Study Finds Dramatic Growth in Numbers of “Supporting Scientists” on Research Teams

Posted January 2, 2019

As university research has become more complex and interdisciplinary, laboratory teams have grown in size, with increasing numbers of specialists in such areas such as statistical analysis, electron microscopy or mass spectrometry.

A paper published December 10, 2018 in the journal Proceedings of the National Academy of Sciences takes a closer look at these specialists, who perform essential roles in research and resulting academic publishing – but who may never lead production of a journal paper themselves. These supporting scientists often do their work outside the traditional tenure track and may never obtain permanent positions as professors.

“More and more critical scientific work is taking place in teams, and there are people who are building their careers supporting these teams,” said John Walsh, a professor in Georgia Tech’s School of Public Policy and a co-author of the study. “These supporting scientists are here to stay as a part of our scientific workforce, but we don’t really have a career system that accounts for and recognizes the contributions they make.”

Since the 1960s, the ranks of these “middle authors” has grown from 25 to 60 percent of the research authors publishing in the three fields the study examined. With titles such as “research scientist,” “laboratory technician” or “postdoctoral fellow,” these authors often move from one temporary assignment to another, and may drop out of research publishing in as few as five years after receiving a Ph.D.

“There seems to be more volatility in scientific careers,” Walsh noted. “Careers are getting shorter, and the point at which authors drop out, the half-life of a scientist, is getting shorter.” The study found that the time at which half of a cohort has left academic publishing has declined from 35 years in the 1960s to only five years a half century later.

Walsh and colleagues Staša Milojevic and Filippo Radicchi from the School of Informatics, Computing, and Engineering at Indiana University used data from the Clarivate Analytics Web of Science to study the changing demographics of scientific careers by looking at researchers in the fields of astronomy, ecology and robotics. They examined the careers of 71,164 scientists in astronomy, 20,704 in ecology and 17,646 in robotics to determine when publishing careers began and the publishing roles played by individual scientists. The National Science Foundation-supported research analyzed millions of bibliographic items produced during the study period.

The researchers looked for factors that might predict the career length for newly minted Ph.D. scientists. They found that the long-term survival of lead authors correlated with the number of publications during their first five years, while the success of supporting scientists didn’t seem to have a predictor.

The study wasn’t able to determine where the dropout scientists went after they stopped publishing, but Walsh notes there are number of career choices that would utilize Ph.D. skills – teaching, research administration or industrial research – without the expectation of traditional academic publishing.

With academic research based on the conventional roles of principal investigators and graduate students, the supportive scientists necessary for today’s research might not find a place in traditional college and university career paths.

“If you build the hiring, promotion and compensation systems on a model of a principal investigator and graduate students, these important contributors may be left out,” he said. “We may need to rethink the career and reward system because these specialists are becoming a larger and larger share of the scientific workforce.”

What’s causing the shift toward more transient members of research teams? Walsh says factors include the need for large teams to take on big science challenges, and competition for research support that drives the division of labor in team-based approaches – similar to what happened in factories – to get work done faster and more efficiently.

“There is very strong pressure to be good and to be fast,” he said. “One of the effects of what has been called the bureaucratization of research is that as groups get larger, you see more specialization and people who become permanent staff members who help support the team.”

Walsh said the growth of this “temporary workforce” represents a change in the university research enterprise.

“A significant and growing share of authors in each of these fields we studied spent their entire career as part of a research group, but never as a leader,” he said. “We saw this as strong evidence of a transition in the organization of scientific work.”

This work used Web of Science data by Clarivate Analytics provided by the Indiana University Network Science Institute and the Cyberinfrastructure for Network Science Center at Indiana University. This work was supported by National Science Foundation Social, Behavioral & Economic Sciences (SBE) Office of Multidisciplinary Activities (SMA) Early-Concept Grant for Exploratory Research (EAGER) grant SMA-1645585.

CITATION: Staša Milojevic, Filippo Radicchi, and John P. Walsh, “Changing demographics of scientific careers: The rise of the temporary workforce,” (Proceedings of the National Academy of Sciences, 2018). https://doi.org/10.1073/pnas.1800478115

This story by John Toon originally appeared in Georgia Tech Research Horizons in December, 2018.