For decades, blue ribbon reports, studies, panels, and commissions have bemoaned universities’ lack of transparency about the career prospects of their Ph.D. recipients and postdocs. In particular, experts have criticized institutions’ failure to track and report how their doctoral and postdoctoral alumni fare in the labor market. As administrators at 10 U.S. research institutions warned in a Science article this past December, without knowing what kinds of jobs graduates can expect after graduation, students can’t “mak[e] informed choices about their pre- and postdoctoral training activities.”
But finally, like small green shoots hinting at spring, three efforts to close this knowledge gap launched in February. The December Science article announced the formation of the Coalition for Next Generation Life Science, which aims to make “advanced training in the life sciences more efficient and humane,” and the coalition committed to start providing information about training outcomes starting in February. As promised, the coalition’s data went live on the first of the month. That same day, the University of Toronto in Canada unveiled an online database of its own, the 10,000 PhDs Project, which tracks the professional fates of 15 years of doctoral graduates. And a February article in Nature Biotechnology summarized the career outcomes of 15 years of postdocs at the U.S. National Institute of Environmental Health Sciences (NIEHS) using a newly devised taxonomy of scientific jobs, plus related visualizations.
Each of these efforts contributes—though in varying degrees—to ending the career data dearth. But by easing this problem for prospective students, universities could well be creating a different, potentially significant—but ultimately very salutary—problem for themselves.
Mixed results
The University of Toronto’s very useful website lets visitors burrow deep into the data gathered by a team of undergraduates tasked with tracking down each of the 10,886 people who earned Ph.D.s between 2000 and 2015. The student sleuths succeeded admirably, locating 9583 individuals and documenting each one’s first or current employment status from “two or more reliable Internet sources,” according to an overview of the project.
Using an interactive dashboard, visitors can view the data by university division or school, graduation cohort, department, gender, and employment sector, in any combination or all at once. For example, half the molecular geneticists with degrees granted between 2004 and 2007 work in the post-secondary education sector, 46% of those women as research associates and 15% as tenure-track faculty. Of the men in that group, 7% and 67%, respectively, hold corresponding positions. Equally detailed data cover the graduates working in the business, government, and charitable sectors and even identify the entities that employ the graduates in every sector.
Prospective students won’t need long to learn the professional fate of any program’s and era’s graduates and—though, as stock brokerage ads warn, individual results may vary—to surmise what their own futures might hold. The University of Toronto’s alumni are “highly employable,” the overview states, working in a wide range of occupations in countries around the world—hardly surprising for Canada’s highest ranked university. Even a cursory cruise through the science results, however, reveals a steady decline of the percentage on the tenure track.
Compared with the University of Toronto’s comprehensive offering, the effort of the coalition schools—Cornell University; Duke University; Johns Hopkins University; the Massachusetts Institute of Technology; the University of California, San Francisco (UCSF); the University of Maryland, Baltimore County; the University of Michigan; the University of Pennsylvania; the University of Wisconsin; and the Fred Hutchinson Cancer Research Center (Fred Hutch)—is generally a distinct disappointment. On their website, the coalition institutions “commit” to gathering and publishing data on admissions and matriculation; median time to and percentage of degree completion; student and postdoc demographics; median time spent as postdocs; and, most importantly from the standpoint of those considering scientific training, “career outcomes for PhD and postdoctoral alumni, classified by job sector and career type.”
At this point, however, Cornell alone provides robust Ph.D. outcome information, though not for all of its science doctorates. Cornell University Graduate School offers data in considerable, searchable detail—including field, graduation cohort, employment sector, and citizenship status. Weill Cornell Medicine, however, and the rest of the coalition schools, only allow students to discern their chances of admission, how long they’re likely to spend becoming Ph.D. holders, and a demographic picture of their probable classmates. Beyond that, Fred Hutch reports the types of jobs held by 1142 postdocs between 1999 and 2016. Because it does not break the data down by era or indicate whether those dates refer to completion of postdoc appointments or to something else, however, the numbers have little predictive value. Nor, unfortunately, does revealing information from a study of the career outcomes of UCSF postdocs, which we discussed in 2016, appear on the website. So, for nearly all the institutions, prospective students cannot answer the most important—and most individual—question: Does getting a Ph.D. provide career opportunities that justify the time, effort, and forgone earnings involved?
