Friday, May 27, 2005 11:37 AM | rahel luethy | 5 comment(s)
via cafe au lait, i've read this article which disusses that the gender gap at work isn't due to women's insecurities about their abilities but due to different appetites for competition. here's a short summary of the experiment underlying the claim:
women and men were paid to add up five numbers in their heads. in a first round, they worked individually, getting 50 cents for each correct sum. in a second round, they competed in groups of four, with only the winner getting $2.
on average, women made as much as men (with either system).
for the third round, they were free in their choice of system: either take the piece rate or compete in a tournament. the result was that even the best women declined to compete, while almost all men chose the tournament (even the ones who had done worst in previous rounds).
seeing this as a "byproduct of evolution and testosterone" is a rather redunctionist view. however, it actually matches my personal work experience and my own behavior rather well...
3:31 AM | Etienne Studer said...
I love competition!
--Male
9:51 AM | said...
It's very difficult to make a statement on those findings without knowing the frequency of correct answers. If when working alone the frequency of payout is low then it may work out better in the long run to compete. The correct solution according to Nash could be to compete or to work individually.
On the same lines, working together probably requires less effort which implies a payout for no work for some of the team, which you couldn't get if you were working on your own.
Essentially you could just as well conclude (albeit requiring a few more numbers) that women are better at game theory than men (or vice versa).
While I'm not disagreeing with the conclusion, I find statements like this baseless.
The men's eagerness partly stemmed from overconfidence, because on average men rated their ability more highly than the women rated theirs.
That could be true but is probably rubbish. And incidentally, I'm better than you...
10:50 AM | rahel luethy said...
On the same lines, working together probably requires less effort which implies a payout for no work for some of the team, which you couldn't get if you were working on your own.
I don't think that they actually had to work together. As I understood it, the setup in the second round was identical to the first round from an individual's point of view, except that the payoff was determined by comparing the score to the 3 competitors, with only the winner getting 2 bucks.
11:03 AM | nj said...
Ah, you're right - I misread. I do still think my comment about game theory stands though; it's wrong to come to a conclusion of a game without at least touching on game theory.
10:58 PM | said...
Interesting.
It seems like it might be possible (and correct) to set up the experiment so that game theory would not tell you to prefer competition over individual work. Are you saying they did not do this? (As I am too lazy to read the article...)
Plus, what does this have to do with testosterone? Unless you're using "testosterone" as a shorthand for "biological differences between men and women"?
Over here I noted down some findings to the contrary, e.g. studies in which the boys and men who engage most in violent antisocial activity are found to have relatively low testosterone among the men in their group.
