Advertisement
In today's scientific and business organizations, to resolve big questions, people are more likely to build large research teams. But a recent study discovered that smaller teams are more likely to produce overturning and creative content after the analyse for more than 65 million papers, patents and software projects over the past 60 years.
According to the paper published in ‘Nature’, researchers at the University of Chicago show that smaller teams are more likely to introduce fresh ideas into science and technology, while larger teams often consolidate and develop existing knowledge.
While large and small teams are critical to scientific progress, the findings suggest that research policies and the allocation of funds to team projects should be reassessed.
“Large teams are always more conservative. Their papers were like sequels to great work; they are very unsurprising and risk-averse,” says co-author James Evans, professor of sociology and director of the UChicago Knowledge Lab, “Large teams are always looking for the past and depending on the knowledge from yesterday. Small teams try strange, unconventional things -- they go further that it may take others a long time to understand and appreciate their work.”
Advertisement
To finish the study, Evans and colleagues collected 44 million papers from the Web of Science database, with more than 600 million citations; they downloaded 5 million patent documents from the United States Patent and Trademark Office; sixteen million software projects have been collected from Github.
They found that in paper writing, patent invention and software product development, topics grew more monotonous as the number of team members increased. A simple overview of existing knowledge can be more innovative if written by fewer people.
“The nature of small teams and large teams is different,” says co-author Lingfei Wu. “Small teams can always remember forgotten ideas, ask questions and create new areas for research, while large teams chase the hot spots, discard less popular ideas, answer questions and integrate existing knowledge frameworks.”
The analysis shows that both small and large teams play an important role in the research ecosystem, with the former generating novel and promising insights that are rapidly developed and refined by large teams. Some experiments cost is so high (the Large Hadron Collider) that they can only be achieved on a large scale by a single team. But the authors argue that for efficiency reasons, other complex scientific problems might be left to a small, independent, and adventurous team.
“There are a lot of grants going on around the world for larger and larger research teams,” Evans says. “We're suggesting that there should be more diverse science budgets. If you really want to push forward science and technology, you need to act like a venture capitalist rather than a big bank -- you should fund a series of smaller, basically unrelated research projects to increase the likelihood of significant, ground-breaking results.”
He added: “Yes, most of these investments will fail. But the strategy itself is optimal control of the risk of failure, and if you want to discover something, you have to take a bet.”
