Redeeming Gender Studies

The myth of the glass ceiling continues to be debunked. Challenging the presumption that women are underrepresented in high-status jobs because of discrimination, economists at Stanford University and the University of Pittsburgh offer empirical evidence for the more plausible explanation that the two sexes respond differently to competitive environments.

Muriel Niederle and Lise Versterlund conducted an experiment at the Pittsburgh Experimental Economics Laboratory, where 40 men and 40 women were divided into 20 groups (each containing two men and two women seated in rows), to investigate gender differences in the preferences for competition and how those differences, if any, account for economic outcomes. They asked participants to take three simple, five-minute math tests, one of which would be chosen as the basis for compensation at the end of the experiment. If the first test were chosen, participants would be paid 50 cents for each correct answer; if the second were chosen, only the individual in each group with the most correct answers would be paid $2 for every correct answer, while the others would receive nothing. Before taking the third test, participants were asked to select the payment scheme (piece-rate or tournament) under which they would prefer to perform.

While gender differences were nonexistent in all three test performances, twice as many men (73 percent) as women (35 percent) selected the tournament scheme in Test 3, a “robust” difference that remained when the researchers compared the choices of men and women of equal performance ranges. In fact, they also discovered that low-ability men chose to enter the tournament “too often” while high-ability women entered it “too rarely.”

To explain the gap, the researchers asked participants to guess how their performance in Test 2 ranked relative to the others in their group; participants would be paid $1 if their guess were correct. The men were “substantially more” overconfident than the women: 75 percent of them thought they were the best in the group; only 43 percent of the women thought that. Then attempting to control for the overconfidence factor, participants were given the option to participate in a competition where payment would depend upon the number of correct answers in Test 1 and the participant would select the compensation scheme he or she wished to be applied to past performance. Absent the thrill or fear of performing in a competition, the men still preferred the tournament compensation scheme more than the women, confirming to the economists that men are simply more competitive creatures.

Confident that their findings explain why men succeed more in business and the professions, Niederle and Versterlund call upon their fellow economists to give greater attention to the study of sex differences, seeming to lament that gender studies “have been mostly left in the hands of psychologists and sociologists.” Natural family buffs should welcome that.

(Muriel Niederle and Lise Versterlund, “Do Women Shy Away from Competition? Do Men Compete Too Much? The Quarterly Journal of Economics 122 [August 2007]: 1067–1102.)