Thomas Malone of the MIT Sloan School discusses the importance of interpersonal skills and his research measuring intelligence of groups.

Read the full transcript below. (Transcript by Realtime Transcription.)

Kirkpatrick:  The next session was not originally scheduled. But Tom Malone, in the opening session, made a comment about the interpersonal skills and how important they’re becoming in business, and then he elaborated on that further yesterday morning in the session on sort of the future of business. So I asked him if he would come up here and explain some of the things he explained yesterday morning, which you will find fairly surprising, regardless of your gender, although gender may have something to do with it.

Tom runs the Center for Collective Intelligence at MIT, and what he’s going to talk about is why interpersonal skills are more important than you think.

So, Tom.

Malone:  Thank you, David.

So, David asked me to say some of the things I said yesterday morning. So those of you who were in the panel session then will hear some of this again.

Here is why I think interpersonal skills are even more important than I used to think. First, I think it’s becoming increasingly important to think about business and organizations not just in terms of how efficient or how productive they are, but also in terms of how intelligent they are.

Now, if you want to create intelligent organizations, one thing that would certainly be helpful would be a way of measuring the intelligence of a group.

So I’m going to tell you today about some research we’ve done at MIT to do exactly that. We used the same statistical techniques that are used to measure individual intelligence, but we applied them to measuring the intelligence of groups.

Now it turns out with individuals, if you give a bunch of people a bunch of different tasks—math, reading, et cetera—and you look at how well they do on that, there’s a single statistical factor that explains—not perfectly, but significantly—how well a given individual will do on a very wide range of different kinds of tasks. One word for that factor is “intelligence,” and it’s basically this factor that individual intelligence tests measure.

Now as far as we can tell, nobody had ever asked the question about whether there was a similar factor for groups.

So we asked that question. And to answer it, we brought a bunch of groups of people, two to five people in each group, into our laboratory, and we had each group do a bunch of very different kinds of tasks. And it turns out when we analyzed all this statistically that the answer is yes, there is a single statistical factor for a group, just as there is for an individual, that predicts how well the group will perform on a very wide range of very different kinds of tasks.

Now that’s interesting, but what I think is even more interesting is what predicts this collective intelligence factor for a group.

Really, before we did this that it might be that groups had a collective intelligence factor, and it was just the average of the individual intelligences of the people in the group. It turns out that the individual intelligence of the group members is significantly correlated with the group’s collective intelligence. But only moderately so. Only moderately correlated.

So what that means, in a sense, is just having a bunch of smart people doesn’t necessarily make a smart group.

Now you all knew that. But here’s a very precise scientific demonstration of that.

The next question is, if it’s not just having smart people that makes a group smart, what is it? We looked at a bunch of factors that we thought might affect this, and we found three that were significantly correlated with the group’s collective intelligence.

The first was the average social perceptiveness of the group members. Now we measured this by giving everybody a test called reading the mind in the eyes, where you get to see pictures of other people’s eyes and try to guess what the person in the picture is feeling. When you’ve got a bunch of people in a group who are good at that, then the group on average is more collectively intelligent.

The next thing we found that was significantly correlated with the group’s collective intelligence was the degree to which the group members participated about equally in the group discussion. If one or two people dominated the group discussion, then on average the group was less collectively intelligent.

The third thing we found that was significantly correlated with the group’s collective intelligence was the percentage of women in the group. More women was correlated with more intelligent groups.

You see I get a reaction when I say that. Now it turns out that this last result was largely explained statistically by the first; that is, it was already known before our work that on average, women score higher than men on this test of social perceptiveness. So one possible interpretation of our results would be that what you really need for a group to be collectively intelligent is to have a bunch of people in the group who are socially perceptive. And it may not matter that much whether the people are men or women. We don’t know for sure whether that’s true. But that’s one possible interpretation.

The other thing that’s interesting about this result is it’s not a standard diversity result. Standard diversity theory would say that the best performance should come from the groups that are about half and half men and women.

Now we didn’t design our study specifically to look at this, but as best we can tell from looking at the preliminary results we got, half and half seems to be one of the worst performing groups, and it seems to be roughly linear. That is, more women seem to be correlated with more intelligent groups, maybe even all the way up to all women.

Now, one last thing. We’ve done some more recent work not yet published where we gave this same test of collective intelligence to two kinds of groups. In one case, the groups work together face to face; they could talk and do everything as in the original groups. In the other case, they were online groups, and they could only communicate with each other by text chat, just by typing.

Now it turned out that the average social perceptiveness of the group members was equally predictive of the collective intelligence of the group in both cases. In other words, when you have a bunch of people in a group who are good at this reading emotion and eyes test, they are better at working together, they are more collectively intelligent even when they can’t see each other’s eyes at all, and can only communicate by typing.

So what we think that means is that even though this test is just measuring reading emotions and eyes, it must be correlated with a much broader range of interpersonal skills, what you might call social intelligence.

So that’s why I think interpersonal skills are even more important than I used to think. Thank you.