Graduate School of Communication
Master’s Thesis
A Woman’s got to Write
What a Woman’s got to Write
The Effect of a Journalist’s Gender on the Perceived Credibility of News Articles
__________ by Elena Klaas
11717459
Master’s Programme Communication Science Supervisor: Dr Mark Boukes
1 February 2019 Word Count: 7,493
Abstract
This study investigates the effect of an author’s gender on the perceived credibility of news articles. In an experiment, participants (N = 205) evaluated the credibility of two different news articles with a typical male and female topic either written by a male or a female journalist. News articles written by male journalists were seen as significantly more credible overall, but especially when the article had a typical male topic. Both effects were stronger for female participants than for male participants. Younger male participants rated female journalists overall as less credible compared to older male participants. The experiment highlights that audiences still have a preconceived idea about what female journalists are meant to report about, while generally giving preference to male journalists. This underlines the need for a better visibility of female journalists in all areas of reporting to viably change these perceptions.
The Effect of a Journalist’s Gender on the Perceived Credibility of News Articles
Within the field of communication science, and journalism specifically, there is little that is as widely researched as the concept of media credibility. Its tremendous relevance for journalism seems obvious: If an article, and resulting the whole newspaper or media outlet, is not seen as credible, its readership will decrease (Thorson, Meyer, Denton, & Smith, 2008; Weibel, Wissmauth, & Groner, 2008). Researched factors influencing the audience’s perception of media credibility are plentiful and include the distinction between online and traditional media (e.g., Johnson & Kaye, 1998), the audience’s media use (e.g., Rimmer & Weaver, 1987), and source expertise (e.g., Eastin, 2001).
This study will focus on a factor that has taken a dominant place in societal discourse in recent years: the gender of an author. The idea that the audience’s perception of what constitutes a more or less credible news piece might be influenced by the author’s gender is not a new one. Initial studies focusing on gender as a credibility factor lead back as far as 1968, and found that articles written by female journalists were often seen as less credible (e.g. Goldberg, 1968). However, traditional gender roles in society were much more rigid back then compared to how they are today. Further, journalism used to be a male-dominated profession (Sebba, 1994), a development that has changed drastically over the last decades. Currently, the vast majority of all journalism graduates are females (Boateng, 2017).
Nevertheless, gender dynamics in journalism are still asymmetric, as in most countries only a small percentage of female journalists make up the media landscape (Edström, 2017; Usher, Holcomb & Littman, 2018; Ross, Boyle, Carter, & Ging, 2018; Strong & Hannis, 2007; Women’s Media Center, 2017). The 2015 Global Media Monitoring Project showed that worldwide, women only report around 37% of all stories in print,
television and radio combined. In 2015, only 31% of German print journalists were female (Global Media Monitoring Project, 2015b). In addition to that, there is still a striking difference between the topics that male and female journalists are assigned to. Previous research has shown that the majority of women are delegated to cover stereotypical ‘female’ topics, such as lifestyle and fashion, while men continue to dominate ‘male’ topics like technology and economics (e.g., North, 2016; Steiner, 2012).
As a result, despite loosening gender roles in society, the question arises if news consumers are still influenced by the gender factor in journalism. If readers are used to mainly reading certain ‘male’ topics written by men, do they evaluate the articles less
positively if a women wrote the article, and vice-versa? The main aim of this study is thus to examine whether the unexpected gender of the author impedes on the perceived credibility of an article. While maximising credibility is an obvious goal for communication research, investigating this is also an important step to evaluate to what extent gender stereotypes still exist in Western societies. Clear audience perceptions on which topics should be covered by men or women would display a continuing presence of clear gender roles that should be minimised in today’s society.
Existing research about gender differences in print journalism is vastly outdated, with most studies being over 30 years old. Recent research investigating the effect of gender in journalism focuses on TV newscasters. Through the omnipresence of mobile phones and tablets, however, written journalism is becoming increasingly important again.
As a result, this study aims to answer the following research question: To what extent
does the gender of a journalist influence the perceived credibility of a news article?, by
conducting an experiment including a focus on the distinction between typical male and female topics.
Theoretical Framework Journalism and Gender
The first experiment investigating whether female authors were evaluated more negatively than male authors was conducted in 1968 by Philip Goldberg. Goldberg gave female participants professional articles that were either written by ‘John McKay’ or ‘Joan McKay’. The results showed that articles written by ‘John’ were consistently rated higher in terms of value, persuasiveness and profundity than those written by ‘Joan’. This was
especially evident and yielded significant results when the articles were written about
traditional masculine fields such as city planning and law rather than traditional female fields (dietetics and education). His experiment was replicated various times with several
modifications; however, often yielding opposing results.
Many studies did not find any differences between the evaluations of male and female authors (e.g., Baruch, 1972; Chobot, Goldberg, Abramson, & Abramson, 1974) or found that female participants rated articles written by women to be better (Levenson, Burford, Bonno, & Davis, 1975). Paludi and Bauer (1983) also included male participants into their sample and came to the conclusion that both men and women rated articles written by a male author to be more favourable. Important to note is that men showed higher prejudice against female authors. This contrasts Miller and McReynold’s (1973) findings, where female participants showed higher prejudice against females.
