Gender novelty and personalized news coverage in Australia and Canada

Abstract

Are female government leaders more likely than their male counterparts to see their gendered identities and personal lives profiled in news coverage of their ascents? Are non-novel women leaders—those who are the second in their jurisdiction to achieve the top political job—less likely to experience media personalization than did the women who preceded them in office? By analyzing newspaper coverage of 20 Australian and Canadian premiers, ten women and their immediate male predecessors, our study establishes that female premiers were more extensively personalized in news coverage than were male premiers, particularly in the Australian context. However, gender novelty and other factors proved significant. The proposition that an increased presence of women in leadership roles diminishes the salience of private lives and personal characteristics is supported by our study, suggesting that gender stereotyping of female political leaders will decrease over time as more women exercise political power.

Keywords

Women leadership gender stereotypes media personalization premiers Australia Canada

Introduction

When a woman becomes the first to attain a political leadership role, her exceptionalism has immense news value. She is different because she is a woman in a position traditionally held by men. Journalists invariably direct attention to the intimate and ‘private’ dimensions of her life, highlighting her trailblazer status with a first-woman frame. What happens when more than one woman has broken the glass ceiling? Do gendered identity characteristics and family responsibilities lose newsworthiness when the presence of a woman in the top job is no longer atypical? Our study explores this possibility by measuring the frequency and intensity of news personalization, defined as attention to politicians’ personal characteristics and family lives, in newspaper coverage of 20 Australian and Canadian subnational government leaders (10 women and 10 men), called premiers. This sample is unique in featuring four jurisdictions—two Australian states and two Canadian provinces—in which a second woman rose to the premiership. It allows us to identify gender differences in coverage of newly elected leaders, and to determine whether gender novelty shapes news-mediated attention to the personal dimensions of leaders’ lives. We do so by comparing levels of personalization for first and subsequent women premiers. Given the continued scarcity of female leaders of national governments, our study offers a rare opportunity to assess differences in media representations of political leaders based on gender incongruity.

The purpose of this study is to measure the symbolic impact of women’s representation in powerful political roles. If media attention to the personal lives and attributes of female political leaders does in fact decrease when the press becomes habituated to the sight of a woman in power, a significant barrier to women’s progress in political leadership roles is ameliorated. Since most of what citizens know about politicians is communicated by media organizations, the presence of gender stereotypes in news reports can be damaging to women’s political ambitions. As a form of gendered mediation, personalization draws attention to the ways in which women’s bodies and personal lives depart from those of the prototypical government leader. Emphasizing the gender identities, looks and family responsibilities of women suggests they do not belong in political leadership roles. When the news makes it personal for female government leaders, they may be evaluated and treated poorly (Curtin, 2015; Steele et al., 2002; Trimble et al., 2013; Verge and Pastor, 2018).

To determine the extensiveness of personalization in news coverage of Australian and Canadian premiers, we analyzed the first seven days of reporting by national and regional newspapers after each of the 20 premiers was confirmed as their jurisdiction’s new government leader. News coverage during this time period offers powerful first impressions, articulating norms of effective leadership and reinforcing leadership prototypes (Harp et al., 2010; Kahn and Goldenberg, 1991). Content analysis measured both the presence and intensity of references to nine features of personalized coverage: gender identity, racial identity, sexual orientation, sexualization, appearance, age, marital status/relationships, children or childlessness, and upbringing. These measures were analyzed individually and combined into an index to answer two interrelated research questions. Are women leaders more likely than their male counterparts to be personalized in reporting about their rise to power? Does gender novelty influence levels of personalization in news coverage? Our findings show that an increased presence of women in leadership roles diminishes the news salience of private lives and personal characteristics. Although levels of personalization were higher for women than for men, particularly in the Australian context, the differences disappeared when controlling for gender novelty. While women who are the first in their jurisdiction to attain the role experience higher levels of personalization than do men, subsequent women premiers are no more likely than men to be described with references to their personas or personal lives. Our study demonstrates the power of symbolic representation to increasingly normalize women’s presence in powerful political positions.

Media personalization and gender stereotyping

The phenomenon of personalization is succinctly characterized as the heightened interest in the ‘human side’ of political actors (Campus, 2010: 221). Considered a defining feature of contemporary political communication, personalization is manifested in the media strategies of politicians, political parties and governments, and it helps shape citizen perceptions of politics (Holtz-Bacha et al., 2014: 154; McAllister, 2013). Our study focuses on media personalization, which has three distinct, but often inter-related, dimensions. The first, individualization, refers to media interest in, and the resultant news prominence of, individual political actors (Garzia, 2011; Karnoven, 2009: 4). The second is a shift in journalists’ emphasis from ‘hard’ professional characteristics like experience and communication skills to descriptions of ‘soft’ or ‘non-political’ personality traits such as likeability and humility (Holtz-Bacha et al., 2014: 154; van Aelst et al., 2012: 213–214). Both of these features prompt journalists to write about the personal lives and personas of politicians, a trend which constitutes the third dimension of personalization.

Often called privatization, this dimension is investigated in the present study because it reveals and reinforces socially constructed gender stereotypes. By ‘intimizing’ political actors, news reporting brings familial relationships and romantic lives into public view (Hirdman et al., 2005). While it helps journalists tell a compelling story, one that humanizes politicians (Langer, 2010: 10), personalization situates bodies and private lives as relevant to evaluations of politicians’ leadership styles, electoral viability and political effectiveness. In short, the phenomenon politicizes gender identities and roles (Langer, 2010: 61; Sreberny-Mohammadi and Ross, 1996: 103). Personalization presents a barrier to women’s political achievement because the characteristics, attributes and tasks considered personal for women are seen as antithetical to professional and political responsibilities. Locating women in the private realm links femaleness with domesticity (van Zoonen, 1994). As a result, women experience role incongruity: the stereotype-based lack of fit between their culturally inscribed personal characteristics and gender roles and those assumed to be necessary for political leadership (Koenig et al., 2011).

Media personalization imprints gender role incongruity through the use of words that explicitly signify the gender of the individual, such as man or woman, father or mother (Trimble, 2017: 64). These labels present men as ordinary inhabitants of the political role (‘the man who. . .’) while foregrounding women’s unusualness (‘first woman prime minister’) (Falk, 2010; Trimble, 2017). Emphasis on physical attributes such as age, physique, wardrobe choices and sexual attractiveness similarly draws attention to the unexpectedness of women’s bodies in the political sphere (Baxter, 2018; Trimble, 2017; Trimble et al., 2015), as does attention to socially prescribed familial roles such as wife and mother (Jenkins 2006; Thomas and Bittner, 2017; Trimble, 2017). Sexual and racial identities are also marked as novel—hence deviant—because of the rarity of LGBTQ and racialized politicians (Everitt and Camp, 2009; Tolley, 2016). News about female leaders thereby articulates their presumed lack of congruence with elite leadership positions (Curtin, 2015; Falk, 2010; Gidengil and Everitt, 1999; Lawrence and Rose, 2010). As Baxter’s analysis of reporting in UK newspapers about female business, political and mass media leaders found, news stories employ stereotypes to underscore women’s unsuitability for leadership (2018: 23). However, gendered mediation does not always replicate or reinforce gender norms and stereotypes. Trimble (2017: 11) argues that ‘because gender dualisms are artificial and unstable social constructs, they can be exposed as such by news mediation of political life.’ Gender neutral or counter-stereotypical portrayals help affirm the leadership aspirations and competencies of women (Simon and Hoyt, 2012).

Some of the personalizing references in news reports are self-representations, reflecting the internalization of gender norms and politicians’ willingness to discuss their identities and family lives as a way of softening their political personas (Langer, 2010). Yet, regardless of the origin of the ‘privatizing’ message, it is presented to the public and consumed as such, with different consequences for women and men. While men who display their wives and children are applauded for being good family men, women are asked who is taking care of their families while they devote themselves to public life (Thomas and Bittner, 2017: 4). Since media personalization articulates gender dualisms and traditional gender roles, observing the phenomenon measures how potentially de-legitimizing forms of gender stereotyping are communicated by the media to the public.

As yet there are few large-scale comparative or longitudinal studies of the privatizing dimensions of personalization. That research findings are mixed reflects the importance of gender novelty. Gender stereotypes are activated when women are in the numerical minority, or where their emergence is regarded as unusual (Hoyt and Murphy, 2016). If an individual’s gender is regarded as ‘relatively salient, the incongruence between the leadership role and the person occupying it is brought to attention and is more easily encoded’ (Vial et al., 2016: 408). As more women hold elected political positions, the observed incongruity between the role and the gender of the actor decreases (Hoyt and Murphy, 2016: 394). For instance, an analysis of more than 10,000 news articles about the women and men who contested 2010 and 2014 mid-term races for the US House of Representatives found very few references to gender and appearance overall, and stories about female candidates were no more likely to focus on their gender identities, looks or family roles than was the case for male candidates (Hayes and Lawless, 2016: 8, 66). Gender is rarely newsworthy in the context of mid-term House elections because of the declining novelty of female politicians in that legislative context and the lack of emphasis on gender in their campaigns (Hayes and Lawless, 2016: 7).

When women are poised to break a glass ceiling, gender is likely to come to the fore in news stories, heightening the salience of personal details. Such is the case for women who compete for, and especially those who attain, political leadership roles. Research on news reporting about eight female political firsts in Spain found that 72% of the newspaper stories employed a novelty frame, foregrounding these women’s gendered unusualness (Verge and Pastor, 2018: 34). Falk’s comparison of news reporting about nine women who sought the US presidency between 1972 and 2008 and their closest male equivalents revealed that the gender identities, looks, and family lives of the women were more frequently profiled that those of the men (2010: 184–185). A large-scale longitudinal analysis of reporting about Canadian national party leadership candidates from 1975 to 2012 established that women who contested these positions were more likely than comparable men to be described with personalizing references to their appearance, sexuality and marital relationships (Trimble et al., 2013).

In contrast, other comparisons reveal modest differences in media treatment of high-profile women and men on key personalization variables (see Lawrence and Rose, 2010). Trimble (2017) examined newspaper reporting of Australian, Canadian and New Zealand women prime ministers and their primary male competitors during general election campaigns, finding that journalists wrote about the women’s looks with much greater intensity than was the case for men (Trimble, 2017: 124). But, with a few exceptions, male party leaders were as or more likely than the women to be described with overt gender identifiers (Trimble, 2017: 72–73), and more stories about the men discussed their marriages and children than was the case for the female prime ministers (Trimble, 2017: 104–106). Relevant to the present study is the case of Helen Clark, who was the second woman to become New Zealand’s prime minister and held the position for nine years. Over time, journalists wrote considerably less frequently about Clark’s gender identity, physical attributes and familial relationships, signaling the conventionalization of women in the role (Trimble, 2017).

While no previous studies explore personalized reporting about Canadian premiers, qualitative studies of the first two women premiers in Australia confirm the presence of personalizing tropes. Journalists detailed Joan Kirner and Carmen Lawrence’s gender and described their hairdos, wardrobes and family roles (Van Acker, 1999: 148). Kirner was also ‘subjected to cruel commentaries and personal attacks’ (Curtin, 2006; Van Acker, 1999: 152). An analysis of newspaper reporting about 12 Australian women who were the first to become members of parliament, cabinet ministers, or heads of state and territorial governments, found that personal attributes were highlighted, thereby communicating these women’s lack of fit with leadership roles (Jenkins, 1999: 79).

Because comparative studies of news personalization are rare, our research addresses a major gap in the literature by observing the phenomenon in newspaper coverage of a significant number of high-profile women leaders in Australia and Canada. Measuring both the presence and intensity of personalized coverage allows us to make systematic comparisons by country and gender of the leader. Because the two countries share common features and six ‘first’ women premiers and four who were the second in their state or province to hold the job, we are able to examine the effects of gender novelty on media representations of premiers’ bodies and intimate lives.

Cases and methods

Australian and Canadian premiers offer an ideal series of case comparisons as these countries share important institutional and structural features, notably Westminster-style parliaments and federal systems of government (Chappell, 2002; Sawer et al., 2006; see also Elazar, 1997). Because power is divided between national and subnational units, women have multiple venues in which to gain executive-level political experience and to compete for party leadership positions (O’Neil and Stewart, 2009; Stockemer and Tremblay, 2015). Greater opportunities have facilitated a relatively high level of success for women. Australia and Canada combined have produced 26 female heads of government. Each country has, to date, featured one female prime minister and 12 female premiers, including leaders of territorial governments. We focus our analysis on premiers of Australian states and Canadian provinces because of their power and prominence. These constitutionally autonomous subnational units exercise considerable governmental authority and hold jurisdiction over matters of social and economic importance (Thomas, 2018: 2). As of March 2019, the two nations offer 15 women premiers at the state and provincial level, seven in Australia and eight in Canada.

To measure levels of media personalization, we examine news depictions of the five most recently elected women premiers in each country and the men who immediately preceded them in office, for a total of 20 cases.¹ This sample was chosen for three reasons. First, combining data from the two countries increases the sample size of women premiers who attained power in the same decade (2007–2017). Second, it enables a rigorous test of gender and novelty in shaping coverage as, in each country, the group of five most recently elected women premiers includes three who were the first in their jurisdiction to gain the premiership and two who were the second to attain the position. Third, each woman premier’s immediate male predecessor was chosen as a comparator because analyzing personalization of women and men premiers helps to identify the circumstances under which gender identity shapes news reporting about political leaders.

We examined newspaper coverage for the first seven days after a person was chosen to serve as premier because introducing new leaders to citizens is central to the journalist’s brief during this phase. The first week of reporting captures the critical first impressions conveyed by descriptions and evaluations of government leaders. These initial assessments can shape opinion formation by audiences (Harp et al., 2010: 292–293) and they tend to be re-circulated in media coverage throughout a leader’s political career (Trimble, 2017: 34). Selecting a standard time period was methodologically necessary because of variation in how quickly premiers are sworn in, which ranged from a few hours to a few weeks. In cases where the new premier’s party was in government, data collection began as soon as the premier was confirmed as the victor in the party’s leadership contest. In instances of a change of government, we began when the premier’s party was declared the election winner.

To assess how premiers were described to both national and local audiences, news stories were drawn from two sources: the top national newspaper and a high-circulation regional newspaper based in each state’s or province’s largest media market. The national newspapers are The Australian and The Globe and Mail, while the regional newspapers are Brisbane’s Courier Mail, Hobart’s Mercury and the Sydney Morning Herald in Australia, and the Edmonton Journal, Montreal Gazette, Toronto Star and Vancouver Sun in Canada. All stories featuring substantive discussion of the premiers—those mentioning them three or more times—were included in the analysis, for a total of 802 news items, including regular news items, columns and editorials. Since the aim is to understand media personalization, any items not written by journalists, such as letters to the editor, were excluded.

Content analysis was employed to systematically record the message characteristics of news items. We built upon the work of Trimble et al. (2013: 469), who drew from both political communication and gendered mediation scholarship to identify seven indicators of personalized coverage: ‘gender identity, age, physical appearance, sexual identity and sexualization, upbringing, marital situation, and children (or childlessness).’ Our study further develops the operationalization of media personalization by looking for overt mentions of racial identity, and by distinguishing between sexual orientation and sexualization. We measured the presence and intensity of references to nine discrete elements of personalization: (1) gender identity (e.g. ‘first female premier’ or ‘nice guy’); (2) racial identity (labelled white or black or indigenous); (3) sexual identity (labelled LBGTQ or heterosexual); (4) sexualization (intimations of sexual attractiveness, such as ‘hottie’; or flirtatiousness, as indicated by ‘a love affair with voters’); (5) physical appearance (descriptions of hair, clothing, jewelry or body size); (6) age (called young, old or exact age specified); (7) marital status (e.g. ‘his wife,’ or ‘she is twice divorced’); (8) offspring (discussion of the presence or absence of children and grandchildren); and (9) upbringing (e.g. ‘lessons learned on the knee of her father’). Coders read each story carefully to identify explicit mentions of each type of personalization.² Intercoder reliability tests on a random sample of stories found robust levels of agreement among the coders, with the Cohen’s kappa on the personalization variables ranging from 0.719 to 1.000 (Freelon, 2010). As a reliability check on the data, coders reviewed all mentions for each variable and removed any that did not receive unanimous agreement.

This study offers an additional methodological innovation by measuring levels of personalization in two distinct ways: presence in the news story and the intensity with which each form of personalization is employed. Presence of personalization was calculated by determining whether or not at least one mention of the indicator was present anywhere in the story. Intensity was measured by counting the overall number of mentions of each form of personalization, with the exception of the upbringing variable. While it was possible to establish how many times a premier was labeled a woman or man, discussions of a premier’s childhood could not be easily disaggregated into discrete words or phrases. Measuring both presence and intensity gauged the extent to which news reports reflected gender stereotypes by, for instance, making only passing references to the offspring of men premiers while offering more fulsome descriptions of women’s parenting roles.

Research on gendered mediation indicates that levels of personalization will be higher for women than for men, especially in reporting about women who have the news value of novelty, as is the case for six of the ten women premiers in our sample. These observations led us to pose the following hypotheses:

Hypothesis 1: Higher levels of personalization will be evident in news coverage of women premiers than in reporting about their male predecessors.

Hypothesis 2: Women who were the first to attain the role of premier are more likely to experience higher levels of personalization than those who were the second to gain the premiership in a jurisdiction.

Although the limited comparative literature has found a consistent focus on women politicians’ gender identity, bodies and, in some cases, family lives (Raicheva-Stover and Ibroscheva, 2014; Trimble, 2017), the prevalence of single-country case studies in gendered mediation research makes it difficult to develop hypotheses regarding the behavior of Australian and Canadian news organizations. We treat this aspect of the study as exploratory, aiming to identify differences and similarities in how the news media in these countries personalize newly elected premiers.

Results

Gender differences in personalization of Australian and Canadian premiers

To detect variation in personalization indicators, and to explore similarities and differences between Australia and Canada, we conducted bivariate comparisons by premier gender, both overall and by country. The absence of overt mentions of racial identity for any of the premiers, all of whom are white, led us to drop this variable from the statistical analysis. This finding illustrates that whiteness is normalized in media accounts of political leadership.³ Table 1 features the presence results: the number and percentage of stories with at least one mention of each of the remaining eight forms of personalization. Table 2 reports the intensity measures—the average number of mentions per story—on seven elements (all but upbringing).

Table 1.

Presence of personalization variables (N/% of news stories) by gender identity and country of premier.

Personalization variables	Australia			Canada			Both countries
	Women	Men	Overall	Women	Men	Overall	Women	Men	Overall
Gender identity	61 (37.7)	26 (13.7)***	87 (24.7)	69 (31.5)	51 (22.1)*	120 (26.7)	130 (34.1)	77 (18.3)***	207 (25.8)
Sexual identity	0 (0)	0 (0)	0 (0)	11 (5)	0 (0)**	11 (2.4)	11 (2.9)	0 (0)***	11 (1.4)
Sexualization	10 (6.2)	2 (1.1)**	12 (3.4)	4 (1.8)	3 (1.3)	7 (1.6)	14 (3.7)	5 (1.2)*	19 (2.4)
Looks	15 (9.3)	21 (11.1)	36 (10.2)	27 (12.3)	28 (12.1)	55 (12.2)	42 (11)	49 (11.6)	91 (11.3)
Age	22 (13.6)	32 (16.8)	54 (15.3)	24 (11.0)	20 (8.7)	44 (9.8)	46 (12.1)	52 (12.4)	98 (12.2)
Marital status	29 (17.9)	18 (9.5)*	47 (13.4)	19 (8.7)	28 (12.1)	47 (10.4)	48 (12.6)	46 (10.9)	94 (11.7)
Offspring	28 (17.3)	8 (4.2)***	36 (10.2)	26 (11.9)	19 (8.2)	45 (10)	54 (14.2)	27 (6.4)***	81 (10.1)
Upbringing	24 (14.8)	12 (6.3)**	36 (10.2)	7 (3.2)	9 (3.9)	16 (3.6)	31 (8.1)	21 (5.0)	52 (6.5)
N of stories	162	190	352	219	231	450	381	421	802

Note: Pearson chi-square used to assess statistically significant differences between women and men premiers in each country and overall. * p < 0.05; ** p < 0.01; *** p < 0.001. N represents the number of stories with at least one mention of the relevant variable.

Table 2.

Intensity of personalization (average number of mentions per news story) by gender identity and country of premier.

Personalization variables	Australia		Canada		Both countries
	WomenM (S.D.)	MenM (S.D.)	WomenM (S.D.)	MenM (S.D.)	WomenM (S.D.)	MenM (S.D.)
Gender identity	1.20 (1.991)	.27 (.952)***	.85 (1.962)	.49 (1.367)*	1.00 (1.979)	.39 (1.201)***
Sexual identity	.00 (.000)	.00 (.000)	.11 (.614)	.00 (.000)**	.07 (.468)	.00 (.000)**
Sexualization	.11 (.568)	.01 (.102)*	.07 (.657)	.02 (.218)	.09 (.620)	.02 (.175)*
Age	.29 (.995)	.31 (.849)	.15 (.456)	.11 (.386)	.21 (.737)	.20 (.645)
Looks	.20 (.779)	.28 (.994)	.21 (.792)	.16 (.468)	.21 (.786)	.21 (.754)
Marital status	.75 (2.235)	.22 (.825)**	.28 (1.462)	.37 (1.454)	.48 (1.843)	.30 (1.212)
Offspring	.64 (2.122)	.11 (.599)**	.44 (1.743)	.23 (1.122)	.52 (1.913)	.18 (.925)**
N of stories	162	190	219	231	381	421

Note: Independent single sample t-tests used to determine the level of significance between the means. * p < 0.05; ** p < 0.01; *** p < 0.001.

With the notable exception of references to gender identity, Australian and Canadian newspapers do not extensively personalize premiers when introducing them to the public as new leaders. Gender was the only type of personalization appearing in more than a fifth of the stories (presence) and an average of almost once per story (intensity). While news writers occasionally provided some biographical details by reporting the premier’s age, marital status, offspring, upbringing and appearance, they rarely discussed sexuality in relation to an incoming premier, regardless of whether it was to mention their sexual orientation or to sexualize their demeanor. Further analysis confirmed that the ‘privatizing’ dimension of media personalization is not widespread. Almost two-thirds (66%) of the stories do not contain any discussion of a premier’s personal characteristics, while 18.5% include only one indicator of personalization. The remaining 15.6% feature two to seven forms of personalized reporting. No story records all forms of the phenomenon.

The most compelling difference between women and men is the proclivity of journalists in both countries to identify the gender identity of the premier with phrases such as ‘our new premier is a woman!’ and ‘new man at the top.’ Gender differences on this variable are statistically significant for both presence and intensity measures. One-third (34.1%) of the news articles about female premiers explicitly marked their gender, with an average of one mention per story, compared with 18.3% of stories citing the gender identities of men premiers, at an average of 0.39 mentions per story. Table 2 highlights an important country difference: the Australian women premiers received four times as many references to their gender identities per story than their male counterparts, compared with just under twice as many for the Canadian women. This result could reflect the degree of concentration of female premiers, as a relative abundance of women in government leadership roles at the same time may render their gender identities less newsworthy. In Canada, four women held provincial premierships simultaneously for over a year. Australian women’s terms were spread out over time, with only one three-month period featuring three female premiers.

Tables 1 and 2 illustrate that gender and sexual identities are more frequently highlighted in news reports about women and LGBTQ premiers than in stories about men and heterosexual premiers. Sexual identity was noted solely in relation to the lone LGBTQ premier in the sample, Kathleen Wynne, underscoring the heteronormative assumptions circulating in depictions of leaders. The sexual orientation of heterosexual premiers was never overtly discussed or evaluated as relevant to the performance of leadership. In neither country did journalists make more than a few references to the perceived sexual allure of premiers. That gender differences on the sexualization variable are statistically significant for Australia, but not for Canada, reflects the tendency of Canadian journalists to describe men and women premiers alike as flirting with voters. In Australia, only one male premier was depicted with sexualizing references.

Surprisingly, given the gendered mediation literature’s consistent finding of attention to the bodies of women politicians, women premiers did not see their ages or looks mentioned in a higher percentage of stories, or with greater intensity, than their male predecessors. However, Australian women’s family lives were profiled more often than was the case for their male predecessors. The five Australian women premiers—three of whom did not have children—had their childbearing capacities and roles discussed in five times as many news items as their male counterparts, yet the childlessness of the lone Australian male premier was never discussed in newspaper accounts of his first week on the job. All of the Canadian premiers had children, and some had grandchildren, and while references to offspring were more likely to be present and offered with greater intensity in stories about the women premiers, differences between the men and women did not reach statistical significance. Canadian journalists were more inclined to mention the marital relationships of men than those of women, but again the difference is not significant. It appears that the novelty value of Australian women premiers who were not married and/or did not have children sparked attention from journalists accustomed to writing about premiers with traditional nuclear families.

Comparing gender differences in presence and intensity of personalization variables reveals differences in the ways in which each country’s newspapers depict subnational leaders. As Tables 1 and 2 demonstrate, Australian journalists are significantly more likely to draw attention to a female premier’s gender identity, sensuality, marital status, children/childlessness and upbringing than is the case for male premiers, while Canadian journalists do so only for gender identity and sexual orientation. As a result, the Australian women premiers were personalized to a greater extent and in more ways than were their Canadian counterparts, signaling different journalistic norms in the two countries. Baird’s (2004) study of reporting about Australia’s members of parliament found that reporters characterized high-profile women as princesses, cover girls, sex objects, femme fatales, whores, iron ladies, superwoman moms and dumpy housewives. That Australian scribes were more likely than their Canadian counterparts to comment on female leaders’ bodies and family lives may reflect long-standing news practices that cultivate the use of gendered tropes (Baird, 2004).

The effects of gender novelty

Previous research reveals that gender is only one factor influencing women politicians’ news coverage (Wagner et al., 2017). To explore other factors relevant to media personalization, we combined the eight presence indicators into an index and conducted multiple regressions with this index as the dependent variable. The index was constructed by converting each nominal variable into a dummy variable—with a mention coded as 1 and no mention coded as 0—and then tabulated. The maximum value for each premier in each story is eight (8). The index achieved a Cronbach’s alpha of .600. In addition to premier gender (woman or man) and gender novelty (first woman premier), the model ascertains the effect of sexual identity novelty (the presence of an LGBTQ premier in Canada) and pathway to office (whether or not the party was in government when the individual became premier, which occurred more frequently in Australia than in Canada). Influential features of news mediation, notably gender of the reporter, type of news story, and length of the news story, were included to gauge the possibility that articles penned by women and longer opinion pieces may be more likely to include references to identity and personal life. We also controlled for country to allow for the possibility that coverage in one country might be more personalized, regardless of who holds office. Descriptive statistics indicate women averaged .87 (s=1.460) on the index and men .51 (s=.968).

Table 3 presents the regression results for the personalization index. Hypothesis 1, which posited that higher levels of personalization will appear in news coverage of women premiers than in reporting about their male predecessors, was supported in bivariate comparisons. However, significance disappeared when controlling for other variables, notably women premiers’ level of novelty and their pathway to office. Gender novelty mattered most because premier gender only lost significance when the novelty variable was added to the model. The gender identity of the premier, therefore, does not in and of itself predict levels of media personalization.

Table 3.

Regression results for the personalization index measuring the presence of seven indicators of personalization in newspaper coverage of Australian and Canadian premiers.

