Is PhD training too narrow? (TRR-III)

What is a graduate program’s obligation to students who choose careers that do not involve research?

Should students take the wide road or a narrower one? (Credits here and here.)

Conscientious committees give birth to outsized, gangling progeny—like the voluminous Tilghman-Rockey Report (TRR; 1). The TRR issues 19 recommendations, some outfitted with multiple subheadings. Fortunately, readers won’t have to digest 19 posts, because I will consider only a subset of all the recommendations (2).

Today’s topic is what I shall call recommendation 2 of the TRR (2). Here the TRR’s principal concern was that biomedical PhDs in the US are trained for work in academic labs but not for the various different career paths many take after receiving their degree. It collected data showing that 16,000 US-trained PhD students entered biomedical science training programs in 2009, but only 9,000 received a PhD in that year; of these graduates, about 30% skipped postdoctoral training (1). Of PhDs in the post-training biomedical workforce, 43% are academic researchers (23% tenured, 20% outside the tenure track), while the rest are engaged either in research (18% in industry, 6% in government), or in non-research careers related (18%) or unrelated (13%) to science; about two percent of biomedical PhDs are unemployed (1,3). Thus as many as four of every 10 biomedical PhD trainees probably drop out of US graduate programs, and nearly one third of PhD graduates (31%) pursue careers outside research—despite the fact that research is what all these graduates are trained to do.

Something is clearly wrong, but exactly what? Somewhat opaquely, the TRR says: “Given the changing face of the biomedically trained workforce, the working group believes that graduate programs must accommodate greater diversity in anticipated career outcomes for students.” So, recommendation 2 seeks to broaden training programs in order to equip selected students for research jobs in science-related endeavors outside the lab. To do so, it proposes that NIH fund pilot programs aimed at channeling students into doctoral training related to other endeavors and/or MS degrees “designed for specific science-oriented career outcomes, such as industry or public policy.” (The latter approach, says the TRR, would require re-defining ‘success’ in evaluating NIH training grants. MS degrees, it seems, suggest a training program is not doing its job.)

In my opinion, the report aims recommendation 2 at the wrong target. The data shows that our biomedical PhD programs attract students with varied backgrounds and skills, of whom about 44% leave graduate school, while 30% of the rest pursue careers that take little advantage of their research training. From this I conclude that biomedical PhD training is a quite inefficient way to spend scarce resources, but also see no reason why that implies an obligation for PIs and research institutions to train students for non-research careers. Instead, my observations suggest that many applicants don’t understand what it takes to do biomedical research, and students who don’t like what they find in research labs are not furnished escape routes to more congenial pursuits. (Could the dearth of escape routes relate to their PIs’ need to keep those students working in the lab? I wonder.) Moreover, a recent survey (4) found that 89% of first-year biomedical PhD students at UCSF were considering a career as an academic PI, but the percentage dropped to 68% by the end of the third year. This 21% downward shift coincided with the period when they began to learn what research labs do. If my observations and this survey are correct, PhD programs should be obliged to inform applicants, before they matriculate, about their realistic prospects for a research-centered career and to help admitted students who are better suited for other endeavors to find other training paths.

Accordingly, my amended version of recommendation 2 would urge that every NIH-funded graduate training program be required to: (i) inform applicants forthrightly about the rigor of PhD research and the real (documented) career prospects of the program’s graduates (5); (ii) institute mandatory MS degrees for all students who satisfactorily complete three years of training; (iii) inform applicants that after the MS degree some students will go on to earn a research-based PhD, but others will choose (or be asked) to pursue a different course. Applicants should be told about available options for different courses and about obligations the program will (and will not) assume in helping students to follow them. Such obligations would vary among programs, allowing applicants to choose programs depending on whether they offer special tracks devoted to this or that non-research pursuit (6). The amended recommendation would also require reviewers who evaluate training programs to consider that MS degrees should confer no disgrace on students or programs, because: (i) re-direction into non-research careers can be advantageous for some students and necessary for effective graduate training of good scientists; (ii) an MS degree can serve as a branch-point for a student’s development, rather than a mark of her/his ineptitude or of a program’s callous regard for students.

Are these amendments elitist, cruel, or simply too narrow in defining the biomedical PhD? To the contrary: (i) because scientific research is hard and we pay students to learn how to do it, we should make training more effective, efficient, selective, and focused on skills faculty are qualified to teach; (ii) it is far less cruel to ask a few students to take an MS degree and shift to another course of study than to subject them to a long Darwinian struggle, only to discover that they cannot find a good job doing what they were trained to do; (iii) graduate programs are free (but not required) to offer students post-MS options for which the program takes direct responsibility; (iv) other programs will choose not to devise relevant alternative programs, but rather to teach MS graduates how science works, retain as PhD candidates those suited for research-intensive training, and help others find programs that better fit their goals and needs.