Goldberg’s results, and subsequent replications, were subject to a meta-analysis by Swim, Borgida, Maruyama and Myers in 1989. They analysed 123 studies published between 1974 and 1979 and found that, on average, studies showed a lower rating of female authors, albeit effect sizes were considerably small, ranging from -0.05 to -0.08 (e.g., Gallivan, 1991). The meta-analysis also noted that if men were rated more favourably than women, this
difference was greatest when the stimulus material was masculine. Further, they concluded that the less information in addition to the gender was presented, the more stereotypical judgements were made. This is in accordance with previous findings in sex discrimination research: if additional information about a person is present, there is no need to rely on sex-role stereotypes (e.g., Locksley, Borgida, Brekke, & Hepburn, 1980; Tosi & Einbender, 1985). The meta-analysis’s main conclusion, though, was that many results were
insignificant, signaling that Goldberg’s initial findings are given more strength than they should. As a result, the present study sees Goldberg’s initial experiment and its replications as a starting point. While certain aspects of previous research are adopted (e.g. the traditional male and female topics), this study aims to contribute to the current literature by addressing Goldberg’s methodological limitations and including other characteristics of the participants.
Related research investigated gender and its influence on the evaluation of bylines, the author’s information at the beginning of an article. Results were similarly inconclusive: whereas Shaw, Cole, Moore and Cole (1981) showed that male participants made more stereotypical judgements about female bylines, Burkhart and Sigelman (1990) found bylines were evaluated similarly regardless of a male or female author. Most of these studies are characterised by a simple design, small sample sizes and ambiguous measurements for the evaluation of the articles, which could partly explain the lack of significant findings and the inconsistent results. More recent research, including the present study, uses the
well-structured concept of credibility to assess the evaluation of the articles. Gender and Credibility
The importance of credibility in journalism seems evident: When a medium or an author is not seen as credible by the public, channels will be switched or newspapers will not be bought anymore. As a result, credibility in journalism has been a widely researched field
since the 1950s and has since evolved into a multidimensional construct (Golan, 2010). One main distinction is the differentiation between medium and source credibility.
Hovland, Jannis and Kelley (1953) found that people are more likely to believe a fact if the source itself is seen as credible, which was determined by the source’s expertness and
trustworthiness. Since then, the concept of source credibility has been extended into three- or
five-dimensional constructs by several other scholars (e.g., Berlo, Lemert & Merz, 1970; Eisend, 2006). Medium credibility was first measured on a large scale by Gaziano and McGrath in 1986. They developed a 12-item additive index which was shortened into five dimensions by Meyer (1988): A news item is deemed credible if it is unbiased, accurate, fair,
tells a whole story and if it can be trusted. This shortened version of the initial 12-item scale
is now seen as standard for credibility research (West, 1994).
As described, media credibility research is an extensive field among communication science scholars. However, there is only little contemporary research investigating the influence of an author’s gender on the credibility of a journalistic piece. Previous research, such as Goldberg’s experiment, worked with individually designed scales of how the participants were asked to evaluate the articles. Using credibility as a measure to rate the articles, this allows for a much more reliable comparison across studies. One of the first studies measuring gender influences on medium credibility was conducted by Andsager (1990), who investigated the perceived credibility of male and female political columnists. He observed the tendency that male participants rated a male byline to be more credible than a female byline. In a follow-up study, he also found that females preferred columns written by women (White & Andsager, 1991).
Andsager’s most recent study on political columnists and credibility included race as another potential influence (Andsager & Mastin, 2003). Interestingly, albeit using a neutral
topic, females were ascribed greater credibility than males and African-American women were given the most credibility overall. A similar study concerning political blogs and
credibility, however, showed that posts with male authors were rated higher in credibility than those with female authors (Armstrong & McAdams, 2009). This was partly explained by the fact that in the US, only three of the Top 30 political blogs were authored by women (Harp & Tremayne, 2006).
Most recent research on the relationship between gender and journalism investigates newscasters rather than newspaper journalists (e.g., Brann & Himes, 2010; Engstrom & Ferry, 2000). Weibel et al. (2008) examined how gender and age affects newscaster’s
credibility and found that while male newscasters were seen as more credible, messages read by them were perceived as less credible as opposed to messages read by female newscasters. These mixed results underline the ambiguity of the existing research — there is little consensus whether and to what extent gender stereotypes have an influence on people’s evaluation of journalistic work. Precisely due to the mixed research results, this study aims to contribute to the existing literature by re-assessing the question. Furthermore, as this brief overview has shown, the vast majority of existing research is outdated. Given the changes gender roles have undergone in recent years, it is important to investigate whether there has been an attitude change in the last decades. In addition, previous research was almost exclusively executed in an American context, which is why this study will be using a more European approach. Above all, little to no research has explored if there are characteristics apart from the participant’s gender that influence the perceived credibility. This research gap was already criticised by several scholars (e.g., Gross & Geffner, 1980; Weibel et al., 2008). As a result, the present study takes into consideration important characteristics of previous studies, such as the use of different article topics and the distinction between male and female
readers, while attempting to advance this field of research by incorporating other participant characteristics.
