elharo comments on Rationality Quotes January 2015 - Less Wrong

4 Post author: Gondolinian 01 January 2015 02:23AM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Article Navigation

Comments (148)
Tags: quotes

Comments (148)

You are viewing a single comment's thread.

Comment author: elharo 19 January 2015 12:57:08PM 1 point [-]

We next tried to define what characteristics distinguished the smarter teams from the rest, and we were a bit surprised by the answers we got. We gave each volunteer an individual I.Q. test, but teams with higher average I.Q.s didn’t score much higher on our collective intelligence tasks than did teams with lower average I.Q.s. Nor did teams with more extroverted people, or teams whose members reported feeling more motivated to contribute to their group’s success.

Instead, the smartest teams were distinguished by three characteristics.

First, their members contributed more equally to the team’s discussions, rather than letting one or two people dominate the group.

Second, their members scored higher on a test called Reading the Mind in the Eyes, which measures how well people can read complex emotional states from images of faces with only the eyes visible.

Finally, teams with more women outperformed teams with more men. Indeed, it appeared that it was not “diversity” (having equal numbers of men and women) that mattered for a team’s intelligence, but simply having more women. This last effect, however, was partly explained by the fact that women, on average, were better at “mindreading” than men.

--Anita Wooley, Thomas W. Malone. and Christopher Chabris, Why Some Teams Are Smarter Than Others, New York Times, January 16. 2015

Comment author: Nornagest 20 January 2015 12:18:33AM *  2 points [-]

I'm open to the idea that the factors informing group performance might not be identical to those informing individual performance, and it seems plausible that intra-group communication could play a strong role in that, but this is a result suspiciously amenable to the NYT's politics. Probably deserves a grain or two of salt.

Comment author: Manfred 20 January 2015 01:11:38AM *  1 point [-]

You have probably actually heard of this study already - it was in the news briefly when the Science article got published in 2010, this is just a rehash.

Anyhow, to some extent this is factor analysis smoke and mirrors - just because there's this nice factor that correlates well with performance on group tasks doesn't mean that the causal mechanism doesn't go through cognitive skills. This is especially obvious in the case of gender, where it seems implausible that women improve average performance just by exuding some sort of aura. They probably do it by using skills that are distributed differently among genders and weren't captured by the study's emotional-perceptiveness test. So as soon as they include number of women in their c-factor, you know it's correlational and not necessarily telling you useful actions to take (e.g. the intervention "get people to talk more equally" has no guarantee of helping, even though equal time spent talking correlates with success).

But, that said, there is this nice factor that correlates well with performance on group tasks, and if one wants a diagnostic test, and your tasks look like those in the study (e.g. brainstorming, within-group bargaining, playing checkers, designing a building), and your participants are drawn from a population similar to college students, it's more valuable to measure social skills than it is IQ.

EDIT: Nor are they discounting intelligence. From the paper:

If c exists, what causes it? Combining the findings of the two studies, the average intelligence of individual group members was moderately correlated with c (r = 0.15, P = 0.04), and so was the intelligence of the highest-scoring team member (r = 0.19, P = 0.008). However, for both studies, c was still a much better predictor of group performance on the criterion tasks than the average or maximum individual intelligence

The correlation coefficients for intelligence are about half to 2/3 those for the social perceptiveness and turn-taking - and also about half to 2/3 what the correlation with IQ is when doing these tasks alone. this is consistent with the hypothesis that when working in groups, less of the variation depends on IQ and more of the variation between groups is due to different levels of social skills.

Comment author: elharo 20 January 2015 12:37:38PM 1 point [-]

That's pretty much exactly what the article, and the quoted selection, said. The improved performance of teams with more women is attributed to from gender disparity on the test for "Reading the Mind in the Eyes, which measures how well people can read complex emotional states from images of faces with only the eyes visible."

Comment author: Viliam_Bur 22 January 2015 12:32:04PM *  1 point [-]

Possible interpretation:

If a team wants to do something smart together, the team members have to (a) communicate, or at least (b) be really good at guessing what the other team members are thinking.

First, their members contributed more equally to the team’s discussions, rather than letting one or two people dominate the group.

communication

Second, their members scored higher on a test called Reading the Mind in the Eyes, which measures how well people can read complex emotional states from images of faces with only the eyes visible.

mind reading

Powered by Reddit Powered by Reddit