	Personalization index
Independent variables	Australia	Canada	Both countries
Female premier	.240 (.181)	−.147 (.131)	.010 (.108)
First woman premier	.561 (.201)**	.349 (.159)*	.464 (.125)***
Governing party	.043 (.155)	.593 (.114)***	.373 (.092)***
Female reporter	.584 (.167)**	.057 (.124)	.279 (.101)**
Opinion piece	.428 (.134)**	−.265 (.105)*	.035 (.084)
National paper	.226 (.127)	.044 (.108)	.100 (.083)
Word count	.001 (.000)***	.001 (.000)***	.001 (.000)***
LGBTQ premier	–	.078 (.221)	.068 (.213)
Canada	–	–	−.020 (.090)
N of stories	352	450	802
Constant	−.556 (.162)**	−.509 (.161)**	−.555 (.124)***
Adjusted R²	.239	.174	.177

Note: * p < 0.05; ** p < 0.01; *** p < 0.001.

As posited in Hypothesis 2, journalists in Australia and Canada perceive gender to be more salient for ‘first women’ than for subsequent women, as those premiers exhibiting the news value of novelty experienced significantly higher levels of personalization than did the women who followed in their footsteps. This finding is important because it demonstrates that the presence of women in government leadership roles diminishes perceptions of role incongruity. When the sight of a woman as leader is no longer unusual, journalists are less inclined to write about her identity characteristics, looks and family life.

As the country-specific models in the regression analysis indicate, pathway to office matters in Canada but not Australia. The five Canadian politicians who became premier by winning the leadership of governing parties, deemed ‘heirs apparent’ by Helms (2018), were more personalized than the five individuals who brought an opposition party to victory in a general election (‘election winners’). Perhaps because those who assume the leadership of parties in government are not necessarily well known to the electorate before their ascents, Canadian journalists offer personal details to familiarize voters with the newly selected premier. In contrast, this sort of information might have already been provided about election winners during the general election campaign. In Australia, however, the level of personalization for heirs apparent and election winners was generally the same, which is puzzling. Because Australian heirs apparent are selected by members of the parliamentary party in a process that is typically swift and sometimes secretive, we anticipated that Australian journalists would offer more personal information about them than about election winners. It is possible that there are insufficient cases of election winners in the Australian sample to effectively test the impact of pathway to office on levels of news personalization. Only three of the ten Australian premiers (one woman and two men) took the electoral path, compared with five of the ten Canadian premiers (two women and three men). Australian news culture could also be a factor. Journalists in that country may be more inclined to personalize government leaders regardless of how they achieve power.

Table 3 also shows that Australian and Canadian journalists took advantage of the opportunity afforded by lengthier articles to share personal information about premiers. This is not surprising, as in-depth profiles employ techniques of personalization to lend color and entertainment value to the story (Trimble et al., 2013: 473). Interestingly, personal details about Canadian premiers were less likely to be found in opinion pieces than in regular stories, an unexpected finding considering opinion writers have greater leeway than reporters in how they can write about a subject. In Australia, however, opinion writers were more likely than reporters to personalize premiers. That articles written by female reporters were also more likely to include personalizing references appeared to be driven by gender differences among Australian scribes, as it only achieved statistical significance for Australian coverage. Newspaper type was not significant.

Discussion and conclusion

This study analyzed the impact of premier gender and gender novelty on levels of news-mediated personalization of 20 Australian and Canadian premiers, finding that the increased presence of women in government leadership roles moderates levels of personalized coverage. As regression analysis of the personalization index confirmed, the news value of gender unusualness is an important predictor of media personalization. Media attention to a political leader’s persona and personal life is heightened only when that individual is among the first of their social group to hold that position—in this case, the first woman to be premier. While our study is limited in the number of cases where more than one woman has served as government leader in a jurisdiction, the findings are robust, indicating that less attention will be given to the personal dimensions of female leaders’ identities over time as more women exercise political power. Yet the results also suggest that other path-breaking politicians, such as racialized women or men, will see intense personalization in their coverage until members of their respective social groups hold executive positions in larger numbers. Although the sample size of one LGBTQ premier was too low to draw conclusions about the role of sexual novelty in coverage, the findings demonstrate the importance of including racial and/or sexual novelty when investigating media personalization of political leaders.

This large-scale comparison offers a valuable contribution to the literature by finding relatively low levels of personalization in news-mediated introductions of government leaders in the immediate aftermath of their ascension to power. Two-thirds of news stories said nothing about premiers’ gender identities, bodies or family lives, suggesting journalists are more interested in evaluating new government leaders’ ideas and policy goals than in describing their looks and intimate relationships. Another marker of progress is the relative lack of attention to age and looks. The ten female premiers were no more likely than their male predecessors to have their bodies profiled in stories about their rise to the top job. Perhaps the gendered mind/body dualism is being unsettled by the increased presence of women in positions of power. A more discouraging finding is the reification of the prototypical nuclear family as the exemplar for government leaders. That the presence and intensity of references to marriage and children was greater for Australian women premiers than for their male predecessors likely reflects the atypical nature of these women’s family lives. Three of five women in the Australian sample were child free and two were unmarried. Since most political leaders inhabit the traditional nuclear family, the unusualness of these women’s family forms sparked attention. In contrast, the ‘typical’ family situations of the Canadian women premiers were no more likely to be mentioned than those of the men.

This study offers methodological innovations that will be useful in other contexts and especially for longitudinal comparisons. We developed a comprehensive and rigorous method for measuring the frequency and intensity of nine features of media personalization. Creating an index from the presence indicators afforded a rigorous test of the effects of leader gender, country and novelty on the degree of personalized coverage. As emphasized above, our research confirms that the symbolic representation of women, measured as their increased presence in positions of power, shapes media representations of their rise to government leadership roles. Through qualitative analysis, we plan to illuminate the meanings communicated by these representations, including the role played by self-representation. Future research should test additional hypotheses suggested by our study. For example, more comparative analysis is needed to theorize and test the effects of journalistic culture and pathway to office on levels of media personalization. Regarding gender novelty, the relatively small number of cases of ‘second’ women in the sample of Australian and Canadian premiers did not allow us to gauge a potential lag effect, and this is worth investigating in other contexts. Because the symbolic resonances of a woman in charge might fade in memories over time, the longer the time period between the appearance of the first woman and the advent of a female successor, the more likely news organizations will draw attention to the personal dimensions of the second woman’s identity. Subnational jurisdictions offer an ideal focus of analysis for comparative research given the continued rarity of female leaders at the national level.

