The science of teamwork

When PhD candidates at my university were recently asked if they were part of a research team, 6% of them answered “I don’t know.”

But if they don’t know what they have, at least they know what they want. Our eager, young researchers want more collaboration, they want to be part of teams. They know isolation is an ineffective way to do research, and they’re looking for ways to work together.

While researchers are good at thinking about their research problems, not enough of us have thought explicitly about what it means to work in teams. I suspect that many of the remaining 94% in the survey, if they were give the option, would have chosen the answer “it depends on what you mean by team.”

Why is it a good idea to work in teams? What kinds of problems are teams well-suited to solve? What does collaboration in research actually mean? What are the pitfalls? How can we avoid them?

Some of these questions have recently been addressed by Ron Ashkenas, who claims that compliance and cooperation often parade as collaboration. Research teams seem particularly vulnerable to these impostors, and when university or faculty leadership moves enthusiastically towards an organization of researchers into teams, we must be aware of these pitfalls.

When a group of people is given responsibility for achieving a particular goal, compliance refers to a situation in which each member individually takes responsibility for achieving the goal.

Great research teams, for example, are often pushed forward by ambitious graduate students. A group of researchers may decide that they need to increase the size or quality of the applicant pool to their PhD program. Compliance refers to a situation in which each researcher aspires to recruit a few more applicants. The goal may be formulated by the group, but the strategy involves no collaboration. The team does not ask which members have previously recruited the best students, it does not ask how the group can work together to grow the pool.

Teams in which members cooperate with one another also maintain individual responsibility, although they work explicitly to inform other teams members of their activities. Research teams in non-experimental fields are easily snared into thinking of cooperation as collaboration. The team identifies an over-arching problem and the members make their individual projects fit under that umbrella, telling about their work, but not actually working on the projects together.

Collaboration is hard. And we should only pursue it if the prospects are great. If you are leading a research team, do you see the potential? Can you push beyond compliance and cooperation to true collaboration?

Ashkenas studied some groups that tried but failed. Team members, he observes, did not consciously decide against collaboration.

They did what came naturally, which is to work either completely or partially on their own. … [T]rue collaboration … requires subordinating individual goals to collective achievement; it means engaging in tough, emotional give-and-take discussions with colleagues about strategies and ideas; and it often leads to working in new ways that may not be comfortable or easy. So given these difficulties, most teams find it easier to talk about collaboration rather than do it.

The challenges Ashkenas identifies should improve research. Research becomes better through challenging give-and-take, through new approaches, through the uncertainties that appear in teamwork. The challenge of leadership in research organizations is to create these conditions. The why is clear.

Organizing researchers in teams and pursuing the benefits of cooperation has to be done in a way that is clear and explicit. So clear, in fact, that young researchers can have no doubt about what they’re doing and who they are working with; when they are part of a team, they better know it! If you are leading researchers, how will you make this happen?