Half of all tenure-track faculty members in the science and engineering disciplines leave their research universities within 11 years of being hired, according to a study released in the journal Science Thursday. The study tracked 2,966 assistant professors at 14 research universities.
The report , by Deborah Kaminski of Rensselaer Polytechnic Institute and Cheryl Geisler of Simon Fraser University in Canada, studied American research universities, including Cornell University, Virginia Tech and the University of Maryland-Baltimore County.
Men and women stayed on roughly at the same rate over all, though women had lower median retention rates than men in many disciplines; 64.2 percent of the assistant professors moved up to become associate professors at the same institution where they were hired.
Median Time to Leave a STEM Faculty Job
The field of mathematics had the lowest median retention rate, with 7.33 years for men and 4.45 years for women. Mechanical engineering had the highest retention for men, at 16.19 years, while biology had the highest retention rate for women, at 16.36 years.
Median Time to Exit, by Discipline and Gender
Kaminski said the study was unusual because it was longitudinal, instead of comparing the number of faculty members coming into an institution in a particular year against those leaving. One important finding, she said, was that women were retained at the same rate. “But there are not enough women who are Ph.Ds who are going into academics, and science in particular,” she said.
Hiring at universities is in some ways akin to the Air Force hiring a fighter pilot, Kaminski said, because there is a heavy investment in the new hires. According to the study, it can take up to 10 years to recoup the investment of a new hire in a STEM field because start-up costs -- the money required for a research lab and research program -- can be as high as $1.5 million.
“Our work confirms the importance of the late pre-tenure period as a period of critical risk in the retention of faculty in STEM,” the report states. Over all, faculty members were less likely to depart after tenure than before tenure. “If a university expects to grow its faculty, it has an even greater challenge. Simply staying the same size requires considerable hiring and mentoring, even without considering retirement,” the study says.
Geisler said they began working on the study thinking that women were not being retained at the same rate. “It does look like more of an issue of how many women are in the pool,” she said. Women might decide on an academic profession depending on how compatible the career is with their desire to raise a family, she said.
Women represent about 27 percent of STEM faculty members at the 14 universities in the study, and even though their are numbers on the rise, it could be 100 years before 50 percent of faculty members in science and engineering are women, the report states.
Gerald Marschke, a labor economist at the State University of New York at Albany, who looked at the report, cautioned against jumping to conclusions about the analysis. “Science is highly meritocratic, which is a good thing because that makes scientists work hard and be productive and it also attracts scientifically talented people to the enterprise, all good for the country and the economy,” he said. “A disadvantage, of course, is that it discourages good folks who want some balance in their lives from entering science.”
As for mathematics, the low retention numbers in the field might be because mathematics does not have the start-up cost that disciplines like physics or chemistry have when it comes to a new hire. “New math faculty don’t require this sort of commitment, and maybe the departments are more willing to give someone a chance knowing they can let them go if their research is not felt to measure up, without seeing a substantial financial commitment wasted,” said Jim Maxwell, associate executive director at the American Mathematical Society.
Maxwell pointed out that the study looked at only 14 institutions, when more than 200 colleges in the United States award Ph.D.s in mathematics, and said he wasn’t sure the study was representative of all the 200 departments.
Another expert said STEM disciplines tended to be painted with a broad brush, and the importance of this report is that it highlights the differences within disciplines. That might be important because real change can happen at the departmental level, said Kiernan Mathews, director of the Collaborative on Academic Careers in Higher Education at Harvard University.
Mathews said there is no consistent collection of data across universities when it comes to retention rates. “Either they don’t do it at all or it is rudimentary,” he said. “The question needs to be asked as to why these people are leaving. There is an inability or an unwillingness to do it. The method for doing this is an exit interview and the response rates for them tend to be low,” he said.
Another important criteria for future studies, he said, is the importance of race and ethnicity in retention. The impact is meaningful because a sizeable number of STEM faculty tend to be foreign-born, Mathews said. “What impact does citizenship play on retention rates?” he said, especially now when the standard of living in competing countries is rising.
Mathews said the report was a call to action to institutions to start creating good analysis, even when the data might be incomplete.