Where Have all the Researchers Gone?

An article published by Nature in early November1 suggests that almost two-thirds of researchers have considered quitting research…and about 15% have actually done so.

Thousands of Nature readers responded to an online poll on the scientific publishing group’s website. The results, although not surprising, remind us of an ongoing trend in science research: the desire to leave it all behind for greener grasses.

Why is everybody leaving?

Historically, the PhD has served as a training program for future academics. The expectation was that a student would undergo an apprenticeship under the supervision of a more experienced scientist before seeking a faculty position. Unfortunately, the number of newly-minted PhD’s has far surpassed the number of available faculty positions, both in the USA and abroad. Some PhD’s try to wait it out by doing a second or third post-doc, only to find that the current climate in academia will be very unfavorable for many years to come. By some reports, less than 17% of new STEM PhD’s will find tenure-track positions within a few years of graduation3. For many post-docs, a tenure-track position will never surface.

Those who are not on the tenure track can seek employment as adjunct (or contingent) faculty, a position defined by its heavy course load, low pay, and minimal health benefits. One contingent faculty member from Ithaca College in Ithaca, New York says that “I have 16 years of physics teaching experience and make less than my grader”3. This is not a feasible option for doctoral students graduating with thousands of dollars in both undergraduate and graduate debt.

For those who are fortunate enough to find an academic position, research funding has proven challenging. The cutting-edge research that is more likely to be funded often requires access to more powerful and expensive equipment, but government funding has remained stagnant since the 1980’s. Professors also face the rising number of applications per grant and high entrance barriers. What was once considered an early career award is now routinely given to professors with more tractable publishing records. This leaves newer professors struggling to find funding for excellent ideas due to their low paper count.

For professors at universities that require advisers to pay both their students’ stipends and tuition—like Cornell—the funding situation becomes even more problematic. And with the rise of student unions and their associated costs, it is likely that there will be more cash-strapped professors, less PhD students, and lower scientific productivity.

The cost of research—like the cost of higher education—is becoming prohibitively high. Academia is becoming unappetizing to the brightest minds, who know that they can find both a lucrative position and their sanity elsewhere.

 Where has everybody gone?

Academia is the “alternative” career path now. Most students with STEM PhD’s are going into industry, with most working for the private sector, and some working for the government. STEM PhD’s enjoy a wide range of employment opportunities, from analytics to policy to consulting. Some PhD’s go on to receive a professional degree (MD, JD, MBA), enabling them to practice medicine, law, or enter the business world.

There are benefits to finding non-academic employment. First, there are more available positions, more stable jobs, and less unemployment. Another incentive is higher pay. The average post-doc salary in the U.S. is $45,000, with many making much less4. Compare this to the starting salary of a PhD in industry, which is often over $80,000. Data published by the National Science Foundation shows that the median salary of doctoral recipients with a job commitment in industry is higher than that received by those with postdoctoral and academic commitments across all published disciplines5.

Should we do anything about this?

There is a glaring problem with the current PhD program: it is designed to prepare students for an occupation that will likely never be available to them. There are two solutions to this problem: one is to maintain the PhD as is and address the surplus of students relative to tenure-track faculty positions. The second is to reimagine the PhD to better prepare students for non-academic positions.

Keeping the traditional PhD

There is a contingency that believes that the PhD should retain its original purpose as a training ground for future academics. As long as there is an academy, there will be a demand for future academics. The problem then lies in addressing the number of PhD’s.

One solution is to limit the number of PhD students that matriculate each year. This could involve simply rejecting an applicant or funneling rejected doctoral applicants into a Master’s program. Although the Master’s usually helps develop the skills and expertise requisite for employment in industry, students must pay for the degree, often with little financial aid. This presents a problem for students who are already in debt from trying to afford an undergraduate education and do not have the out-of-pocket resources to pay for another degree.

The alternative is to admit PhD students at the current rates, but make the selection process throughout the PhD program much stricter. Programs could implement a qualifying exam if one doesn’t exist, and raise the passing standards for those that do. Programs could require tangible progress in the research process (not necessarily in terms of publications, but in the form of conferences, technical talks, or a development of expertise in the field). Admission to Candidacy exams could be more difficult to pass. Programs could require students to take their Admission to Candidacy exams no later than the first semester of their third year, that way if they do not pass, they can graduate with a Master’s without having wasted too much time in academia.

