Beaman, Lori, Raghabendra Chattopadhyay, Esther Duflo, Rohini Pande, Petia Topalova. 2009. “Powerful Women: Does Exposure Reduce Bias?” The Quarterly Journal of Economics, 124(4): 1497-1540
In this paper, the authors examine whether exposure to female leaders influences public opinion about female leaders, such as taste preferences for male leaders and perceptions about effectiveness of female leaders. To address this question, the authors study village councils in West Bengal India, where one third seats in the Panchayat level were reserved for women. However, reserved seats are rotated across Panchayats and some Panchayats have been reserved once, twice or never. In this study, the authors compare villager attitudes towards hypothetical and actual women leaders across councils (Panchayats) which have been reserved once, twice or never. The authors hypothesize that taste preferences are more deeply held while perceptions about effectiveness may change with more exposure to female leadership.
Hypotheses:
Taste discrimination, conceptualized as higher preference for male leaders remains unchanged with political reservations.
Statistical discrimination, measured as the difference in the evaluation of male and female leader effectiveness will reduce with political reservations.
Data and Methods:
The data for this study come from a survey of 495 villages across 165 Panchayats from a random sample of 5 households per village. The survey was complemented by Implicit Association Tests (IAT) to measure implicit feelings towards male and female leaders. In this, villagers were asked to categorize images of male and female leaders with normative categories of good and bad. To measure gender-occupation stereotypes, villagers were asked to categorize male and female names with leadership and domestic tasks. This study was methodologically innovative as it sought to use IATs in a low income setting. As an extension of the survey, the authors administered speech and vignette experiments where villagers were asked to listen to a short tape-recorded speech adapted from an actual village meeting, where a male or a female leader responds to a villager’s complaint. Each respondent also heard a vignette where a leader makes a decision in a situation of resource scarcity.
Measures of Voter taste
The survey had explicit taste measures with questions that asked villagers to rate their feelings towards male and female leaders. There were two IATs. One was a categorization task where villagers were asked to match male and female names with good and bad attributes. The second IAT asked villagers to match attributes with male and female politicians.
Measures of Leadership Stereotypes
To examine changes in gender stereotyping of occupations, the IAT task evaluated whether villagers exposed to reservation were less likely to associate women with domestic tasks and men with leadership tasks.
Measures of Leadership effectiveness
This was measured with the help of the vignette task and some survey questions. The leader’s gender was varied randomly for the vignette. Thus, the authors could examine whether the same action was evaluated differently depending on the leader’s gender. Further survey questions asked villagers to rate their actual leader’s effectiveness.
Findings
The findings are examined in categories of never reserved, once reserved and twice reserved Panchayats to explore whether exposure to women leaders influence villager’s perceptions and biases.
Male villagers rank male leaders higher than female leaders. For female villagers, the difference in rating is smaller but male leaders are ranked higher. This remains the same even after exposure to women leaders.
Both men and women show same gender preference for positive attributes. Male villagers associate positive attributes with men and female villagers attribute positive attributes with women and neither is affected by reservations.
In terms of gender occupation stereotypes, male villagers in never reserved Panchayats were faster in associating women with domestic activities than with leadership activities. Male villager exposure to female leaders reduced the association of a female leader with domestic activities. Thus, the authors conclude that exposure to female leaders lowers the stereotype linking.
Male villagers in never reserved Panchayats rate the hypothetical female leaders to be less effective than the hypothetical male leader. Exposure to female leaders reverses this bias. However evaluations of female villagers do not change with exposure to female leaders. The authors suggest that this is because female villagers have less exposure to local politics.
When villagers were asked to evaluate their actual leaders, female leaders in first reserved GPs scored less than leaders in never reserved GPs in their ability to respond to village needs, general effectiveness and in preparing BPL beneficiaries. However villagers’ ratings of leader effectiveness in twice reserved GPs were statistically indistinguishable from never reserved GPs, where the Sarpanch is usually, a man. This shows that exposure to female leaders influences how villagers evaluate female leader effectiveness.
The authors also find that male leaders do not outperform female leaders, and in fact seem to suggest that female leaders perform better. The authors find that the average number of repairs in or constructions in reserved GPs were higher than in unreserved GPs. Moreover, villagers were seen to have paid lesser bribe in reserved GPs.
Thus, villagers perceive first time female leaders more negatively even though they seem to be performing equal to or better than male leaders. This bias is not seen in twice reserved GPs. Women in first reserved GPs seem to report their gender as a difficult in performing their role. The authors also find that female leaders invest in women preferred goods.
In conclusion, the authors find that deep rooted biases and social norms are difficult to change but perceptions about female leadership may improve as a result of affirmative action and mandated exposure to female leadership. Exposure to women leaders reduces stereotypes regarding gender roles and changes perceptions about effectiveness of female leaders among male villagers. Moreover, in the recent elections, compared to never reserved Panchayats, almost twice the number of women contested and won seats in unreserved Panchayats, which had reserved seats for women earlier. Thus, the authors argue that affirmative action help remove barriers for women’s political participation.