The Prevalence of Political Gender Stereotypes An Experimental Research Design Focusing on Flanders

The starting point of many studies on gender and politics has been the observation that men are overrepresented in politics, whereas women constitute only a small minority of elected officials (Ballington, 2005; Murray, 2014). Scientific literature points to two possible explanations for women’s underrepresentation: ‘voter bias’ (whether voters reject women at the ballot box) and ‘systemic bias’ (whether party, and other crucial actors do not grant enough support to female candidates) (Murray, Krook, & Opello, 2012; Verge & Troupel, 2011; Wauters, Weekers, & Maddens, 2010). A major aspect of ‘voter bias’ is constituted by political gender stereotypes. Voters lack resources, time or interest to become informed about all candidates in elections. Therefore, they rely on voting cues to make their choices (McDermott, 2009). Gender could function as such a cue. On the basis of gender, voters ascribe particular personality traits, capacities and opinions to candidates (Huddy & Terkildsen, 1993). These cues are called political gender stereotypes, which have the potential to determine voting behavior (Brown, 1994; Dolan, 2014; Fox & Smith, 1998). Since the existence of political gender stereotypes has been extensively documented in countries using a majoritarian electoral system, most typically the US (Dolan, 2010; Fox & Smith, 1998; Huddy & Terkildsen, 1993; Koch, 2000), we suggest to switch the institutional context by studying whether political gender stereotypes also prevail in a Proportional Representation (PR) system. It is our aim to identify the presence of gender stereotypes in Belgium. To these ends, a quasi-experimental design will be set up. Respondents will be confronted with hypothetical candidates, who present themselves in an audio message in which they mention their position on the lists and their policy position on a particular issue. The experiment will be set up as ‘between subjects’ design, in which different groups are compared with each other (Druckman, Green, Kuklinski, & Lupia, 2011). The candidate’s gender and the list position will be manipulated as between-subject factor (Mutz, 2011; Torgerson & Torgerson, 2008). By adopting this kind of experimental design, we can control for a large number of intervening variables. The methodology used for this project is original. (Quasi-) experimental methods are rather scarce in political science, but its use tend to increase in recent years (e.g. Blais, Lachat, Hino, & Doray-Demers, 2011). Also for the analysis of voter bias, an experimental approach is not common. Moreover, most experimental studies on gender stereotypes ask voters to assess women in general (e.g. Dolan, 2010) or present real-life candidates (e.g. Koch, 2000; Matland, 1994). As such, intervening variables could unintentionally cause bias, which will not be the case in our design with hypothetical candidates. On this conference, we will present our fully-fledged theoretical framework on voters’ opinions about women’s candidates, and we will link these theoretical insights with our experimental design, which comes to the heart of what this panel on experimental research aims to do.