Hypotheses
Traditional gender roles stem from the patriarchal notion of the public versus private sphere (Donovan, 1992). Stereotypically, there was a clear assignment of men and women to these spheres: Women were responsible for the private sphere including the home and the family, whereas men guarded the public sphere in the government and the political realm (Smyth, 2008). Albeit gender roles are slowly changing, these stereotypes are still
incorporated into people’s opinions, often expressed by the assumption that men and women have different interests. Orwant and Cantor (1977) proved this by asking communication students to judge men’s and women’s interest in journalism. They found that women were consistently judged as less interested in politics and business, and more interested in fashion and cooking. As a result, if this opinion is engrained in people’s minds, it could be extended to the idea that male and female journalists should also report about these separate topics.
This development is very much visible in the media landscape, as proven by Parker (2004). His findings that male news anchors are more frequently assigned to hard news, whereas women focus more on soft news and human interest stories, are also reflected in several content analyses (e.g., Cann & Mohr, 2001; ‘Fairness and Accuracy in Reporting’, 2001; Helwerth, 2002, Lumby, 1994; North, 2016). It seems that there is a clear distinction between topics that are considered ‘female’ and ‘male’, and are thus allocated to female or male journalists, respectively. As argued by Djerf-Pierre (2007), this distinction stems from the above-mentioned theory of the public versus private sphere. She highlights that
masculinity is associated to power, prestige and status, and thus male journalists are assigned to cover topics associated with those characteristics.
This reflects that typically male and female topics exist in journalism. News about politics, economy, or technology covering the public sphere are typically male-orientated topics (North, 2016). On the contrary, stories about fashion, childcare or cooking covering the private sphere are considered female-orientated topics (Cann & Mohr, 2001). Previous
research, using articles with typical male and female topics showed that if the topic did not fit with the author’s gender, the articles were rated more negatively (Goldberg, 1968; Paludi & Strayer, 1985; Shaw et al, 1981). This phenomenon is based on role congruity theory (Eagly & Karau, 2002), which posits that if both women or men write about topics that are seen as incongruent with their gender, bias against them and scepticism about the article’s credibility will be stronger. Role congruity theory has been used to explain gender biases in other fields, for example demonstrating biases against female scientists (Nosek et al., 2009) or in
scholarly communication (Knobloch-Westerwick & Glynn, 2013). Based on these findings, the first hypotheses for this study are as followed:
Hypothesis 1: Articles written by authors with gender-congruent topics will be
perceived as more credible as opposed to articles written by authors with gender-incongruent topics.
Hypothesis 1a: Specifically, an article with a typically male topic will be perceived as
more credible if written by a male as opposed to a female author.
Hypothesis 1b: An article about a typically female topic will be perceived as more
credible if written by a female as opposed to a male author.
A characteristic that has been found to influence the evaluation of the news pieces is the participant’s gender. Most studies indicate that male participants make more stereotypical judgements (e.g., Andsager & Mastin, 2003; Shaw et al., 1981) and rate articles written by males higher in credibility (e.g., Andsager, 1990; Armstrong & McAdams, 2009). Generally,
similarity with the source or in this case author is seen to increase the perceived credibility (Aune & Kukichi, 1993). Furthermore, men are believed to use more heuristic cues –
including stereotypes – in information processing (Armstrong & Nelson, 2005) and generally hold more traditional gender views than women (Bolzendahl & Meyers, 2004). As a result, the following hypothesis will be tested:
Hypothesis 2: The role-congruity effect will be stronger for male as opposed to
female readers.
Based on gender studies and research on attitudes towards feminism, other characteristics of the participant could influence his or her assessment of the article’s
credibility. Bettencourt, Vacha-Haase and Byrne (2011) showed that age influences people’s attitudes towards feminism: compared to older adults, younger people have a more positive attitude towards feminism. Several other studies have displayed a similar conclusion (Dambrot, Papp, & Whitmore, 1984; Spence & Helmreich, 1979). This is explained by the fact that previous generations have grown up with more traditional views on gender and family (Oláh, Kotowska, & Richter, 2018; Twenge, 1997). These gender roles have changed significantly over the last 50 years, as more children are growing up with a single parent and/ or working mothers (Kiecolt & Acock, 1988; Willetts-Bloom & Nock, 1994). In one of the replications of Goldberg’s study, participants were split between students and older people averaging 43 years (Gross & Geffner, 1980). They found that older participants evaluated women’s articles less positive compared to the younger age group. This results in the third hypothesis of this study:
Hypothesis 3: The role-congruity effect will be stronger for older readers as opposed
Furthermore, another characteristic that is said to influence people’s proneness to gender stereotyping are conservative attitudes. Conservative attitudes are commonly associated with a more traditional viewpoint on family and gender roles, leading to more traditional attitudes towards women (e.g., Bettencourt et al., 2011; Davis & Greenstein 2009). This is due to the fact that conservatism is an ideology that is resistant to change (Jost,
Glaser, Kruglanski, & Sulloway, 2003). More conservative people are less open to new experiences (Carney, Jost, Gosling, & Potter, 2008), and are thus less likely to adapt new gender roles but prefer to stick to traditional views. People who have a more liberal political view, by contrast, are more likely to have feminist attitudes and less gender stereotypes (e.g., Lottes & Kuriloff, 1992; Twenge, 1997), as they are more open to change and new
developments (Cichocka, Bilewicz, Jost, Marrouch, & Witkowski, 2016). Based on these characteristics, the final hypothesis of the study will be the following:
Hypothesis 4: The role-congruity effect will be stronger for readers with more
conservative views as opposed to readers with more liberal views. Method
Design
A between-groups 2×2 factorial design was used. The factors consisted of the gender of the author (male or female) and the topic of the article (technology or fashion). The survey-embedded experiment was administered via the online self-completing questionnaire tool
Qualtrics.
