Small research teams produce ‘riskier, more disruptive science’

Smaller research teams are more likely to develop newer, riskier ideas, but their value is at risk of being ignored by funders, according to the authors of a new study.

As contemporary science asks bigger and more complicated questions, the prevalence of smaller working groups and solitary researchers has dwindled since the mid-20th century in favour of larger, multidisciplinary teams – but both small and larger teams “are essential to a flourishing ecology of science”, a group of US researchers write.

To investigate the effects of team size on scientific output, the group analysed a dataset of more than 65 million papers, patents and software products contained in depositories including the Web of Science, over a 60-year period.

The findings, published in Nature, suggest that investigations undertaken by small groups of collaborators produce more disruptive science by “developing ideas from older and less popular work”.

Conversely, larger teams were found to be more risk-averse on account of their projects being less flexible, requiring more money, resources and ultimately “success” in matching their findings to a funded proposal.

Speaking to Times Higher Education, James Evans, professor of sociology at the University of Chicago and co-author of the report, said that the benefit of “small team nimbleness…holds across all times, places, domains [and] topics”.

The biggest surprise, according to Professor Evans, was that, despite the evidence pointing towards the value of smaller teams, those that were well-funded tended to be similarly “conservative [and] developmental” in their work to those of larger teams.

This suggested that “when these teams are funded, they are likely funded through a conservative process that does not value what small teams offer: risk”.

For each dataset, researchers assessed the degree to which each work disrupted its scientific field, for instance by introducing something new. Nobel-prizewinning papers, for example, were found to be of the top 2 per cent most disruptive.

They found that solo authors were just as likely to produce highly cited papers as teams with five members. But solo-authored papers were 72 per cent more likely to be highlighted as “disruptive”.

As teams grew from one to 50 team members, the measured disruption of their papers dropped by 70 percentage points, the study says. Their patents also became significantly less disruptive.

“It was striking to us that the effect is so strong that a researcher deciding whether or not to add one more to the team will typically see a difference in disruptiveness,” said Professor Evans, suggesting a direct correlation between group number and disruptive effect “that decreases for every additional team member. Too many collaborators will, on average, change both the search process and the disruptiveness of a work.”

“I believe that there has been an unequivocal push and support for large-scale teams in science and technology over the last half-century, which…has likely been driven by funders and management, but also by teams that are trying to hedge their bets,” Professor Evans concluded.

“These systems need to understand the value of risk that small teams bring…the real message is to managers and funders who, if they want to flee diminishing marginal returns and support breakthroughs, will need to fund science like venture capital, with a much higher tolerance for failures.”

rachael.pells@timeshighereducation.com