Can Women’s Representation Reduce the Effect of Stereotype Threat on the Gender Gap in Political Knowledge?
Download data and study materials
Sample size: 1038
Field period: 06/11/2013-04/11/2014
Evidence from social psychology suggests that reminding test-takers of gender differences in performance increases the gender gap in actual performance. The mechanism that explains such effect is called stereotype threat. Meanwhile, recent studies find that stereotype threat also produces gender gaps among surveys respondents that answer political knowledge questions. However, survey research shows overwhelming evidence for the gender gap even when respondents do not receive explicit reminders of gender differences in performance. This study aims to understand why there is such a gap in the first place. The paper uses an experimental design that manipulates the gender composition of batteries of feeling thermometers towards prominent politicians in order to prime stereotype considerations among respondents.
H1: Male respondents should outperform female ones when respondents are shown a battery of feeling thermometers towards 8 male politicians prior to answering the knowledge questions;
H2: Male respondents should not outperform female ones when respondents are shown a battery of feeling thermometers towards 4 male and 4 female politicians prior to answering the knowledge questions.
There are three experimental conditions in the study:
- in the first, respondents are asked to rate their feelings towards 8 male politicians said to be the "most prominent" in the country prior to answering the knowledge questions;
- in the second, respondents are asked to rate their feelings towards 4 male and 4 female politicians said to be the "most prominent" in the country prior to answering the knowledge questions;
- in the third, respondents don't receive the battery of feeling thermometers prior to the knowledge battery.
Conventional political knowledge battery (Q6-Q11).
Results show a gender gap in political knowledge when respondents are asked to rate a list of male politicians, while a much smaller gap is found when respondents rate a list containing both male and female leaders. Moreover, these results are driven by respondents' propensity to guess and/or say "don't know". When respondents see the list of 8 male politicians, men become more likely to guess and women more likely to pick the DK option. These gender differences disappear when respondents see a list with both male and female politicians.
Gendered Political Contexts and the Gender Gap in Political Knowledge