Sample
Due to financial restraints, participants were collected using a convenience sample over the period of December 2, 2018 until December 16, 2018. To achieve a more
reaching participants covering different aspects of the population. To reach as many participants as possible, the questionnaire was sent out in both English and German. As a result, participants were collected via social networks and email, using the snowballing technique.
The questionnaire was filled in by a total of 226 participants of which 118 were female (57.1%), 87 male (42.4%) and one genderfluid (0.5%). As 21 people failed the manipulation check, which will be explained below, they were excluded from the study, leaving a final sample of 205 participants. The respondents came from 21 different countries, with 147 participants from Germany (71.7%), followed by 13 participants from the
Netherlands (6.3%) and 10 from Italy (4.9%). The remaining 35 respondents were from other European countries as well from North America, Asia and Australia. The participants’ age ranged from 18 to 68 years with an average age of 30.78 years (SD = 12.81). The
participants’ education level was diverse, with most participants having a high school degree (55 participants, 26.8%), followed by a Master’s degree (50 participants, 24.4%), and a Bachelor’s degree from a university (50 participants, 24.4%) or a university of applied sciences (35 participants, 17.1%). The minority obtained less than a high-school degree (13 participants, 6.3%) or a Doctor’s degree/PhD (2 participants, 1.0%). Finally, the participants political orientation was measured on a scale from 1, left-wing to 11, right-wing and showed a slightly more left-wing sample with a mean of 4.24 (SD = 1.69).
Stimulus Material
To investigate the role-congruity effect, it was important to find article topics that are considered typically male or female. Based on previous studies, four topics were chosen that were deemed to be typically male (technology and banking) or female (childcare and
in an article about Tesla’s profits in 2018, a Goldman Sachs scandal, social media bans for nannies as well as the launch of a new H&M collection.
A pilot study with 34 participants was administered to select the final articles. As previously done by Paludi and Strayer (1985), participants were randomly assigned to two of the four articles and had to subsequently answer whether they thought the article was written by a male or a female author. The results showed clear indications for three of the four articles. 89.5% of participants said the Tesla article was written by a male author and 87.5% said the Goldman Sachs article was written by a male author. Similarly, 82.4% said the H&M article was written by a female author. The article about child care was not as conclusive, however, with 56.3% of participants indicating a male author and 43.8% a female author.
Further, similar to White and Andsager (1991), participants were then asked to rate to what extent they believed the article was a men’s or a women’s topic. To test this, a paired samples t-test between all four articles was performed. As the participants were asked to rate two articles each, the paired samples t-test allowed for a direct comparison within the groups. A significant difference was found between the articles about Tesla and H&M: participants rated the Tesla article to be more of a male topic (M = 2.50, SD = 0.55) than the H&M article (M = 3.67, SD = 0.52). This is a large significant difference: t (5) = -3.8, p = .013, d = 1.56. All other pairs did not show significant differences. Thus, these results led to choosing the article about Tesla as a typically male article topic, and the article about H&M as a typically female article topic for the main study. Given the length and style of the articles, as well as the pilot study’s results, the articles and their topics are comparable, adding to the
measurement validity of the study.
To manipulate the gender of the authors for the final stimuli, the pilot study also included the assessment of four different author photos. The photos were previously
constructed using FaceApp and Photoshop to design two male-female pairs with the same facial features, but a different gender. Based on existing scales (Ohanian, 1990; Reysen, 2005; Surawski & Ossof, 2006), participants had to rate one male and one female photo from different pairs on the following dimensions: competent-incompetent, qualified-unqualified,
knowledgeable-unknowledgeable, likeable-unlikeable, approachable-unapproachable, reliable-unreliable, trustworthy-not trustworthy, attractive-unattractive, friendly-unfriendly.
In line with Reysen’s (2005) likeability scale and Ohanian’s (1990) expertise scale, these dimensions were transformed accordingly; one expertise scale (α = 0.86, M = 8.18, SD = 2.72) and one likeability scale (α = 0.89, M = 9.82, SD = 3.88), compromising all items apart from attractiveness, which was not in tune with either scale and was thus excluded from the analysis. In order to test that the two corresponding author pairs were not rated significantly different on these dimensions, independent samples t-tests were performed. For the first author pair, significant differences between both scale measures were found. The second author pair yielded no significant results for Likeability (t (31)= 0.27, p = .788) and Expertise (t (31)= 0.18, p = .862). As a result, the second pair of author photos was used, as they were rated to be similar in all dimensions while still portraying a different gender.
Based on these results, four stimuli were constructed for the final study. Both the
Tesla and the H&M article were used twice; once including a male author name (Daniel
Evans, or Daniel Beck in the German translation) and photo, and once including a similar female name (Danielle Evans, or Daniela Beck in the German translation) with the matching photo. The author names were chosen based on the most popular first and last names in the UK and Germany around 1980 to match the approximate age of the pair in the photos. The articles were designed to look like an online news article but without any additional layout of a specific news site in order to not compromise the study’s internal validity. The four
different stimuli were used as the independent variable of the study (see Appendix for the stimulus material).
Procedure
In the final study, the participants were randomly assigned to one of the four stimuli. After reading one of the articles, they were asked to rate its credibility by evaluating several factors. Afterwards, they were asked to complete a range of demographic questions about their gender, age, nationality, and level of education. They were then asked to self-assess their political orientation, as well as answer a range of questions about contemporary social issues. Finally, participants had to identify the topic of the article they read as well as the author’s gender.
Measures
Dependent variable: perceived credibility. The credibility of the news article was assessed using an adapted form of Flanagin and Metzger’s (2000) 5-item scale that is based on dimensions commonly used in other media credibility research. Participants were asked to what extent they found the article to be (a) accurate, (b) trustworthy, (c) biased, (d)
comprehensive and (e) believable on 7-point Likert scales ranging from 1 = strongly disagree to 7 = strongly agree. Scores for bias were reverse-coded, so that a higher score for all
dimensions adhered to the perception of more credibility. The resulting scale had a reliability measure of α = 0.74 with a mean of M = 4.65 and a standard deviation of SD = 1.01.
Moderator: conservative cultural attitudes. To measure the conservative cultural attitudes of the participants, they were asked to agree or disagree with four statements that were loosely adapted from the social and economic conservatism scale (Everett, 2013) and a scale developed by Malka, Lelkes and Soto (2017): (a) “Same sex couples should have the same right to adopt children as heterosexual couples.”, (b) “Abortions should be legal in all
European countries.”, (c) “There should be stricter immigration laws in Europe.”, (d) “Mosques should not be allowed to be built in the city centres of European cities”.
Responses were measured on a 7-point Likert scale ranging from 1 = strongly agree to 7 =
strongly disagree. Statements (c) and (d) were reverse coded so that a higher score on all
dimensions indicated a more conservative cultural attitude. A total scale was constructed by averaging the means of all items, resulting in a scale with a reliability measure of α = 0.57 (M = 2.32, SD = 0.96). As the reliability of the scale is quite low, the results need to be
interpreted with caution.
Manipulation check. To assure that the stimuli were perceived as intended, the final set of questions contained the manipulation check. First, participants were asked if they remembered the topic of the article, and had to choose between politics, healthcare, sports,
economics, fashion, and “I do not remember”. Further, participants were asked about the
gender of the author, choosing between male, female, “There was no sign of the gender of the
author” and “I do not remember”. 80.1% of all participants chose both the right topic and
gender. Ten participants did not remember the topic or chose a wrong topic while also indicating that they did not remember the gender of the author. Seven participants chose the wrong gender, while four said there was no indication of the author’s gender. As it is assumed that those participants did not read the article properly, they were excluded from the study.
Finally, 26 respondents said they did not remember the author. It was chosen not to delete these participants as they did not remember a wrong gender, and thus did not base their answers on a wrong memory. As a result, it is assumed that they could have been
unconsciously influenced by the gender without consciously noticing it. It is also interesting to note that in the different conditions, most people (20%) did not remember the male author writing about Tesla. Contrary, only 8.3% of participants could not recall the female author
with the same article. This could lead to the conclusion that if the author ‘fits’ with the topic, people pay less attention to it, whereas if the fit is not congruent, people will notice more.
Randomisation check. To check if the participants’ age, gender, and political orientation was distributed evenly across the four conditions, two one-way analysis of variance (ANOVA) and a chi-square analysis were conducted. The chi-square analysis with gender as the grouping variable was not significant, χ2(1, 205) = 7.14, p =.308. The first ANOVA, with the four conditions as independent variable, and age as dependent variable was also not significant, F[3, 204]= 1.54, p =.207. The second ANOVA, with the four conditions as independent variable, and political orientation as dependent variable, was also not
significant, F[3, 204]= 1.09, p =.356 showing that the randomisation across the four conditions was successful in terms of age, gender and political orientation.
Results
Main Effect: Gender Congruity
To test whether articles with gender-congruent topics were perceived as more credible than articles with gender-incongruent topics, an independent t-test was conducted. The independent variable consisted of two different groups: gender-congruent, including the
Tesla article written by a man and the H&M article written by a woman, and
gender-incongruent, including the Tesla article written by a woman and the H&M articles written by a man. The dependent variable used was the perceived credibility. The t-test showed no significant results, t (199.795)= -1.48, p = .140. Participants that were exposed to the stimuli in which the author and the topics were congruent did not evaluate the credibility to be higher (M = 4.76, SD = 0.89) compared to participants exposed to the stimuli in which the author and topics were incongruent (M = 4.55, SD = 1.10). As a result, Hypothesis 1 was rejected.
However, albeit insignificant, the direction of results indicates that role-congruent articles were rated slightly higher in credibility than role-incongruent articles.
To test the participants’ differences in the perceptions of credibility between both the gender of the author as well as the topics of the article, a two-way analysis of variance
(ANOVA) was performed with credibility as the independent variable and the author’s gender and article topics as fixed factors. The main effect showed a significant difference between the four groups, F[3, 201] = 3.12, p = .025, η² = 0.05, displaying an effect of medium
strength. The ANOVA indicates that the stimuli influenced the participants’ perception of the credibility. Table 1 gives an overview of the mean differences between the four stimuli.
The table shows that overall, both articles written by a male author were seen as more credible, but especially the article about Tesla written by a man. The Tesla article written by a female author was seen as least credible overall. Albeit the overall interaction model of gender and article topic was not significant, F[1, 201] = 2.78, p =.097, post-hoc comparisons revealed a significant difference (Mdifference = 0.54, p = .008) between the male and female
difference was found between both the male and female author of the H&M article (p = .698). Thus, Hypothesis 1b was rejected. These findings lead to the conclusion that the
role-incongruity effect only applies for female authors.
Finally, the ANOVA also revealed a significant result for the author’s gender overall,
F[1, 201] = 4.86, p = .029, η² = 0.02, displaying an effect of low strength. This indicates that
articles written by a male author were seen as significantly more credible (M = 4.79, SD = 0.95) compared to articles written by a female author (M = 4.49, SD = 1.04).
Moderation: Gender of the Participants
To test whether the role-congruity effect was stronger among male than female participants, a two-way ANOVA was performed with the participant’s gender as a moderator variable and congruity (vs. role-incongruity) as the independent variable. Results displayed no statistically significant interaction, F[1, 200] = 0.11, p = .746. While the main effect of participants’ gender was also not significant, F[2, 200] = 1.70, p = .185, it was visible that male participants rated the articles as somewhat less credible (M = 4.52, SD = 1.19) than female participants (M = 4.74, SD = 0.99).
Since a significant difference was found for the role-congruity effect for female authors, the decision was made to test whether there was a moderating effect of the participants’ gender. To test the effect, two independent t-tests for male and female
participants separately were performed. Testing only male participants showed no significant results, t (41)= 2.03, p = .182, indicating that male participants did not rate the credibility of the Tesla article written by a man (M = 4.70, SD = 0.86) to be significantly higher than the same article written by a woman (M = 4.28, SD = 0.86), albeit the mean differences point in the hypothesised direction. Surprisingly, the same test for female participants showed significant results, t (53)= 2.46, p = .017, indicating that female participants rated the
credibility of the Tesla article written by a man (M = 4.93, SD = 0.81) to be significantly higher than written by a woman (M = 4.26, SD = 1.10). Female participants, thus, were more sensitive to the role-congruity effect, but only for female and not male journalists.
Further, similar results were found when comparing both articles written by a male author with both articles written by a female author: while male participants did not view articles written by a male author to be significantly more credible (t (85) = -1.03, p = .307), female participants did (t (104.199) = -2.14, p = .034). Concluding, Hypothesis 2 was rejected, as the results point in the opposite direction and apply for female rather than male participants.
Moderation: Age of the Participants
To test whether the role-congruity effect of H1 was stronger among older
participants, a moderation analysis was conducted with PROCESS (Model 1; Hayes, 2013). The overall model was significant, F(3, 201) = 5.44, p = .001, R2 = 0.08. However, the interaction effect between role-congruity and age was not significant, b = -0.002, t (201) = -0.24, p = .810. Figure 1 displays the differences in the participants’ ages on the credibility.
Figure 1. Effect of participant’s age on the credibility of the congruent vs.
gender-incongruent news articles. The y-axis scale denotes the perceived credibility and the x-axis denotes the participant’s age.
The graph shows that, on average, credibility for all articles decreased with age. However, the difference in credibility between gender-congruent and gender-incongruent topics remained roughly equal across all ages. Hence, the effect of congruity is not conditional upon age.
To test whether age influenced the perception of credibility of all articles written by a male author compared to all articles written by a female author, another moderation analysis was conducted with PROCESS (Model 1; Hayes, 2013). Albeit results showed an overall significant model, F(3, 201) = 6.66, p < .001, R2= 0.09, the interaction effect was not
significant (b = -0.017, t (201) = -1.63, p = .104). As a result, Hypothesis 3 was rejected: Age does not moderate the effect of the author’s gender on the perceived credibility of a news article.
It is interesting to note, however, when looking at female and male participants separately, a clear difference was found. While the age of female participants did not play a role in the evaluation of male versus female authors, a significant interaction was found for male participants and age. The overall model of the moderation analysis was significant, F(3, 83) = 2.30, p = .035, R2= 0.10, as well as the interaction term, b = -0.04, t(83) = -2.49, p = .
Figure 2. Effect of male participant’s age on the credibility of male versus female authors.
The y-axis denotes the perceived credibility and the x-axis denotes the male participant’s age. The graph displays a visible difference between younger and older male participants with surprising results: the older the male participant, the more credible the female author was evaluated while the male author decreases in credibility.
This tendency was reversed for younger male participants. Moderation: Conservative Social Views
To test whether the role-congruity effect was stronger for participants with more conservative social views, a moderation analysis was conducted with PROCESS (Model 1; Hayes, 2013). The results showed that both the main effect (F(3, 201) = 1.97, p = .120, R2=
0.03) as well as the interaction effect (b = -0.24, t (201) = 1.64, p = .104) were not significant. This indicates that the conservative views of the participants did not have an effect on their
evaluation of the credibility between role-congruent and role-incongruent topics. As a result, Hypothesis 4 was rejected.
Discussion
This study sought to assess whether the perceived credibility of a news piece is conditional on the author’s gender, especially in combination with an article topic that is either congruent or divergent from the gendered role expectations. It was found that role-congruity only applies to female journalists: an article about a typical male topic is seen as significantly less credible if it is written by a woman. This coincides with previous research, where male journalists were consistently rated higher if the topic was considered male (e.g., Goldberg, 1968; Swim et al., 1989). In contrast, the study also showed that a male journalist reporting on a female topic does not suffer from a negative impact on the credibility.
The present study’s findings display a discrepancy in role expectations between women and men in the journalistic field. As female journalists are still more likely to be assigned to specific topics coherent with a female gender role (Cann & Mohr, 2001; North, 2016; Parker, 2004), a divergence from that role is noted by the audience. Male journalists, on the other hand, seem to have less rigid role expectations and are thus more free in what they can report about credibly. The presence of clear role expectations for women has also been supported in other fields, for example in science communication (Knobloch-Westerwick & Glynn, 2013), in leadership roles (Ritter & Yoder, 2004), as well as athletic administration (Burton & Grappendorf, 2011).
Additionally, this study has shown that irrespective of the article’s topic, less
credibility is given to female journalists overall. This tendency has been observed by earlier research, but predominantly in the context of male and female TV newscasters (Brann & Himes, 2010; Weibel et al., 2008). The present study included author photos, and thus,
similarly to TV newscasters, the gender of the author was more visible to the participants. Compared with previous studies that did not include author photos, this strengthened the manipulation of the experiment. As a result, the internal validity of the study was improved. On the other hand, since many real news articles do not include author photos, the external validity was partly compromised. Nevertheless, the clear results display that the audience still regards journalism to be more of a male-dominated profession. This coincides with the minority figures of female journalists in many newsrooms (e.g., Global Media Monitoring Project, 2015b), and clearly displays the need for an increased visibility of female journalists.
Distinguishing between the participants’ gender in the analysis provided interesting results. While previous research led to the assumption that male participants would portray more stereotypical perceptions (e.g, Armstrong & McAdams, 2009), this study has shown the opposite. The results indicate that female participants were more critical towards female journalists—not only in the context of a woman writing an article about a male topic, but also less credibility for female journalists overall. It seems that since Goldberg’s discovery in 1968 – in a journalistic context, women are prejudiced against women – not much has
changed. Albeit women might be consciously aware of gender inequalities, this indicates that implicit biases derived from a patriarchal societal structure are overruling conscious thoughts. These implicit cognitive biases of women against women have been proven in many
experiments surrounding workplace hiring (e.g., Moss-Racusin, Dovidio, Brescoll, Graham & Handelsman, 2012) or women’s self-placement in seemingly gender-incongruent subjects such as mathematics (Nosek, Banaji & Greenwald, 2002).
What is perhaps most surprising about these results is that women showed even more prejudice against themselves than men, although previous studies found both men and women to be similarly prejudiced against women (e.g., Kalra, 2018; Paludi & Bauer, 1983;
Paludi & Strayer, 1985). This may partly be explained by a general tendency of women being severely critical towards each other in other contexts, for example in workplace success (Katsarou, n.d.), or attractiveness (Vaillancourt & Sharma, 2011). However, the circumstance of this experiment was different, as the women were not comparing themselves to a direct female competition. Thus, future research should focus on what other possible reasons might have been at work here.
Contrary to an earlier study by Gross and Geffner (1980), displaying that older participants evaluated articles written by women as less credible, the present study did not confirm this effect. There was no visible difference in age between the role-congruent and role-incongruent stimuli, nor a difference between all articles written by male compared to female journalists. This could potentially be explained by the convenience sample used, which might not be a completely adequate representation of the population. As a result, since very little research has focused on the moderation effect of age, the findings of this study could serve as a starting point for future research focusing on this aspect.
As described above, male participants overall did not evaluate women’s articles to be less credible. However, a visible difference was observed between older and younger male participants. Younger male participants rated the female journalists’ articles less credible than the male journalists’ articles, whereas this relationship was reversed for older male
participants. No previous research has found this relationship so far, as this study was the first to combine two participant characteristics (i.e., gender and age) into one analysis. As a result, the question remains whether this was a singular result due to a potentially skewed sample, or whether similar results could be observed in future studies. Nevertheless, the results are surprising, as one would think that younger men are less influenced by gender stereotypes.
Lastly, this study has not found a relationship between the role-congruity effect and participant’s conservative views. This seems surprising, given evidence from research in gender discrimination that sees a clear relationship between conservatism and more rigid views on traditional gender roles (e.g., Bettencourt et al., 2011; Davis & Greenstein 2009). A factor that might explain the lack of results in the present study is the low reliability of the scale measuring conservative views, leading to less precise estimates, and therefore low statistical power.
Apart from the low reliability of the conservatism scale, several other limitations of the study need to be addressed. First of all, as credibility is such a widely used concept, there are frequent discussions within literature about which are the most valid measurements (e.g., Powell & Wanzenried, 1995). However, the scales of the present study were assessed to have a high reliability and were found valid in other studies (e.g., Gaziano & McGrath, 1986).
Lastly, as mentioned above, the sample of the participants cannot be seen as fully representative of the population. Due to the convenience sample, as well as a limited number of participants, the sample is slightly skewed in terms of education, political orientation and, most importantly age. However, given previous research on factors adding to gender
stereotypes, this makes the study’s findings even more interesting. A sample of comparatively young participants with a higher education level than the average and a more left-wing political attitude is rather expected to be less biased toward female journalists. However, the results point in the opposite direction, thus adding to the strength of the results. This, in turn, helps to shed light on the ambiguity of previous research results, as the bias against female journalists is underlined despite the left-leaning sample.
