In 2012, President Barack Obama won reelection, maintaining the streak of 44 consecutive men to serve in that office. At no point in America’s history has a major political party nominated a woman to be its presidential candidate. In contemporary elections two women, Elizabeth Dole in 2000, and Hillary Clinton in 2008, were formidable contenders in primary races, but neither woman was able to clutch their party’s nomination. So few are competitive female runs at the White House that we can count them on one hand. “The Office of The President” has established itself, with no variation, as a man’s domain.

In 2016, the presidential race is likely to feature a female candidate (if Hillary Clinton ultimately decides to run), and an end to this streak is possible. Yet, given the difficulties women face when running for most elected offices, including the challenges of gendered media coverage discussed below, her road to the White House may be a long and arduous one. One possible means of overcoming negative media coverage would be for Hillary to campaign not as a woman, per se, but as an experienced, gender-neutral candidate, given what we know from past research that assesses how media treat female candidates.


In political science, there is an active and rich research agenda interested in the underrepresentation of women in politics. One vein of such scholarship explores the degree to which media coverage perpetuates an association between women and femininity, and whether negative attitudes about women’s potential in politics is the result. Numerous political scientists have examined the degree to which the media perpetuates the association of women candidates with femininity by analyzing media coverage of candidates during elections.

Man Enough?: Gender, Media and the American Presidency Brown Bag

Meredith Conroy will give a talk entitled, “Man Enough?: Gender, Media and the American Presidency’” at noon, Friday, April 10 in the Bishop Barnwell Room of the Student Union Building at Boise State. The free talk is part of the Politics and Policy Brown Bag Series. See TBR Events for future Brown Bag talks.

Even though many women who run for office deliberately avoid using feminine terms to describe themselves and shy away from emphasizing their families (which might remind voters of their familial obligations), the media consistently covers women in feminine terms, focuses on their appearance over more substantive policy issues and uses masculine metaphors (i.e. comparing the election to a boxing match), thereby maintaining the presumption that men are more appropriate as politicians than women. Overall, there appears to be a positive bias for male candidates in media coverage of elections.

However, I suggest the bias may not necessarily be for male candidates, but instead for masculine candidates, and thus, in races where two men are running it is possible that the candidate perceived to be less masculine is subject to the same media bias observed in general for female candidates. Indeed, not all men who run for office comply with our stereotypical notions of manliness and masculinity, embodied by that independent, singularly focused, win-at-all-cost, mentality. What is more, regardless of an individual’s actual personal characteristics, the media may still deem particular male candidates as more feminine than their male counterparts.

Similarly, female candidates perceived as more masculine may in fact receive less biased media coverage. The basis of this argument stems from notions of leadership in the United States, which largely embrace masculinity and reject femininity, as well as the Western cultural construction of gender, which treats masculinity and femininity as defined in opposition to each other, and thus largely incapable of coexisting within the same individual.

In American politics, femininity is often considered synonymous with weakness, and antithetical to leadership. Typically, masculine traits are preferred in public officials, as is expertise on more masculine issues, such as national security. Particularly in the context of the White House, feminine traits and feminine issues are largely deemed as less relevant. See Jennifer Lawless, “Women, War, and Winning Elections,” 2004 and  Rossenwasser and Dean’s, “Gender Role and Political Office,” 1989. Yet feminine characteristics are not bereft of leadership potential. Empathy, deliberation, cooperation and understanding are characteristics that are valuable leadership traits. Yet, due to our cultural expectations of leadership as masculine, to express feminine traits is to be a weak leader. The focus of my research is the extent that the media plays a role in maintaining this gender hierarchy, where masculinity is the norm and preference in American politics, and in particular in the presidency.


My expectation for this study was that the media largely use feminine traits as a means of discrediting male presidential candidates;masculine traits are a more likely source of praise. “Gender conflict framing,” coverage of the candidates in a news article — often focused on a candidate’s personal character — is gendered in nature to the extent that one candidate is framed as more masculine and the other is framed as more feminine. This gendered distinction between candidates is consistent enough to project to readers (and voters) that one candidate is the more masculine choice, whereas the other candidate is the more feminine choice.

In order to assess the degree to which the media frame male presidential candidates in gendered terms, I studied presidential election print media coverage for the 2000, 2004, 2008 and 2012 elections, and focused on gendered language used in news reports, namely the traits and characteristics used by print media to describe the major parties’ nominees. For each of the election years, I read a sample of articles that focused on the candidates’ personal characters, published between September 1 and Election Day, and recorded traits that reporters used to describe the candidates. I analyzed a total of 300 character focused articles, largely from The New York Times and USA Today, with attention paid to described traits, behaviors and personal styles on which journalists chose to focus. I use this data to assess several different components of gender discourse in presidential elections.

