As was previously reviewed, there is a bulk of literature to support the phenomenon that the media coverage of female politicians is different, and typically more sexist, than the coverage of their male counterparts. Two dominant theories, media framing and media bias, are typically used by scholars to classify these two types of media coverage. These theories are not mutually exclusive and are often used in combination with each other in order to relay information to readers and/or voters in ways which can shape their perception of the candidates they read about.
Media Framing. The first variable I operationalize is media framing. There is a multitude of definitions of media framing used by scholars, however, for the purposes of the current study, media framing is defined as journalistic descriptions embedded in news stories in order to create different depictions of news subjects (Devitt 2002). This definition hones in on news subjects specifically which is of particular use to this study given that it focuses exclusively on female political candidates as news subjects (Devitt 2002). It has also been the definition utilized by previous scholars using framing as the mechanism through which to analyze differences in news coverage between male and female political candidates (Devitt 2002).
There are several forms of media framing which have been historically used to cover female politicians differently than male politicians. News coverage of female politicians is more likely to prioritize personal information about the candidate than coverage of their positions on political and policy issues such as information about the candidate’s gender, marital statues, age, appearance, family, etc. (Miller et al. 2010; Braden 1996; Carroll and Schreiber 1997; Kahn 1996; Norris 1997; Woodall and Fridkin 2003). Even when female candidates do receive news coverage of their policy stances, it tends to focus on policy stances that are typically characterized as feminine issues such as healthcare or education over other issue stances (Atkeson and Krebs 2008).
Use of specific biased language is often used to discuss and describe female candidates differently than male candidates. Burke and Mazzarella examine speech verbs according to their emotional content in Canadian election news coverage and find that verbs that contained emotional overtones and that are characterized as “negative and aggressive language” such as “attack,” “boast,” and “complain” were used more often to describe speech acts be female candidates (Burke and Mazzarella 2008). Even the name that is used by the media when covering female candidates is often different. Female candidates are more often referred to by their first name than by their formal titles which has been shown to to detract from their power, legitimacy, authority, and support the overarching stereotype that women are less fit for positions of power such as elected office (Uscinski and Goren 2011; Lawrence and Rose 2010). Finally, I include coverage that explicitly focuses on the gender of the female candidate. This includes coverage which uses what has been called the novelty frame which emphasizes the rarity of female candidates running for office (Lawrence and Rose 2010; Wolbrecht, Beckwith and Baldez 2008). This also includes references to female candidates as an explicitly female or woman candidates which is extremely rare among male candidates (Dolan 2004).
Thus, for this analysis, I operationalize several indicators of media framing in order to evaluate media framing in the French news coverage of female political candidates. These indicators include: personal versus issue coverage, language choice, candidate references, the novelty frame, and explicit mentions of the candidate’s gender.
Media Bias. In addition to media framing, media bias is another technique which can be used in order to shape the type of coverage that a female candidate receives. For the purposes of this study, media bias is described as individual news stories, commentators, or news outlets exhibiting pervasive patterns of sexism or other prejudice (Lawrence and Rose 2010). This definition hones in on news stories specifically which is of particular use to this study given that it focuses on individual news stories as a mechanism of media bias.
One dominant indicator of media bias is the frequency of coverage in which female candidates receives overall less media coverage than male candidates (Dunaway et al. 2013). Common examples of this include focusing on a female candidate’s viability, discussing them as a less competitive candidate than the male candidate, or describing the female candidate’s gender as barrier to electoral victory (Kahn 1994b; Atkeson and Krebs 2008). Within this already limited coverage, news stories that focus on the electoral viability of the female candidate are another example of media bias (Dunaway et al. 2013; Kahn 1994b; Atkeson and Krebs 2008). In some studies, female candidates were described as overall less viable, competitive, and likely to win when compared to their male counterparts running for office (Kahn 1994b; Atkeson and Krebs 2008). Finally, negative media coverage of a female candidate is classified as media bias. This includes what is called a negative gender distinction in which the gender of the female candidate is described as a hindrance to their ability to win and adequately perform in office (Devitt 2002).
Thus, in addition to the media framing indicators, I operationalize the following indicators in order to evaluate media bias in the French news coverage of female political candidates: mentions of a candidate’s electoral viability and negative gender distinctions.