Collective Intelligence

October 23, 2010 6:54 AM

Number of Women in Group Linked to Effectiveness in Solving Difficult Problems
posted by kliuless (54 comments total)

Well, duh. They can actually focus on more than one thing at once.

But you can't have the positive stereotype without the negative one: Number of Women in Group Linked to How Late Group's Meeting Starts.
posted by Mayor Curley at 7:10 AM on October 23, 2010 [2 favorites]

What a ridiculous premise. Believe me, I've worked on some of these problems; I've tried yelling at them, I've tried punching them. I've even poured some beer on them, yelled at them some and then asked them about the game. I headbutted one of them. They just don't get solved.

These are just intractable problems, that's all. I don't see what women have to do with it.
posted by mhoye at 7:11 AM on October 23, 2010 [18 favorites]

Believe me, I've worked on some of these problems; I've tried yelling at them, I've tried punching them. I've even poured some beer on them, yelled at them some and then asked them about the game. I headbutted one of them. They just don't get solved.

Really? Headbutting is the number one problem solver in my office.
posted by indubitable at 7:18 AM on October 23, 2010 [1 favorite]

Ah, more propaganda from the radical left wing Women Are Beneficial to Society agenda.
posted by nomadicink at 7:19 AM on October 23, 2010 [16 favorites]

This explains the almost comical ineffectiveness of the Three Stooges.
posted by PlusDistance at 7:22 AM on October 23, 2010 [37 favorites]

Two things:

1. "So having group members with higher social sensitivity is better regardless of whether they are male or female," Woolley explains.
Aaargh, I hate this sort of move where they extrapolate the salient point into a headline-worthy statement full of gender binary nonsense! Believe me, I know plenty of women with not one lick of social sensitivity, and plenty of men who are amazing at understanding and exploiting group dynamics. My mother and father, one of each, just as some examples. I would like to take this moment to flail my arms about in a feminist rage.

2. That insufferable stock photo. Why do they have pens? There is no visible paper! They are using the pens to point at a laptop, but you can't write on a laptop! Do the pens make it businessy-er? What if they need to make a note, are they going to write it down on their hands? Stock photographers, please stop doing this!

Okay okay got that out of my system.
posted by Mizu at 7:22 AM on October 23, 2010 [25 favorites]

It's true, look at how well those nuns solved the Maria problem...
posted by Huck500 at 7:22 AM on October 23, 2010 [20 favorites]

If you want to look at the actual study (and have access to the freakishly expensive Science online), here's a citation:

Woolley, Anita et al. "Evidence for a Collective Intelligence Factor in the Performance of Human Groups." Science Xpress 30 Sep. 2010 DOI: 10.1126/science.1193147

also, the abstract:

Psychologists have repeatedly shown that a single statistical factor—often called "general intelligence"—emerges from the correlations among people's performance on a wide variety of cognitive tasks. But no one has systematically examined whether a similar kind of "collective intelligence" exists for groups of people. 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.

posted by GenjiandProust at 7:24 AM on October 23, 2010 [6 favorites]

Abstract of Science article, "Evidence for a Collective Intelligence Factor in the Performance of Human Groups," pdf, podcst of interview with the lead author.
posted by nangar at 7:29 AM on October 23, 2010 [2 favorites]

> I would like to take this moment to flail my arms about in a feminist rage.

Thank you for so accurately describing my reaction to most gendered stories in the mainstream media.
posted by djfiander at 7:29 AM on October 23, 2010 [4 favorites]

Related Stereotyping: Number of Women in Group Linked to Complete Batshit Distractedness of Men in Group
posted by pwally at 7:33 AM on October 23, 2010

Dave, you old bastard, what picture are we going to use for this article?

Pretty standard stock photo, John. Hot business chick and an old man at a laptop, maybe an ethnic guy in the back.

Sounds fine to me. Did Alice sign off on it?

She's stuck in traffic.