From the sample of character focused articles, for all election years under analysis, 1,545 traits were recorded. Of these traits, most were noted as neutral (56 percent); 30 percent were noted to be masculine traits, and 14 percent were noted as feminine traits. Positive masculine traits used to describe the candidates included resolute, decisive and courageous; negative feminine traits included weak, indecisive and erratic; gender neutral traits included liar, intelligent and relaxed.

Given the masculine nature of politics, and the fact that all candidates in the elections under analysis were male, it is unsurprising that there are more masculine than feminine traits being used in the discussions of the candidates’ characters. Masculine language was even invoked adversely, such as when a candidate was referred to as “not strong enough” or “not tough.” References such as these to the candidates’ characters  explicitly acknowledge the candidates’ lack of masculinity, while not overtly feminizing them. Yet, descriptions framed in this manner reiterate the importance of masculinity in our political leaders, and the understanding that those unable to live up to the masculine standard need not apply.

It would be very unlikely, for example, to see the news media refer to a candidate in terms of kindness or warmth because there is not an expectation for American presidents to be kind or warm. In other words, it is acceptable for a president to lack positive feminine qualities, but it is unacceptable for a president to lack positive masculine qualities, especially those related to strength. When feminine traits are used to describe candidates, it is those traits that are not socially desirable; feminine traits are rarely used to describe candidates in positive terms.

To appreciate more fully the degree to which feminine and masculine traits are negative or positive, I randomly selected an appropriate number of feminine and masculine traits from the full population of gendered traits to be analyzed for tone. For the feminine traits sample, 69 percent of the traits used were negative in connotation; 31 percent were positive. For the masculine traits sample, 67 percent of the traits used were positive in connotation; 31 percent were negative and 2 percent were neutral.

I also use the data to assess the existence of gender conflict framing. Media coverage of presidential elections — regardless of its utility as a journalistic practice — tends to be reductionist, characterizing candidates as occupying opposing sides of a two-sided conflict. The prominent dichotomy of gender as masculine versus feminine lends itself to the negative and combative nature of elections, and journalistic practices and routines of political reporting. Thus, the value of gender conflict to the media is accentuated by the provocation of disagreement in political news reporting. I expected that in the context of each article studied, the candidates’ portrayed gender depictions would be dissimilar; furthermore, the expectation was that dissimilarities would be relative, as a means of portraying conflict. To test the theory of gender conflict framing I used paired sample t-tests as a statistical method.

A paired sample t-test is used when there is an expected relationship between dependent variables (in this case the candidates’ portrayed gender scores), within the same unit (in this case within a character focus article). The expectation here is that the within-article gender score for Candidate 1 and Candidate 2 will be highly dissimilar, reflecting norms of conflict in media coverage, and the dissimilarity will be in terms of the candidates’ described characters, with respect to gender. Thus, for each article, a paired sample t-test assesses the differences in the gender scores for Candidate 1 and Candidate 2.

As displayed below, I find media to invoke gender conflict framing in their within-article coverage of the presidential candidates for the election years under analysis. In other words, within the context of individual news articles, the media was likely to discuss the candidates in opposition to one another, in terms of the gendered media descriptions of their personal characteristics and behaviors. The difference in mean gender score values in the final column represent the P-values, which suggest that difference in gender scores for the candidates within each article are indeed statistically distinct.

Election Year Low and High Gender Score Means Within Article Difference in Mean Gender Scores (P)
Low High
2000 -.07 .32 .001
2004 -.11 .38 .001
2008 -.03 .45 .001
2012 -.23 .45 .006
All Elections -.10 . 40 .001

My analysis found that gender is relevant in presidential elections, even though up to this time a female candidate has yet to be nominated by one of the major political parties. Gender makes itself relevant through the language the media uses to describe candidates, as a means of cryptic, disguised or even explicit evaluation of the candidates, and as an established means of manufactured conflict and disagreement that the  media consistently invokes. Yet, I suggest that the negative consequences of the media’s use of gendered language in presidential elections extends far beyond the candidate whose image is feminized. The consequences influence our general perceptions of women, their capacities for leadership and their representation in our government.

Share:Tweet about this on TwitterShare on FacebookShare on LinkedInShare on Google+Share on TumblrBuffer this pageEmail this to someonePrint this page

The views and opinions expressed here are those of the writer and do not necessarily reflect those of Boise State University, the Center for Idaho History and Politics, or the School of Public Service.