A laughing matter? How responses to sexist campaign attacks influence candidate evaluations

Although the number of women in political office is slowly increasing over time, they are still strongly underrepresented in almost all countries around the globe. A wide variety of explanations for this continuing underrepresentation of women in politics are brought to the fore, although still little is known about how the specific campaign dynamics contribute to this situation. Is the use of sexist political attacks against women during election campaigns - and the way female candidates react to sexist attacks targeted at them - important at all in shaping candidate evaluations? And for which types of voters? Rather surprisingly, evidence about how (different) voters evaluate different styles of responses to misogynistic campaign attacks is lacking. This paper fills these gaps by experimentally testing to what extent an (implicit vs explicit) sexist attack of a male candidate towards a female opponent and her reply (in an educational vs argumentative vs empathic vs humorous manner) have an impact on the candidate evaluation of both the attacker and the target politician and to what extent these effects are conditional upon voters’ sexist attitudes. We expect a backlash effect of a campaign attack on the evaluation of the attacker (H1a), as it violates general norms of gender equality and egalitarianism, especially if the sexist attack is explicit (H1b) and for citizens with strong anti-sexist attitudes (H1c). Moreover, we expect that a response to the attack by the target based on a prejudice confrontation style will positively affect the evaluation of the target politician (H2a), especially if the response in humorous or argumentative (H2b), moderated by citizens’ level of sexist beliefs (H2c). To test these expectations, we plan to conduct two between-subjects online survey experiments, one in the USA (MTurk sample, planned data collection: December 2021) and one in Germany (SoSci sample, planned data collection: January 2022). Testing these expectations in two countries that are different in terms of their political culture, political system, level of personalization, level of negative campaigning, and descriptive representation of women in politics, allows for investigating the broad scope of the expected effects - and provides a first indication as to whether the dynamics of sexist attacks and targeted retaliation are also a function of contextual conditions. All participants will read a newspaper article about a debate between two fictitious candidates in which the male politician attacks the female politician. We manipulated the attack to be explicitly sexist (“as a women, she is too emotional and lacks leadership qualities”), implicitly sexist (“she is too kind for this job and does not have what it takes”) or not sexist. Then, the female politician’s response to the attack is manipulated as well. The results of this study will shed more light on how sexist campaign behavior and responses to these attacks help or hinder political candidate and, ultimately, contribute to the underrepresentation of women in politics.