The implications of our empirical findings reach well beyond the specificities of Australia and Canada. Female government leaders are widely covered by the media and likely known by most residents of the jurisdiction they lead, hence prolonged exposure to women in executive leadership roles should challenge the association of political power with men and masculinity (Jalalzai, 2016: 446, 448). Our research shows that news coverage can conventionalize the presence of women in male-dominated political spaces in circumstances when the leader’s gender is no longer regarded as unusual. As women’s presence in political leadership roles continues to grow, it is essential to investigate the extent to which media coverage reflects the symbolic effects of higher levels of representation.

Supplemental Material

sj-pdf-1-ips-10.1177_0192512119876083 – Supplemental material for Gender novelty and personalized news coverage in Australia and Canada

Supplemental material, sj-pdf-1-ips-10.1177_0192512119876083 for Gender novelty and personalized news coverage in Australia and Canada by Linda Trimble, Jennifer Curtin, Angelia Wagner, Meagan Auer, V K G Woodman and Bethan Owens in International Political Science Review

Supplemental Material

sj-pdf-2-ips-10.1177_0192512119876083 – Supplemental material for Gender novelty and personalized news coverage in Australia and Canada

Supplemental material, sj-pdf-2-ips-10.1177_0192512119876083 for Gender novelty and personalized news coverage in Australia and Canada by Linda Trimble, Jennifer Curtin, Angelia Wagner, Meagan Auer, V K G Woodman and Bethan Owens in International Political Science Review

Footnotes

Acknowledgements

The authors thank the anonymous reviewers for their helpful insights and suggestions.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by a Social Sciences and Humanities Council of Canada Partnership Development Grant [number 890-2016-0003].

ORCID iD

Linda Trimble

Supplemental material

Supplemental material for this article is available online at

Notes

Author biographies

Linda Trimble is professor of Political Science at the University of Alberta, Canada.

Jennifer Curtin is professor of Politics and Director of the Policy Institute at the University of Auckland, New Zealand.

Angelia Wagner is a post-doctoral fellow at the University of Alberta.

Meagan Auer is a doctoral student at the University of Alberta.

V. K. G. Woodman is a doctoral student at the University of Auckland.

Bethan Owens received a BA Honors in Media Studies from the University of Auckland in 2018.

References

Baird

Julia

(2004) Media Tarts: How the Australian Press Frames Female Politicians. Melbourne: Scribe Publications.

Baxter

Judith

(2018) Women Leaders and Gender Stereotyping in the UK Press. London: Palgrave Macmillan.

Campus

Donatella

(2010) Mediatization and Personalization of Politics in Italy and France: The cases of Berusconi and Sarkozy. International Journal of Press/Politics 15(2): 219–235.

Chappell

Louise

(2002) Gendering Government. Feminist Engagement with the State in Australia and Canada. Vancouver: UBC Press.

Curtin

Jennifer

(2015) Review Essay: The prime ministership of Julia Gillard. Australian Journal of Political Science 50(1): 190–204.

Curtin

Jennifer

(2006) Joan Kirner: The first feminist. In Strangio

Paul

Costar

Brian

(eds) The Victorian Premiers 1856-2006. Sydney: Federation Press, 351–362.

Elazar

Daniel J

(1997) Contrasting unitary and federal systems. International Political Science Review 18(3): 237–251.

Everitt

Joanna

Camp

Michael

(2009) One is Not Like the Others: Allison Brewer’s leadership of the New Brunswick NDP. In Bashevkin

Sylvia

(ed.) Opening Doors Wider: Women’s Political Engagement in Canada. Vancouver: UBC Press, 127–144.

Falk

Erika

(2010) Women for President: Media Bias in Nine Campaigns. Urbana: University of Illinois Press.

10.

Freelon

Deen

(2010) ReCal: Intercoder reliability calculation as a web service. International Journal of Internet Science 5(1): 20–33.

11.

Garzia

Diego

(2011) The Personalization of Politics in Western Democracies: Causes and consequences on leader-follower relationships. The Leadership Quarterly 22(4): 697–709.

12.

Gidengil

Elisabeth

Everitt

Joanna

(1999) Metaphors and Misrepresentation: Gendered mediation in news coverage of the 1993 Canadian leaders’ debates. International Journal of Press/Politics 4(1): 48–65.

13.

Harp

Dustin

Loke

Jaime

Bachmann

Ingrid

(2010) First Impressions of Sarah Palin: Pit bulls, politics, gender performance, and a discursive media (re)contextualization. Communication, Culture, and Critique 3(3): 291–309.

14.

Hayes

Danny

Lawless

Jennifer

(2016) Women on the Run: Gender, Media, and Political Campaigns in a Polarized Era. New York: Cambridge University Press.

15.

Helms

Ludger

(2018) Heir Apparent Prime Ministers in Westminster Democracies: Promise and performance. Government & Opposition: 1–23. Advance online publication. doi 10.1017/gov.2018.22

16.

Hirdman

Anja

Kleberg

Madeleine

Widestedt

Kristina

(2005) The Intimization of Journalism: Transformations of mediatized public spheres from the 1880s to current times. Nordicom Review 26(2): 109–117.

17.