All right, go with it.
posted by condour75 at 7:34 AM on October 23, 2010 [10 favorites]

I'm confused. The conclusion arrived at in the headline is not supported by any numbers I can see in the linked article. Maybe I need my girlfriend to explain it me.
posted by PareidoliaticBoy at 7:45 AM on October 23, 2010 [2 favorites]

A quick look at the study itself reveals it wasn't double blind so the other headline would be:

"Number of scientists in group not linked to proper experimental design."
posted by storybored at 7:45 AM on October 23, 2010 [10 favorites]

What I want to know is if the researchers and reporters would have been just as willing to announce these results if they had been favorable to men or unfavorable to women. Based on everything I've observed about how the media reports on gender differences*, I doubt it.

*Their rule seems to be: interpret everything to flatter women whenever possible.
posted by John Cohen at 7:46 AM on October 23, 2010 [2 favorites]

I'm recommending the podcast now that I've listened to it. It goes into more detail (and doesn't require a subscription).

Note: The study found little or no correlation between group performance and the IQ's of individual group members. However, there is little correlation between IQ and performance on problem solving tasks, even on individual tasks, so the lack of correlation may simply be because the task the groups were asked to perform required different skills and abilities than those measured by IQ tests.
posted by nangar at 8:13 AM on October 23, 2010

storybored, what would double-blindness consist of in this type of observational study? It's not like you can do a randomized controlled trial, there's no treatment effect to measure.
posted by demiurge at 8:15 AM on October 23, 2010 [1 favorite]

What I want to know is if the researchers and reporters would have been just as willing to announce these results if they had been favorable to men or unfavorable to women. Based on everything I've observed about how the media reports on gender differences*, I doubt it.

Without having read the article (as it's behind a paywall) - it seems like they are looking at a mix of personality traits, some of which are stereotypically linked to female gender. But presumably a group composed just of this trait might also be less productive. So the headline could just as well read Number of men in the group, etc., etc.

Well actually I wish the headline would read something else altogether. This is awful, awful science reporting.
posted by carter at 8:23 AM on October 23, 2010 [1 favorite]

Mizu: "
2. That insufferable stock photo. Why do they have pens? There is no visible paper! They are using the pens to point at a laptop, but you can't write on a laptop! "

I pretty much can't point to stuff on my monitor to my boss without a "talking pencil". Spreadsheets, CAD, drafting letters...it's actually pretty useful.
posted by notsnot at 8:30 AM on October 23, 2010

Headbutting is the number one problem solver in my office.

We use thumbwrestling, and, more recently, pop & lock battles.
posted by elizardbits at 8:30 AM on October 23, 2010 [5 favorites]

What was the methodology? How did they control for the possibility that, while all of the socially aware people were off doing fluffy happy social stuff, one person was grinding out the work that the whole group would take credit for?
posted by MrVisible at 8:40 AM on October 23, 2010 [3 favorites]

carter, I think while the statement "Number of Men in Group Linked to Effectiveness in Solving Difficult Problems" would be valid as well as "Number of Women in Group Linked to Effectiveness in Solving Difficult Problems" on the face of it, but most people think "linked" in these kind of statements implies a positive correlation. If that's true, then saying "Men" here would be misleading.

Finally, c was positively and significantly correlated with the proportion of females in the group (r = .23, P = .007). However, this result appears to be largely mediated by social sensitivity (Sobel z = 1.93, P = .03), since (consistent with previous research) women in our sample scored better on the social sensitivity measure than men; t(441) = 3.42, P = .001. In a regression analysis with the groups for which all three variables (social sensitivity, speaking turn variance, and percent female) were available, all had similar predictive power for c, though only social sensitivity reached statistical significance (β = .33, P = .05) (12).

I'm not a statistician, but what this means to me is that their "social sensitivity" trait affects c, and females were more likely than males to have a high rating for that trait. Even then, the effect isn't too high (.23 Pearson correlation) even though it has a low p-value. Of course, maybe that's an excellent correlation value for these type of studies, with all the confounding factors.

However, I agree with you on the science reporting. If I read this paper before I saw the news article, I could have guessed what the press release would lead with.
posted by demiurge at 8:44 AM on October 23, 2010

A quick look at the study itself reveals it wasn't double blind ...

How would conduct this study as double-blind experiment? What would the treatment be? What would you use as a control? Would you experimentally give some members higher intelligence, social sensitivity or other personality traits to see how this affected group performance? Would you use 'no personality' as a placebo?

This was essentially a survey design, not an experiment, (though not a pencil-and-paper survey). The researchers tested participants on intelligence and a number of personality traits, collected demographic data and information about the groups' patterns of interactions, and looked at how those factors correlated with group performance. That doesn't make the study invalid or improper.
posted by nangar at 8:45 AM on October 23, 2010 [10 favorites]

TEAM ANGRY MANBOY NERD FOR THE WIN!

Wait, did we win?
posted by Artw at 8:58 AM on October 23, 2010 [1 favorite]

So, social sensitivity is the predictor, and women (in this sample) have more of it on average. That's kind of neat, that the social sensitivity was the most correlated factor. (Though all the recent meta-science has made me kinda jumpy about assuming anything based on p values…)

Also, John Cohen, do you have to whine about men being hard done by every time something about women comes up?
posted by klangklangston at 9:03 AM on October 23, 2010 [14 favorites]

It's interesting that gender affects the ability to work in groups, but I'm more interested in a negative result they found:

This "c factor" is not strongly correlated with the average or maximum individual intelligence of group members

Why on earth not? We know that individual performance on tasks like visual puzzles is strongly predicted by g, the general intelligence factor, so if that doesn't show up in group tests, being in a group must somehow suppress individual performance. This raises a few questions: what results would we see comparing group performance to individual performance on the same task? Does being in a group suppress the contribution of g in favour of that of c, or is it that low c suppresses g? What effects does designating a team leader have? I think this one is important because evolutionary psychology predicts a greater positive effect the more men are in the team.
posted by topynate at 9:04 AM on October 23, 2010

Based on my observations both as a grad student and as a high school teacher, group work seems to produce higher-level thinking than does individual work, in most cases. I would have no idea how to design a study to prove this. And I hardly think we need one. (Gender is a minor issue here, although it's red meat to headline writers. Oh, and how do Muslims do in group work?)
posted by kozad at 9:23 AM on October 23, 2010

I wonder how much of this "social sensitivity" is socially trained. If you're not the dominant group, and used to having people talk over you, or ignore you, or you need a man to say the same damn thing you just said for it to be acknowledged - you develop survival tactics which require more attention to the group dynamics.

Would this apply also to queer folks or people of color? Or is it primarily gender based?

And what would the differences be between groups placed by the researchers vs. self-selected groups? Groups that self-select in ways closer to actual demographic odds, would they be formed with less bias to begin with?
posted by yeloson at 9:29 AM on October 23, 2010 [3 favorites]

But you can't have the positive stereotype without the negative one: Number of Women in Group Linked to How Late Group's Meeting Starts.
posted by Mayor Curley at 3:10 PM on October 23

I've always found it far more usual that it's the men in the group - especially senior men - who like to turn up late to meetings. Seems to be a dick-swinging thing - "Oh, look at me. Not only am I incredibly busy with other stuff more important than this meeting, I don't care about making you peons wait for me, ha ha." Revolting fuckers.

I used to walk out of any meeting in which we were still missing someone ten minutes after start time. God, I do not miss office life one little bit.
posted by Decani at 9:30 AM on October 23, 2010 [3 favorites]

> What I want to know is if the researchers and reporters would have been just as willing to announce these results if they had been favorable to men or unfavorable to women.