Martin Rosenberg, the chief scientific officer at Promega Corp., recently wrote a much more forceful argument for major change in the focus of biomedical graduate (and perhaps postdoctoral) education (7). Changes in the economy and other factors, he points out, have led to elimination of thousands of science-based jobs from industry, as biotech investment focuses on “nearer-term product development” and pharmaceutical companies undertake a “strategic shift” that is creating opportunities for academic labs and stimulates job growth in various support functions. He lists 16 of these private sector jobs—e.g., preclinical analysis, safety assessment, regulatory affairs, project management, etc.—which are not based on bench research and “have little to do with the focus of our entire academic training system.” Academic biomedical researchers know little or nothing about these opportunities, he says, and arrogantly disdain them as undignified in comparison to their “Ivory Tower” commitment to research. On both counts, I suspect he is right. He adds that at present “training is focused on driving independent bench research for faculty to achieve their funding, status and advancement.” But, he concludes, “bench experience is a necessary part of training, but it no longer suffices to prepare students for today’s job opportunities.” Here again, he is right.

Rosenberg’s prescription: “The science-education system [should] seek expertise . . . from the private sector to convey these employment opportunities to their trainees,” he says. His sketchy pronouncement—which we should certainly follow, by the way—differs from my prescription. While biomedical researchers should not try (or be required) to take responsibility for teaching specific skills necessary to profit from the job opportunities Rosenberg presents, three years of study for an MS degree would be ideal way to learn how science is done before undertaking further training in areas where he says job opportunities abound—e.g., business, regulatory affairs, project management, etc. In the meantime, researchers must also do their damnedest to make sure that they and the US biomedical research enterprise preserve the capacity for discovery that will drive continuing job growth for our trainees throughout the 21st century.

Having delivered myself of these pronouncements, I am assailed by an obvious question. Who am I to contradict the ideas of an expert working group chosen for its scientific prowess, knowledge of graduate education, and teaching experience? Treading this more personal ground depletes my confidence. Still, this is a blog, not wisdom literature, so I can risk a tentative answer. Perhaps the difference is that I feel more strongly than the TRR’s experts that scientists teach best what they best know how to do, leaving to students the prerogative and duty to decide what they really want to learn. As a teacher, I should be obliged to teach a selected group of students what a scientist does, for a period of up to three years, but not to guide them for three to four additional years, if they are not suited for research or decide they want to do something else.

Nonetheless—to argue for a moment against myself—I do realize prospective students may seek to avoid programs with a mandatory MS degree, where they may be told to choose between seeking further schooling elsewhere or finding a job that doesn’t require a PhD. So clear a breakpoint could reduce the number of applicants and matriculating PhD students, and perhaps (although I think it unlikely) their quality as well. And I am undeniably less generous than the TRR’s experts when I point out that society, not the student, pays for the training, and suggest, further, that the MS degree can furnish a useful opportunity for both students and teachers to decide whether a particular candidate should continue to prepare for a demanding profession.

Dare I speculate as to why the TRR’s experts take a broader view of their and my duties to graduate students? Is it incorrect, or even outright uncharitable, to guess that the broader view might reflect a desire to keep student workers in their labs, where (they may think) young people can learn what real competition is and choose to follow in the PI’s footsteps? Certain colleagues argue strongly for retaining a marginally performing student in the lab, in order to provide a complete learning experience (not, they insist, to get the project done). The same professors may also extol the value of Darwinian struggle for graduate students, who need to learn how to fail in order to learn how to succeed.

The personal ground now begins to tremble in earnest, so I’ll stop.


1. Biomedical Research Workforce Working Group Report. Pdf here.

2. Because the report does not number its recommendations, for convenience I give numbers to those recommendations I choose to discuss. This numbering system will be consistent in successive posts, but does not conform to the order of their appearance in the TRR.

3. Data for these outcomes was collected in 2009 or earlier, and is discussed on pp 23-28 and 32-33 of the TRR, cited in note 1, above. This data does not include PhDs trained outside the US, who represent an unknown fraction (less than half, perhaps as much as one third) of the post-training biomedical workforce.

4. CN Fuhrmann, DG Halme, PS O’Sullivan, B Lindstaedt, Improving graduate education to support a branching career pipeline: recommendations based on a survey of doctoral students in the basic biomedical sciences. CBE-Life Science Education 10:239-49 (2011). Pdf here. This study showed that most of the drop of interest in a research career reflected a specific loss of interest in becoming an investigator at a research-intensive university. It was temporally associated with a converse increase in interest in non-research careers. Interest in other research-centered careers (e.g., industry) did not change significantly. It should be noted that this study’s conclusion accords with the TRR’s recommendation, and differs from my own. That is, the study concluded that universities like UCSF should offer students branching pathways to prepare for careers that do not involve research, while I suggest that a mandatory MS degree would allow students to choose whether or not to pursue further PhD training, wherever it can be found.

5. As noted in the TRR (p. 35), Duke University posts information about outcomes of its PhD students by program. A quick look reveals, however, that the data from different programs varies in level of detail and does not extend to outcomes more than a few years after the PhD.

6. For instance, programs can affiliate with degree programs in business, public policy, or journalism to offer post-MS training suitable for who will seek non-research jobs in, respectively, industry, science policy, or science news. In my opinion, most students with research-intensive PhD degrees and postdoctoral training already qualify for laboratory research jobs in biotech or the pharmaceutical industry.

7. M Rosenberg, An “honorable” career in academia vs. an “alternative” career in the private sector. ASBMB Today, August 2012, Web here.


About biomedwatch
Professor Emeritus of Cellular and Molecular Pharmacology, University of California, San Francisco

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: