What makes one team smarter than another? | Anita Williams Woolley, Carnegie Mellon University


Thank you. Well, it’s great to be here. I’m Anita Woolley. I’m a professor at
Carnegie Mellon University. And I want to start
out by just giving you a bit of an intuition for
how we got interested in that what I’m going to
tell you about. So I’m going to start off
by giving you two examples. In each pair I describe,
I want you to think about, which team do you think
is going to be successful. So the first example consists
of two men’s ice hockey team from the Olympics. So the first team
is made up of stars from professional leagues,
from all around the world. Their home country is
hosting the Olympics, prominent politicians
in the country have said that the
billions of dollars they spent to prepare
for the Olympics would all be worthwhile if
this team brought home gold. OK? So that’s motivation. Compare this to the second team. This team was actually
convened less than a year before the
Olympics– can we go back one– made up of
amateur and collegiate athletes. And nobody really expected much. They just hoped that they
wouldn’t embarrass the country since they were also hosting
the Olympics that year. So most of us would expect
that the team on the left would be more successful. But those of you
who know hockey know that this is the Russian
men’s team from the 2014 Olympics, who was
eliminated from contention before the medal
rounds even began. And this is the 1980
US men’s hockey team, who brought home the
gold medal despite all that was working against them. All right, so let’s
think of another example. This one from
presidential cabinets. So you might think,
it’s election season, you’re electing a person. Well, in fact, you’re electing
a person and his or her team of advisors. So in this cabinet,
this cabinet was made up of what one historian termed
the best and the brightest, highly accomplished Ivy
League educated individuals, strong interpersonal
relationships, including even some family
members of the president. Compared to this
cabinet, made up of men who lacked some formal
education in some cases, and were bitter rivals
in a hotly contested presidential primary. Most of us, again, would expect
that the team on the left would be the more
successful, but scholars of American history will
know that this is the Kennedy cabinet, which was responsible
for some huge decision making debacles. This is the Lincoln cabinet,
who passed historic legislation despite being in a
deeply divided country. What these examples really
illustrate is that a, we’re really bad
at predicting which teams are going to perform
well in the future, in part because we have a
tendency to focus a lot on individual
attributes and less on how the group
actually works together. So a question that’s
important to ask is, why were these
team successful. And what my colleagues and
I would put forth to you is that one potential answer
is collective intelligence. And so some of the research
I’m going to tell you about will support this
idea and, hopefully, leave you with about two
different ideas for how you can build smarter teams. We have a strong
tendency to focus on hiring smart
individuals, but we don’t know enough about
how to build smart teams. And one of the
reasons why we focus on individual intelligence
is because there’s some very good metrics for it. So where we’re all familiar
with G for general intelligence. This is the idea that
underlies IQ tests and is highly predictive
of how individuals perform in a variety of domains. We started our
research wondering if there is an analogous factor,
c, for collective intelligence. Are there teams that
are consistently good at working together
across many different domains? And can we use that information
to predict which teams will perform well in the future? So we started our
research to explore if collective
intelligence even exists. We had teams come to our lab. They spent many hours
together performing a whole variety of tasks. We found that teams that did
well on one kind of task, let’s say a creativity
task, were also good at mathematical tasks and other
sorts of problem solving tasks. When we calculated a score
based on how they performed on all of these tasks, we
were able to then predict with a pretty high
degree of accuracy how they performed
in the future when we brought them back to perform
another more complex task. And we were able to
do so much better than simply knowing the
individual IQs of the team members themselves. So we’ve replicated this
finding a few different times. And repeatedly find that
collective intelligence is a much better predictor
of how teams perform than individual intelligence,
whether you look at the average intelligence
of team members or even the intelligence of the
smartest person in the room. So then we set about
trying to figure out, well, if it’s not individual
IQ that determines collective
intelligence, what does. And some of what we found
was rather surprising. So one of our first
observations was that the proportion
of women in the team is related to
collective intelligence. And at first we thought it
was a linear relationship, but now that we’ve collected
data on several hundred teams, we find that it’s more of
a curvilinear relationship. So on this graph, this
is average collective intelligence, and
what you’ll notice is that when teams include
less than 50-50 females, they tend to oscillate
around average. But once you have more than
half of the group female is when you see that teams are
consistently above average. However, there’s still
a benefit to diversity. It’s not the case that
all-female teams are always way above average. So one of the reasons,
though, as we dug deeper into this, why we see
this relationship is another trait that we measure,
which is social perceptiveness. So social perceptive
is an ability to pick up on subtle, nonverbal
cues from other people. We give all of the participants
in our studies a test called the reading the
mind in the eyes test. In this test, they see only
the eye region of the face and they have to draw
inferences about what this person is
thinking or feeling based on a list of choices. We find that women score
higher on this task than men. And that teams
that include people with higher scores
on tests like these are more collectively
intelligent. We also measure a number of
attributes of communication in the groups, and have
particularly noticed that the distribution of
communication is important. Specifically, if
you have one or two people who dominate
the conversation, the team is much
less collectively intelligent than if you
have more equal distribution of conversation. We’ve also conducted
these studies with teams working together
online and collaborating by a text chat, and we find
a very similar results. Equality in communication is
still important, even when they’re using text chats. It’s also important
even when you look at who’s contributing
what to their shared products. And similarly, in
these online teams, we surprisingly find that
social perceptiveness is just as important. So reading the mind
in the eyes test is predictive of
collective intelligence even when team members
are not seeing one another’s facial expressions. Research that’s just coming out
from a team in the Netherlands further shows that collective
intelligence is really driven by the lowest
scoring member on tests of things like reading
the mind in the eyes. In other words,
including somebody who has really poor
ability in that domain really seems to drag down a team
that otherwise would be high performing. So with that, I want to leave
you with two ideas that are, you know, based on the
consistent findings of the studies that
I’ve told you about, as well as others
that we’ve conducted. First, is that it’s
really important when you’re convening a team
to set egalitarian norms. Over and over again, we see that
the equality of contribution is important. This really comes from
convening a team in which there are no stars, as well as no
people who are slacking off or loafing. The second piece
is, you really need to pay attention to the
skills, the collaboration abilities of the
people in the team, and specifically avoid
bringing in people who are going to drag the team down. People who are very
negative, who are domineering can exert a disproportionately
negative effect on collective intelligence. You what the people who
are really good, also, but avoiding the people
who really drag things down is equally important. So hopefully, by
paying attention to a few of these attributes, we
can not only hire smart people, but also create smart teams. So that, thank you very much.

Leave a Reply

Your email address will not be published. Required fields are marked *