The researchers, sure. Gender differences weren't something they were looking for for, though they did collect demographic information, and having more women in a group turned out not to be a significant factor when they partialed out social sensitivity, as measured by test of reading facial expressions. That women (or American women anyway) tend to be somewhat better at this is not a new finding.

Woolley pointed out in the interview that having more women in a group don't help unless they are also good at social sensitivity, and having men with this trait in the group does help. (In other words, women tend to be better at this, but some men are good at this, and some women suck at it.)

However, I doubt if the media would have picked this up if there wasn't a gender politics angle or something else controversial-sounding they could play with.
posted by nangar at 9:30 AM on October 23, 2010 [1 favorite]

I've always found it far more usual that it's the men in the group - especially senior men - who like to turn up late to meetings.

I don't notice this as much, but then I work in academia, where we are more likely to have women in senior positions. Our Provost (male) is notoriously late, usually being 1--15 minutes late for nearly everything, but this year I am on a lot of committees with him, so I see how a meeting runs over, then he is late for the next, which runs over, and so on. It may be poor time management, but I don't think it's necessarily arrogance.

Also, since I am a lot more involved in faculty governance this year, I have discovered that no one schedules meetings with transit times (which we do with courses), so if I have a 10-noon meeting followed by a noon-2 meeting, I have to a) leave early, b) arrive late, or c) figure out how to cross campus instantaneously. If i work out c), I am going to be rich enough that I may never attend meetings again.
posted by GenjiandProust at 9:44 AM on October 23, 2010

TEAM ANGRY MANBOY NERD FOR THE WIN!

Wait, did we win?

Do we ever?
posted by mhoye at 10:03 AM on October 23, 2010

Also, since I am a lot more involved in faculty governance this year, I have discovered that no one schedules meetings with transit times

That, in a nut-shell, explains why academics should never be put in charge of anything.
posted by PareidoliaticBoy at 10:27 AM on October 23, 2010

Pretty standard stock photo, John. Hot business chick and an old man at a laptop, maybe an ethnic guy in the back.

I like that they invite us to "enlarge" this critical element that is key to understanding the piece.
posted by nev at 10:49 AM on October 23, 2010

Number of Women in Group Linked to Effectiveness in Solving Difficult Problems unless said problem is a spider.

No, seriously though, I read an article not too long ago about how studies that find fewer/no gender differences are significantly less likely to be published, and are often actively discouraged from publication. I thought I'd saved it, but I didn't. Anyone know what I'm talking about?
posted by honeydew at 11:21 AM on October 23, 2010 [1 favorite]

http://www.sciencedaily.com/images/2010/09/100930143339-large.jpg

<insert caption here>
posted by kuatto at 11:23 AM on October 23, 2010

Number of Women in Group Linked to Effectiveness in Solving Difficult Problems unless said problem is a spider.

Spiders aren't difficult. They're very well-mannered creatures.

I was pulling an all-nighter for an exam, and had a can of Diet Coke to hand. While I was studying, I would reach for it without looking and take a sip.

You see where this is going.

IN THE FUCKING CAN, PEOPLE. IN THE CAN! I ALMOST DRANK A SPIDER. I JUST DON'T LIKE THEM, OKAY?! IT'S NOTHING PERSONAL!

[NOT SPIDERIST]
posted by tzikeh at 11:52 AM on October 23, 2010 [5 favorites]

These things are interesting but also dangerous. Assuming that the optimal gender composition of a group doesn't equal the proportion seen in society I think that smaller-minded people and lawyers could easily use it as grounds for justifying discrimination.
posted by vapidave at 1:01 PM on October 23, 2010 [1 favorite]

Did anyone notice this little nugget farther down the page:
The Fancier The Cortex, The Smarter The Brain? (July 19, 2009) — Why are some people smarter than others? A new article describes how certain aspects of brain structure and function help determine how easily we learn new things, and how learning capacity ... > read more
Is MeFi's cortex fancy enough?
posted by Cranberry at 1:07 PM on October 23, 2010

insert caption here

"See, look at that, the laser is burning right through the LCD screen but the beam remains invisible!"

"And yet it tastes like sour apple. Amazing!"

"I'm so bored I've blurred myself."
posted by griphus at 1:52 PM on October 23, 2010 [1 favorite]

What was the methodology? How did they control for the possibility that, while all of the socially aware people were off doing fluffy happy social stuff, one person was grinding out the work that the whole group would take credit for?

I had the same thought. I guess I'm still bitter from group work in high school.

But here they found that taking equal turns in the conversation was positively correlated with collective intelligence. If a few people were talking way more or way less than the rest of the group, that predicted the group would perform badly.

Of course, that doesn't totally rule out your scenario. Maybe the conversation in the high-performing groups went like this:

Alice: YAY TELEVISION WOOO!
Bob: YAY FOOTBALL WOO!
Carol: gritting teeth Okay guys I'm carrying the two now....
Alice: Thanks, Carol! Okay, my turn.... WOO FOOTBALL YAY!
etc.

Somehow I doubt it, though.
posted by nebulawindphone at 1:56 PM on October 23, 2010

How did they control for the possibility that, while all of the socially aware people were off doing fluffy happy social stuff, one person was grinding out the work that the whole group would take credit for?

That person is usually my girlfriend.
posted by clarknova at 2:16 PM on October 23, 2010

Number of Women in Group Linked to Effectiveness in Solving Difficult Problems unless said problem is a spider.

My friend, you have never seen my husband when he was afraid a spider might touch him. It's like Barishnykov having a seizure. I don't like them, but don't actually think their touch will kill me.
posted by emjaybee at 3:03 PM on October 23, 2010 [2 favorites]

I pretty much can't point to stuff on my monitor to my boss without a "talking pencil"

At least you point at your own monitor. I had a boss who liked to point at my monitor like that chump in the stock photo. Stupid bastard *underlined* something on my monitor with a Sharpie, laughed it off with an "Oops" and then refused to sign off on a replacement when I couldn't get the mark off the screen. I quit shortly thereafter.
posted by jamaro at 4:31 PM on October 23, 2010 [1 favorite]

Three statisticians go out hunting. They come across a large buck. The first one fires and misses by a foot to the left. The second fires and misses by a foot to the right. The third on whoops and yells "We got it!!"
posted by empatterson at 7:18 PM on October 23, 2010 [2 favorites]

That's because performance on cognitive tasks is related to arousal, and women are hott.
posted by cogneuro at 7:59 PM on October 23, 2010

No, seriously though, I read an article not too long ago about how studies that find fewer/no gender differences are significantly less likely to be published, and are often actively discouraged from publication.

This is true throughout science. Publishing negative results doesn't happen very often. Historically, I suppose there were reasons for this. As I'm banging my head against a problem, I may try 10 different approaches before finding one that works well. If each of those failures was a paper, it would have made journal subscriptions rather unwieldy and much less useful.

Now that everything is online, a handful of scientists are starting to stand up and say "hey, we should be announcing our failures as well, so others aren't doomed to make the same mistakes". In my opinion, these people have an excellent point.

So there are two major ways that this can come about. The first is to be encourage more openness when publishing papers. In the methods, or at least the supplement, authors should include a pretty solid description of what techniques turned out not to be helpful and why they failed.

This isn't common practice now, mostly because of reputation and ego. Journal articles are always written as though the experiments were a nice, linear process. We did A, then B, then got result C. This isn't a very accurate description of the process, and everyone knows it, but it makes everyone involved look smart. If you're clawing your way towards tenure or angling to land a good post-doc position, you don't necessarily want to broadcast your failures. Probably more importantly, it makes for a nice, easy to communicate story.

