Hello. My name is Chris Kluwe, and for eight years I was the punter for the Minnesota Vikings. In May 2013, the Vikings released me from the team. At the time, quite a few people asked me if I thought it was because of my recent activism for same-sex marriage rights, and I was very careful in how I answered the question. My answer, verbatim, was always, “I honestly don’t know, because I’m not in those meetings with the coaches and administrative people.”
This is a true answer. I honestly don’t know if my activism was the reason I got fired.
However, I’m pretty confident it was.
Despite advances in many areas, women are still underrepresented in the upper levels of corporations, electoral politics, and some scientific research fields. This lack of parity is all the more striking because, in much of the developed world, women’s educational achievements have surpassed those of men. (In 2009 in the United States, 57 percent of currently enrolled students were female, a trend that has been fairly stable over several decades.) This imbalance has been ascribed to two factors: continuing gender discrimination, and lower desire for competitiveness among women.
Focusing on the competitiveness aspect, a new study indicates that policy-based initiatives can increase women’s participation and competitiveness in math and the quality of the resulting work. The particular experiment performed by Loukas Balafoutas and Matthias Sutter, released February 2 by Science, involved three methods that provided an initial advantage to women in a math competition. The authors found that, in each case, women entered the competitions more readily, but the aggregate performance of the participants was unaffected, and sometimes even improved.
The common feature in all three methods is an affirmative action approach: the active promotion of the underrepresented group. Passive methods (such as increasing potential rewards for everyone) do improve participation by women, but they also improve men’s performance as well, which leaves the gender gap in place. Affirmative action, on the other hand, not only changes the odds of success by women, but (according to the authors of a related study) also increases their confidence and willingness to compete in the first place.
If the competitors know they have a higher chance of success before the contest even begins, the authors suggest, they are more likely to enter it with a positive view of their own abilities. In addition, excessive confidence in male competitors is reduced because they know rewards will be distributed in a different manner.
Balfoutas and Sutter created math competitions with rewards, drawing participants from an experimental pool of 360undergraduate students broken into groups of 6—three women and three men. The three methods they examined were based on initiatives used in European parliaments or public-sector job searches:
- A quota system was implemented, requiring that a certain proportion of the winners be female;
- Two related subcases: the highest-performing woman was given preferential treatment in one, while in the other, if a tie occurred in the competition, the woman was always selected;
- The competition was repeated until a certain fraction of the winners were women.
Note that none of these constitute “reverse discrimination,” an accusation affirmative action plans often face. In no case was a top-performing man denied a reward if he outperformed everyone else. The main effect the researchers found was an increase in the number of able women willing to participate.
A significant feature of the study is that it addressed concerns that men may not be willing to cooperate with women after the end of the competition. Critics of some programs have suggested that men may be resentful if they feel a woman had an unfair advantage. To examine this possibility, Balfoutas and Sutter paired men and women in a coordination game after the contest. They found performance in this game did not decrease compared to the controls, within the statistical errors.
Not all methods produced equal outcomes: repeating the competition produced the best cooperation after the contest, but it provided the weakest incentive for women to enter. The preferential treatment options had the opposite effect, though again the statistical variations were larger than the differences, so it’s not possible to say definitively that one method is more effective than another.
This study has focused on whether people enter into existing competitions, which is relevant to things such as job openings or political campaigns. It doesn’t address other issues, like the number of available “winners” (number of available jobs, promotion opportunities, and so forth); likewise, the type of reward given is not considered. Nevertheless, Balfoutas and Sutter have effectively provided a better understanding of gender-based affirmative action, opening up the possibility of broader experimentation.