Collective intelligence factor

Summary: for best results from creative group work, DO NOT compose the group randomly because it will negatively affect participants' performance. Unless, maybe, they happen to be socially sensitive to each other.
---

Having finished teaching my class at Stanford CSP, I'm back at blogging! Today's topic is a research paper in Science about Collective Intelligence. An excerpt from its abstract:

In two studies with 699 individuals, working in groups of two to five, we find converging evidence of a general collective intelligence factor that explains a group's performance on a wide variety of tasks. This "c factor" is not strongly correlated with the average or maximum individual intelligence of group members but is correlated with the average social sensitivity of group members, the equality in distribution of conversational turn-taking, and the proportion of females in the group.

This is consistent with my experience moderating multiple brainstorms and invention sessions. The important lessons from the paper are:
a) random groups perform worse than individuals;
b) the group's social dynamics is more important than participants' intelligence;
c) having women in the group helps the dynamics.

Also note the type of tasks the researchers used for measuring performance:

Tasks included solving visual puzzles, brainstorming, making collective moral judgments, and negotiating over limited resources.

From what I understand reading the paper's references, another task was:

to work on a creative, open-ended task together with their team members during a one-hour laboratory session. Specifically, they were asked to use a set of building blocks to build a house, garage, and swimming pool, which were scored according to a set of complex scoring criteria (see Appendix A). The scoring of the task was intentionally complex and devised to force trade-offs.

Working on open-ended problems is better than solving puzzles, but the approach still follows the good old path of standard engineering and management training practices where people take trade-offs for granted. As we know from the history of innovation, best solutions emerge when problem solvers break through trade-offs, which would impossible to re-create and measure in this type of experiments.

tags: trade-off, psychology, brainstorming, social, research