus on recruitment, not competition. Direct recruitment efforts on truly broadening participation by emphasizing potential instead of already proven ability. Be willing to take risks in admissions, and then erect scaffolds of support to ensure success. Competing with other selective institutions for the few highly sought applicants who stand out in traditional metrics does little to address the needs of the national STEM workforce.
  • Involve key decision-makers in programmatic design and oversight. Faculty who lead graduate admissions must be active stakeholders in the process of matriculating, supporting, and monitoring students. Deans who oversee academic units must commit to work with — and place accountability on — programs that fail to retain students.
  • Stop using the GRE as a filter. Instead, use metrics (such as “grit”) that have been shown to be less biased against minorities and women, and that have been shown to be far more predictive of the types of qualities we (should) actually care about in our graduate students — the promise and potential to succeed to the PhD and beyond.

Keivan Stassun
keivan.stassun@vanderbilt.edu

References

(Endnotes)

1  National Science Board, 2003, “Report of the National Science Board Committee on Education and Human Resources Task Force on National Workforce Policies for Science and Engineering,” NSB 03-69

2  Norman, D. 2009, “Underrepresented Minorities in Astronomy: Higher Education.” A Position Paper submitted to the Astro2010 NAS Decadal Survey in Astronomy and Astrophysics:
https://arxiv. org/ftp/arxiv/papers/0903/0903.4506.pdf

3  Stassun, K.G. 2003, “Enhancing Diversity in Astronomy: Minority-Serving Institutions and REU Programs: Strategies and Recommended Actions,” Bulletin of the American Astronomical Society, 35, 5, 1448

4  Lange, S. 2006, “The Role of Masters Degree Transitions on PhD Attainment in STEM Disciplines for Students of Color,” PhD Thesis, University of Washington

5  Syverson, P. (2003, March). Data Sources. Graduate School Communicator, XXXVI, 5

6  Miller, C. “Admissions Criteria and Diversity in Graduate School.” American Physical Society. APS News, February 2013, 22.

7  Miller, C., & Stassun, K.G. 2014, “A Test that Fails”, Nature, 510, 303

8  Stassun, K.G., Sturm, S., Holley-Bockelmann, K., Burger, A., Ernst, D., & Webb, D. 2011, “The Fisk-Vanderbilt Master’s-to-Ph.D. Bridge Program: Recognizing, Enlisting, and Cultivating Unrealized or Unrecognized Potential in Underrepresented Minority Students”, American Journal of Physics, 79, 374


These contributions have not been peer-refereed. They represent solely the view(s) of the author(s) and not necessarily the view of APS.