Presumably the schools will provide the further data they promise, though they don’t specify when. That certainly shouldn’t be hard for institutions that pride themselves on researching the world’s most challenging questions. In 2012, we reported on Michigan State University’s successful experience tracking the careers of 20 years of Ph.D. recipients by hiring undergraduates to search social media. Many of the 3000 alumni turned up in minutes, though documenting some who had changed their names or moved abroad took longer. Still, $10 a head sufficed overall, and 2012 was eons ago in internet time.
Finally, in an effort to simplify and clarify collecting such data, the NIEHS article introduces a classification system to describe a large group of postdocs’ career choices. (We reported on a similar effort this past November). The authors demonstrate their creation’s utility through unexpected insights it helped them uncover. NIEHS postdoc alumni from the United States, for example, go into applied research in the for-profit sector much more often than do the international former postdocs, who are twice as likely as the Americans to land on the tenure-track doing basic research—but mostly not in the United States, where the majority of tenure-track jobs go to locals.
Also intriguing is the fact that postdocs’ “career outcomes … are heavily influenced by the types of jobs within close proximity” to the postdoctoral institution, the authors write. Comparing their results with information from the “broader NIH postdoctoral community,” the authors find that far fewer alumni of North Carolina-based NIEHS—the only one of the National Institutes of Health located outside metropolitan Washington, D.C.—go into science policy work than do their counterparts at the institutes close to the national capital’s many governmental and policy organizations.
Such findings illustrate the benefit of using a well thought-out system for classifying careers. One hopes that when—or, more likely, if—the academic community finally comes around to providing such needed information as a matter of course, institutions can agree on a standard set of career categories—either this taxonomy or some other—that will allow clear comparisons among universities.
A useful problem
Should the day finally come that not only the coalition schools but many others join the move to transparency on outcomes, a significant problem that has long bedeviled aspiring scientists could well be solved. But I’m betting that increased career transparency could present universities with a whole new problem: If prospective students start making “informed” decisions about their futures, who will do the work in the labs?
The academic research enterprise has long functioned as a pyramid scheme based on cheap workers, notes labor market economist Paula Stephan in her magisterial book, How Economics Shapes Science. University scientists staff their labs with graduate students and postdocs recruited “with funding and the implicit assurance of interesting research careers,” she writes. Principal investigators (PIs) “especially” seek “students with academic aspirations” as the best “worker bees in the PI’s lab.”
Traditionally, many of these aspiring researchers have come to understand their true job prospects in the glutted academic labor market only after years of low-paid toil. But if even a cursory look at career data gives the lie to the “implicit” assumption of a waiting academic career, and if students start deciding based on that knowledge, will they still opt to become graduate students and postdocs? What if more of the informed students that the worthy transparency efforts produce act instead in their own practical interests and opt to use the next 11 years to become, say, neurosurgeons instead of Ph.D. holders and postdocs? Or to stay in academic labs long enough to get Ph.D.s, then opt for industry’s much better financial rewards? Or to spend 2 years earning Professional Science Master’s degrees and entrée to nonacademic science-based careers? Or even to devote 2 years and good math skills to securing MBAs and lifetimes of paychecks beyond the average academic’s dreams of avarice?
Should the old bait-and-switch or any intrinsic allure that graduate and postdoc research may hold no longer work, universities will still need to lure able young scientists to labor in their labs. Perhaps they will simply import lots more international students and postdocs, for whom entry into the United States and, potentially, its labor market is often itself a strong incentive. Or perhaps institutions will come up with something truly radical—say, attracting scientists by offering good salaries and appealing career ladders that don’t involve faculty status. Now that’s something that would be transparently good.
[“Source-sciencemag”]