The second way to better announce our failures is to practice open-notebook science, where raw data are published to the web in real time (or after a short delay). What's holding this back is that scientists worry that by revealing too much, their competitors will get a leg up and publish the next big paper before they can. In this era of crushingly low paylines, where less than 20% of NIH grant applications get funded, things have gotten pretty cutthroat. Stories of being "scooped" abound, and although some people feel that these claims are exaggerated, it certainly does happen, sometimes with career-tanking results.

So to make a long story short, no, negative results aren't often published, even though doing so would probably be a boon to scientific enterprise as a whole. The good news is that there's a pretty strong movement underway that is slowly making science more open, transparent, and reproducible.
posted by chrisamiller at 10:02 PM on October 23, 2010 [4 favorites]

As far as I can tell, this study might have been reported as "Groups With Longer Average Hair Length Exhibit Greater Effectiveness in Solving Difficult Problems." Or "Neckties Linked to Decreased Group Effectiveness in Solving Difficult Problems."

Having one or more women who don't display social sensitivity in a group will presumably not increase that group's effectiveness, while having one or more men who do, will. A group that includes a woman who dominates the discussion and decision-making will presumably do worse than a group of all men in which the conversational turns are more evenly distributed.

And "Groups Whose Members Watch More 'Gilmore Girls' More Effective At Solving Difficult Problems" is probably misleading.
posted by taz at 10:32 PM on October 23, 2010 [2 favorites]

Wait a minute: working in a group is in itself a problem/task, isn't it? If you're working on a problem in a group, you first have to be able to work as part of a group, and then you have to try to solve the problem. And working in the group might have challenges associated with it which are nothing to do with the problem at hand, additional challenges which have to be overcome before any work can be done on the problem.

If that is the case, obviously the problems are going to be solved better by groups whose members are more socially sensitive, and therefore better at working in groups. Perhaps the groups who perform badly on the tasks do so because the issues associated with working in a group prevent them from tackling the task.

There are definitely mental challenges associated in co-operative working. What would be interesting is if the study showed that collectively people can complete tasks which no single member would have been able to complete by themselves. What this study actually seems to be looking at is how well people work in groups, which makes the result (people who work better in groups work better in groups) rather unsurprising.
posted by rubber duck at 3:28 AM on October 24, 2010 [2 favorites]

or you could go down the 'race' route or, more generally, just identifying the genetic markers for 'intelligence'...
posted by kliuless at 7:21 AM on October 24, 2010

I wonder if the study used random participants or if they had pre-screened them to remove bad or worthless players, for which I have noticed gender correlations along these lines.

MEN:
1) "my way is the only right way"

sometimes in the enhanced variant

2) "what the others are saying is nonsense. Here is my approach"

which winds up being essentially the same as what the others were saying expressed in slightly different words. Basically this sort of guy cannot understand anything unless he came up with it himself.

WOMEN:
1) "I understand X"

which may mean she knows some of the terminology used with X and can sometimes string the words into coherent sentences but has no grasp of the underlying meaning and doesn't understand that other people do. When such a person says

2) "I agree / I understand"

this often becomes a signal that she doesn't and might proceed as if the opposite was true, thus causing an extra step of finding out what she thinks she is agreeing to.

I have found these two types of players very frustrating. The gender correlations come from my experience though I have not seen any studies along these lines.
posted by MonkeySaltedNuts at 12:47 PM on October 24, 2010 [1 favorite]

The supplement (PDF) is available for free online and describes all of the group tasks that they measured.
posted by MonkeySaltedNuts at 3:21 PM on October 24, 2010 [1 favorite]

It's true, look at how well those nuns solved the Maria problem...

With the fucking Anschluss--not the best solution to a problem I've seen.
posted by MikeKD at 9:26 PM on October 24, 2010 [1 favorite]

With the fucking Anschluss--not the best solution to a problem I've seen.

So nuns ordered the occupation of Austria... it explains so much.

Yes, I had to look it up, and my comment goes out of its way to imply that I knew what the Anschluss was. Wheeee!
posted by Huck500 at 9:47 PM on October 31, 2010