Skip to Content

New Study: The Longer You’re in a Ph.D. Program, the More You Look Outside Academia for Work

This blog post was edited on 7 May.

Often missing from the discussion over whether there are too many Ph.D. scientists being produced and too few jobs available in academia for them are empirical data on whether those scientists do, in fact, desire academic positions–and whether that desire changes over time. 

“We always assume that people want to get academic positions and not many get them, and so we think, ‘Oh, they all want it and there’s a big imbalance,’ ” says Henry Sauermann, a behavioral economist at the Georgia Institute of Technology in Atlanta. “But maybe nobody really wants them. Maybe those people, when they’ve watched their advisers, have realized that it’s a really tough job, not much time for research, a lot of time spent writing grants and so on. It could go either way. We don’t really know what these preferences look like.” 
In a paper out today in the Public Library of Science ONE, Sauermann and his colleague  Michael Roach, a decision scientist at the University of North Carolina in Chapel Hill, weigh in with some findings that could be important for framing future discussions: As students progress through their doctoral programs, they become less and less likely to want an academic job after they graduate.
To find that out, the researchers surveyed 4109 Ph.D. students across 39 tier-one U.S. research universities on their career preferences and how attractive they viewed academic, industry, government, and other career options. The students were enrolled in the life sciences (59% of those surveyed), physics (23%), or chemistry (18%).
Sauermann and Roach found that, on average, students in the later stage of their Ph.D. programs (defined as those actively looking for a job or planning to do so within the next year), held less favorable views of faculty teaching and research jobs than did students in earlier stages. While faculty research was rated the overall most attractive career path for all respondents in both the life sciences and physics, the percentage of life sciences students who rated faculty-research positions as either “attractive” or “extremely attractive” fell from 78% in early-stage students to 67% in later-stage students. In physics, those numbers fell from 81% to 72%. In chemistry, where the majority of students rated industry jobs as “most attractive,” those who rated faculty-research jobs as “attractive” or “extremely attractive” fell from 62% to 47%, while those who preferred a job in industry jumped from 70% in early-stage students to 76% in late-stage students.

Because the study wasn’t longitudinal–the early-stage and late-stage students are different cohorts–the authors also asked the later-stage participants whether their views toward various career options had changed over their time in a Ph.D. program. While the authors admit that such questions are subject to flawed memory, their survey did reveal some interesting findings. Faculty research positions decreased in attractiveness in 18.3% of life science students (compared to 8.7% for whom it increased in attractiveness), 13.4% of chemistry students (8.4% increased), and 19.9% of physics students (7.1% increased). 
Inexplicably, the job sector that experienced the biggest gains in relative attractiveness among late-stage students assessing their own change was government. Here, 17% of physics students, 18.6% of life sciences students, and 22.9% of chemistry students reported that they found government jobs more attractive later in their grad programs than they did when they first started (compared to 11.9%, 9.9%, and 7.7%, respectively, who reported that government jobs were now less attractive). 
Strangely, this effect isn’t seen in the inter-cohort numbers from the study; there’s little to no change in students’ preference for government jobs when you look at the ratings of both early- and late-stage students. Saurermann told Science Careers that he’s not entirely sure what motivated the late-stage students in his survey to report that their appreciation of government jobs had grown, but he thinks it has something to do with a perception (whether’s it’s true or not) that such jobs are secure, have access to funding, and allow scientists to do academic-like work without the added stress of tenure, promotion, and teaching.
More broadly, Sauermann says that it’s difficult to tell from his findings what could be motivating students away from academic research positions the longer they spend in grad school. It’s possible, he says, that once students see the reality of the academic research life up-close, they become disillusioned by it and seek out alternatives. On the other hand, they might encounter more and more job opportunities over their time in grad school, making them realize they’d prefer a career they hadn’t previously considered.
Whatever the explanation, Ohio State University economist Bruce Weinberg is heartened that such empirical data is finally beginning to accumulate.
“This is a set of issues that people have been thinking about and wringing their hands over for many, many years, and I think there isn’t going to be a single piece of evidence that we’re going to be able to identify and answer this question with,” Weinberg says. “[But] I think this is one very valuable component to answering that question.”
Paula Stephan, an economist at Georgia State University (whose book, How Economics Shapes Science, was reviewed by Science Careers in January), adds that, “it’s pretty encouraging that not all of them say they want research faculty positions, given the small number of those positions out there.”
Both Weinberg and Stephan say the study is limited by the fact that the survey didn’t track students over time but instead looked at different cohorts who entered their Ph.D. programs at different times. It’s unclear whether there may have been economic or social factors that changed students’ preferences in the last few years, and such factors could undermine the study’s results. But Sauermann says he plans to follow the cohorts over the next several years and re-survey them on their experiences post-graduation–data that should address Weinberg’s and Stephan’s concerns.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>