This is not an ideal alternative. It highly exploits students and puts too much pressure on them to succeed. This leads to high levels of competition, stress, burnout, data forging, and other undesirable feedback. It also encourages universities to ignore its original purpose—to educate the next generation—instead of trying to weed them out at every opportunity. Weeding, if it occurs, should happen during the application process, not after matriculation.

In the absence of sufficient PhD students, faculty may resort to hiring staff scientists to conduct experiments. Not only would this give STEM scientists jobs, but the research output would likely be higher and of better quality. There are associated costs with this strategy. Staff scientists will request higher pay for their expertise. Also, they will be considered employees and entitled to overtime and benefits, just like employees at larger companies in industry. Unlike graduate students, staff scientists are also unlikely to teach classes, grade papers and serve as teaching assistants. They are also unlikely to work the 60+ hour schedules that many graduate students keep. Although graduate students are expensive, they are paid very little compared to the tasks they perform on behalf of the university and the hours they work on behalf of their research groups. In this respect, hiring staff scientists will be costlier for a professor.

Develop a PhD for Non-Academics

Most STEM PhD’s find employment in industry. The PhD isn’t always necessary to secure a stable, lucrative position in industry, but the skills developed by the PhD (ie., complex problem solving, ability to work in a collaborative environment, communication to both technical and lay audiences) are highly sought by employers. Some jobs in the government and private sectors often require PhD’s for mid-level positions, such as science policy.

Because PhD’s are required for non-academic positions, PhD programs should not restrict non-academics from applying. However, the PhD should be revamped to help these students pursue their career goals. For instance, many students want to learn about leadership, management, business, and negotiations. These are desirable skills to potential employers in industry. These skillsets are not developed on the bench. Consequentially, students that are interested in learning these skills must do so with their little free time. A non-academic PhD would require at least two courses—and perhaps a formal exam—on any of these topics. This would be in lieu of one or two science-based core classes.

Lack of relevant work experience is the greatest challenge PhD’s face when entering the corporate sector. This experience could be attained through internships and co-op programs. Many professors discourage their students from pursing any opportunity that would detract from the time spent in lab. Some students are paid through fellowships that limit their working hours outside of the lab, making it impossible to work through a co-op program. A non-academic PhD would require participating in either an internship or a co-op lasting no less than an academic semester and concluding with a formal presentation to the student’s academic committee.

Opponents to the non-academic PhD are likely to cite that industry doesn’t care about publications as much as the academy. Consequentially, students in non-academic PhD programs will prioritize developing skills over publications. Since publications are the currency of academia and are used as metrics that determine tenure and funding, professors will be less likely to fund students who don’t share their desire to publish regularly and often. Consequentially, a non-academic PhD will be most successful if the student is supported by an industry fellowship and working under the supervision of tenured faculty members, whose careers are less dependent on publication record.

The future of STEM research in the US depends on the number and quality of its researchers. Talented students who dream of the professoriate need to be encouraged to stay in academia, and professors need more support to be successful in academia. Industry-bound students need to be appropriately trained. As it stands, the PhD is failing both the academy and the industry. This has lead to poorer research and insufficiently-trained students. This threatens the future of STEM in America, and compromises our ability to innovate solutions to the most challenging problems facing our society today.


  1. “Hard work, little reward: Nature readers reveal working hours and research challenges”, Kendall Powell, Nature Trend Watch, 04 November 2016. Link.
  1. “Science and engineering indicators 2012”. National Science Foundation. January 2012. Link.
  1. “Students launch online campaign for Ithaca College contingent faculty demanding decent wages, job security”, Jolene Almendarez, The Ithaca Voice, 19 October 2016. Link.
  1. “US law could increase postdoc pay—and shake up research system”, Heidi Ledford, Nature News, 19 May 2016. Link.
  2.  “Median basic salary of doctorate recipients with definite commitments in the      United States, by position type and field of study: 2014,” in “Doctorate Recipients from U.S. Universities,” National Center for Science and Engineering Statistics Directorate for social, behavioral, and economic sciences, 2014.Link.


Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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