For hundreds of years the most ubiquitous conception of the scientist has been that of the lone empiricist, forever building on the progress of those giants on whom (s)he stands, figuratively going where no one has gone before. But a quick tour around the present’s scientific endeavors reveals a very different story–that of a collective project requiring the concerted effort of many talented professionals.
SCIENTIFIC RESEARCH IS A TEAM EFFORT
The percentage of science and engineering papers composed by teams increased at a near constant rate from 1955 to 2000, from roughly 50% to 80%, and the mean number of co-authors has nearly doubled in the same period of time.
Two sociologists at the School of Public Policy at the Georgia Institute of Technology in Atlanta, John P. Walsh and You-Na Lee, speculated as to how the size of the scientific team affects the way scientific research is carried out. They’ve concluded that research teams often use division of labor, task standardization and hierarchies within their teams to maximize empirical profit–transforming the scientific process into something more like an epistemic production line.
YOUNG SCIENTISTS MAY TRAIN FOR THE JOB, NOT THE JOURNEY
Walsh and Lee argue that this has direct practical import for the training of young scientists, and the potential path(s) of their ongoing academic career. The fact is that many graduate students and postdocs intend to pursue a career as academic principal investigators (PI), which requires a much broader diversity of skills than the usual, highly specialized science degree. If a young scientist in academia is put into a team where the labor is divided in a way such that each member is a specialist instead of a generalist, then “we are failing to train them for the job they are trying to have,” i.e., as a PI, “[a]nd the jobs we are training them for”–generally research specialists–“are marginalized” in the present structure of academic research, Walsh and Lee explained in an email to Science Careers.
As evidence for their conclusions, Walsh and Lee studied how labor was divided across a wide range of scientific fields and institutions. Using data from the Institute of Scientific Information Web of Science database, the two social scientists took a survey of U.S. scientists who published a research article from 2001 to 2006. Their controls were the project of study, the size of the team, and how labor was divided and organized. 2,327 scientists responded, and half of them (1,223) were from scientists working in a university or hospital, each with at least one other team member. Walsh and Lee then focused their study on those teams and published their results on an online journal called Research Policy last month.
PRELIMINARY CONCLUSIONS BEFORE INTERMISSION
To surmise, the survey found that the average scientific research team had seven members, but the list of co-authors excludes three of its contributors, like postdocs, Ph.D. candidates, undergrads and technicians. “[J]ust under half the members of the teams were non-authors, suggesting that authorship credit, while critical for the reputation system of science, may not be accounting for many members of the team,” Lee and Walsh added in their email to Science Careers.
More on this survey, and on Walsh and Lee’s conclusions about the scientific implications of the social organization of scientific teams later this week.
–Original correspondence and coverage of this survey by Elisabeth Pain of Science Magazine