Holtz-Bacha

Christina

Langer

Ana Ines

Merkle

Susanne

(2014) The Personalization of Politics in Comparative Perspective: Campaign coverage in Germany and the United Kingdom. European Journal of Communication 29(2): 153–170.

18.

Hoyt

Crystal

Murphy

Susan

(2016) Managing to Clear the Air: Stereotype threat, women, and leadership. The Leadership Quarterly 27(3): 387–399.

19.

Jalalzai

Farida

(2016) Shattered not Cracked: The effect of women’s executive leadership. Journal of Women, Politics & Policy 37(4): 439–463.

20.

Jenkins

Cathy

(1999) Media Treatment of Women in Politics in Australia. In Ewart

Jacqui

(ed.) Journalism Theory and Practice: Proceedings of the 1998 Journalism Education Association Conference, Sydney: The Australian Press.

21.

Jenkins

Cathy

(2006) Women in Australian Politics: Mothers only need apply. Pacific Journalism Review 12(1): 54–63.

22.

Kahn

Kim Fridkin

Goldenberg

Edie

(1991) Women Candidates in the News: An examination of gender differences in U.S. Senate campaign coverage. Public Opinion Quarterly 55(2): 180–199.

23.

Karvonen

Lauri

(2009) The Personalisation of Politics: A Study of Parliamentary Democracies. Colchester: ECPR Press.

24.

Koenig

Anne

Eagly

Alice

Mitchell

Abigail

Ristikari

Tina

(2011) Are leader stereotypes masculine? A meta-analysis of three research paradigms. Psychological Bulletin 137(4): 616–642.

25.

Langer

Ana Ines

(2010) The Politicization of Private Persona: Exceptional leaders or the new rule? The International Journal of Press/Politics 15(1): 60–76.

26.

Lawrence

Regina G

Rose

Melody

(2010) Hillary Clinton’s Race for the White House: Gender Politics and the Media on the Campaign Trail. Boulder and London: Lynne Rienner.

27.

McAllister

Ian

(2013) The personalization of politics in Australia. Party Politics 21(3): 337–345.

28.

O’Neill

Brenda

Stewart

David

(2009) Gender and political party leadership in Canada. Party Politics 15(6): 737–757.

29.

Raicheva-Stover

Maria

Ibroscheva

Elza

(ed.) (2014) Women in Politics and Media: Perspectives from Nations in Transition. New York: Bloomsbury.

30.

Sawer

Marian

Tremblay

Manon

Trimble

Linda

(ed.) (2006) Representing Women in Parliament: A Comparative Study. Abingdon and New York: Routledge.

31.

Simon

Stefanie

Hoyt

Crystal

(2012) Exploring the effect of media images on women’s leadership self-perceptions and aspirations. Group Processes & Intergroup Relations 16(2): 232–245.

32.

Sreberny-Mohammadi

Annabelle

Ross

Karen

(1996) Women MPs and the Media: Representing the body politic. Parliamentary Affairs 49(1): 103–115.

33.

Steele

Claude

Spencer

Steven

Aronson

Joshua

(2002) Contending with Group Image: The psychology of stereotype and social identity threat. In Zanna

Mark

(ed.) Advances in Experimental Social Psychology 34. New York: Academic Press, 397–440.

34.

Stockemer

Daniel

Tremblay

Manon

(2015) Federalism and Women’s Representation: Do federations have more women legislators than centralized states? Publius 45(4): 605–625.

35.

Thomas

Melanee

Bittner

Amanda

(eds) (2017) Mothers and Others: The Role of Parenthood in Politics. Vancouver: UBC Press.

36.

Thomas

Melanee

(2018) In Crisis or Decline? Selecting women to lead provincial parties in government. Canadian Journal of Political Science 51(2): 379–403.

37.

Tolley

Erin

(2016) Framed: Media and the Coverage of Race in Canadian Politics. Vancouver: UBC Press.

38.

Trimble

Linda

(2017) Ms. Prime Minister: Gender, Media and Leadership. Toronto: University of Toronto Press.

39.

Trimble

Linda

Raphael

Daisy

Sampert

Shannon

Wagner

Angelia

Gerrits

Bailey

(2015) Politicizing Bodies: Hegemonic masculinity, heteronormativity, and racism in news representations of Canadian political party leadership candidates. Women’s Studies in Communication 38(3): 314–330.

40.

Trimble

Linda

Wagner

Angelia

Sampert

Shannon

Raphael

Daisy

Gerrits

Bailey

(2013) Is it Personal? Gendered mediation in newspaper coverage of Canadian national party leadership contests, 1975-2012. The International Journal of Press/Politics 18(4): 462–481.

41.

Van Acker

Elizabeth

(1999) Different Voices: Gender and Politics in Australia. S. Yarra: MacMillan Education.

42.

Van Aelst

Peter

Sheafer

Tamir

Stanyer

James

(2012) The Personalization of Mediated Political Communication: A review of concepts, operationalizations and key findings. Journalism 13(2): 203–220.

43.

Van Zoonen

Liesbet

(1994) Feminist Media Studies. London: Sage.

44.

Verge

Tània

Pastor

Raquel

(2018) Women’s political firsts and symbolic representation. Journal of Women, Politics & Policy 39(1): 26–50.

45.

Vial

Andrea Jaime Napier

Brescoll

Victoria

(2016) A Bed of Thorns: Female leaders and the self-reinforcing cycle of illegitimacy. The Leadership Quarterly 27(3): 400–414.

46.

Wagner

Angelia

Trimble

Linda

Sampert

Shannon

Gerrits

Bailey

(2017) Gender, competitiveness and candidate visibility in newspaper coverage of Canadian party leadership contests. International Journal of Press/Politics 22(4): 471-489.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.09 MB

0.34 MB

0.09 MB