Today (September 22nd) is the *last day* to submit comments to the National Academies of Sciences, Engineering and Medicine study, “Revitalizing Graduate STEM Education for the 21st Century”. Click here to find out how you can submit today and make yourself heard. The Board of Directors of Future of Research has submitted the following statement:
Graduate students are an investment in the future of science, and therefore are in the best shape to guide the future of the scientific enterprise. However, they are not equipped with the necessary resources to succeed within the current scientific climate, in particular as it relates to their transition into productive, independent careers, either within or outside academia, on a long-term basis. Even well-meaning academic mentors are not always aware of these needs or are unable to help trainees in their lab transition into careers outside academia. This is particularly important given that the majority of trainees will be employed in non-academic careers (National Science Foundation n.d.); that 80% of U.S. biomedical PhDs are currently pursuing postdoctoral training (Sauermann and Roach 2016; Kahn and Ginther 2017); and that the number of biomedical PhDs in the U.S. labor market exceeds the number of jobs requiring biomedical PhDs (Mason et al. 2016). Many graduate students may not have the opportunity to improve their training due to limited professional development resources at the university level.
There is a definite need for more offerings from various stakeholders for helping scientists become more prepared for particular careers. Online courses could be useful, but the Committee could further discuss the means by which institutions and mentors can best prepare students for careers outside of academia, and/or consider the specific needs of mentors themselves. More seminars and workshops discussing the types of skills needed and potential ways to obtain them would be helpful to implement as a system at the university level across the country. The Committee could address this by paying attention to those programs which are highly successful at a particular university. In addition, particular attention should be paid to the role that Master’s degrees can play as a career decision point. Those that may not have otherwise entered PhD programs may be encouraged to do so by this experience; likewise, this can provide an opportunity for junior scientists who choose not to enter PhD programs to still have a valuable scientific training without investing in further education that would not be suitable for their career goals.
Incentives for mentors to encourage trainees to pursue their desired career path need to be more prominent, regardless of whether or not that is a tenure-track faculty position at a research-intensive institution, currently the direction incentivized by the culture of biomedical academe. In that regard, we recommend the Committee discuss potential barriers that mentors might see in terms of training graduate students for non-academic careers, and provide them with resources to better guide their students. Students are most often splitting their time between lab work and outside activities which might be valuable for their career progression, therefore incorporating these activities as part of their professional development and getting more faculty buy-in is critical for their professional development. Training should be a central component of graduate programs, not an afterthought to be pushed outside “working hours”, nor requiring graduate students to consider additional employment (Dolgin 2017). Given that graduate students are trainees, and their training is evaluated over the course of their PhD program, so should the mentoring of mentors be evaluated in parallel, to ensure that those providing training are doing so in a competent fashion, and are able to undergo improvement themselves. Institutions and mentors should ensure that they are fulfilling their responsibility to train future scientists.
The lack of diversity in graduate populations is an issue that begins from the point of admissions (Posselt 2016; Posselt and Grodsky 2017). There is an over-reliance of metrics such as the GRE in graduate admissions, and a wide body of literature has recently demonstrated that this metric not only does not provide any prediction of scientific success, but also acts as a barrier to attracting the best talent by selecting for unrelated factors, such as socioeconomic status (summarized by the University of Michigan PIBS program here, which has recently dropped the GRE requirement for admissions). The Committee should make recommendations in terms of how to increase diversity at the university level, using information from programs such as the Mellon Mays Undergraduate Fellowship (http://www.mmuf.org/) and the Meyerhoff Scholars Program (http://meyerhoff.umbc.edu/). This could be done via updating policies at the time of admission into graduate school, creating additional policies to attract a more diverse population of students to attend a particular university (for example more scholarships, and/other types of financial aid), and training and evaluating mentors to ensure that graduate programs foster an inclusive environment.
The lack of data on career outcomes for trainees is one of the biggest challenges in training the next generation of scientists. Having these data available from specific institutions would positively impact the training programs offered in universities to better prepare graduates for different career paths. Clear data on this topic would also help with determining the types of skills that mentors need to teach their students at the university level to allow them to be successful in their careers. We advocate that any institution with a graduate program should be required to publish career outcomes data in a standardized format that can allow national aggregation of data, which is currently being encouraged by the AAU, NIH BEST Consortium, and Rescuing Biomedical Research (see AAU and BEST posts).
Recent data has also shown that there is an egregious problem, reaching epidemic proportions, in the mental health of graduate students. A recent study in Belgium demonstrated that one in two PhD students experiences psychological distress and one in three is at risk of a common psychiatric disorder (Levecque et al. 2017). Data from a survey at the University of California, Berkeley (here) suggests that this trend also exists in the U.S. While in and of itself, this is an alarming statistic, and should be taken into account when considering the general health of graduate students, studies have also shown that mental health generally affects productivity (e.g. see (Rajgopal 2010; Bubonya et al. 2017)), and so likely severely affecting the productivity of graduate students.
Overall, we urge the committee to consider as fundamental to graduate education the tenets that trainees receive training, and mentors provide mentoring. While increasing productivity is an important component of the development of a scientist, an over-reliance on metrics such as publications, and ignorance of the development of skills that can directly or indirectly improve scientific quality and productivity, can lead to perverse incentives that may adversely affect the quality of science produced, and reduce the joy in scientific discovery that graduate education can and should instill.
Submitted on behalf of the Board of Directors of the Future of Research by Executive Director Gary S. McDowell, PhD.