In conclusion, the study’s findings show that the gender of a journalist can influence the perceived credibility of a news article. Female journalists are not only subject to a more
rigid expectation of which topics coincide with their gender roles, but are also seen as less credible overall. The effect of gender bias against female journalists is especially strong among a female audience as well as among younger men. This gives a rather bleak outlook towards the credibility of the media: According to this study, to increase the credibility, male journalists should be given precedent and continue to write about male topics while female journalists should remain in the background reporting about ‘female’ topics such as fashion.
However, the study’s findings are much more pressing considering its societal
relevance. Given that stereotypes against women, and clearly pinpointed female gender roles are alive and well even amongst surprising groups, the need for change becomes obvious. To change the audience’s perceptions of female journalists, the first step is to increase their presence in the media; not only by balancing the gender ratio in journalism, but also by visibly increasing women’s voices in typically male areas of reporting. This might not only help to combat gender biases in the journalistic field, but also loosen existing gender roles in society overall, given the continued impact of the media on today’s society. That being said, it is duly hoped that if this study is replicated a decade from now, the findings will have
References
Andsager, J. L. (1990). Perceptions of credibility of male and female syndicated political columnists. Journalism Quarterly, 67(3), 485-491. doi:10.1177/107769909006700302 Andsager, J.L. & Mastin, T. (2003). Racial and regional differences in readers’
evaluations of the credibility of political columnists by race and sex. Journalism and
Mass Communication Quarterly, 80(1), 57-72. doi:10.1177/107769900308000105
Armstrong, C. L., & McAdams, M. J. (2009). Blogs of information: How gender cues and individual motivations influence perceptions of credibility. Journal of
Computer-Mediated Communication, 14(3), 435-456. doi:10.1111/j.
1083-6101.2009.01448.x
Armstrong, C.L., Nelson, M.R. (2005). How newspaper sources trigger gender stereotypes. Journalism and Mass Communication Quarterly, 82(4), 820-837. doi:10.1177/107769900508200405
Aune, R. K., & Kikuchi, T. (1993). Effects of language intensity similarity on perceptions of credibility, relational attributions, and persuasion. Journal of Language and Social
Psychology, 12, 224–238. doi:10.1177/0261927X93123004
Baruch, G.K. (1972). Maternal influences upon college women’s attitudes towards women and work. Developmental Psychology, 6, 32-37. doi:10.1037/h0032220
Berlo, D. K., Lemert, J. B., & Merz, R. J. (1970). Dimensions for evaluating the acceptability of message sources. Public Opinion Quarterly, 33, 563–576. doi:10.1086/267745
Bettencourt, K. E. F., Vacha-Haase, T., & Byrne, Z. S. (2011). Older and younger adults’ attitudes toward feminism: The influence of religiosity, political orientation, gender, education, and family. Sex Roles, 64, 863-874.
doi:10.1007/s11199-011-9946-z
Boateng, K. J. A. (2017). Reversal of gender disparity in journalism education-Study of Ghana Institute of Journalism. Observatorio (OBS*), 11(2), 118-135.
Bolzendahl, C. I., & Myers, D. J. (2004). Feminist attitudes and support for gender equality: Opinion change in women and men, Social Forces, 83, 759-789. doi:10.1353/sof.2005.0005
Brann, M., & Himes, K. L. (2010). Perceived credibility of male versus female television newscasters. Communication Research Reports, 27(3), 243-252.
doi:10.1080/08824091003737869
Burkhart, F. N., & Sigelman, C. K. (1990). Byline bias? Effects of gender on news article evaluations. Journalism Quarterly, 67(3), 492-500.
https://doi.org/10.1177/107769909006700303
Burton, L. J., Grappendorf, H., & Henderson, A. (2011). Perceptions of gender in athletic administration: Utilizing role congruity to examine (potential) prejudice against women. Journal of Sport Management, 25(1), 36-45.
doi:10.1179/1935739714Z.00000000016
Cann, D.J., & Mohr, P.B. (2001). Journalist and Source Gender in Australian Television News, Journal of Broadcasting & Electronic Media, 45(1), 162-174.
doi:10.1207/s15506878jobem4501_10
Carney, D. R., Jost, J. T., Gosling, S. D., & Potter, J. (2008). The secret lives of liberals and conservatives: Personality profiles, interaction styles, and the things they leave behind. Political Psychology, 29, 807–840. doi:10.1111/j.1467-9221.2008.00668.x Cichocka, A., Bilewicz, M., Jost, J.T., Marrouch, N., & Witkowska, M. (2016). On the
grammar of politics – or why conservatives prefer nouns. Political Psychology 37(6), 799-815. doi:10.1111/pops.12327
Chobot, D., Goldberg, P.A., Abramson, I.M., & Abramson, P.R. (1974). Prejudice against women - replication and extension. Psychological Reports, 35, 478. doi:10.2466/ pr0.1974.35.1.478
Dambrot, F. H., Papp, M. E., & Whitmore, C. (1984). The sex-role attitudes of three generations of women. Personality and Social Psychology Bulletin, 10, 469-473. doi:10.1177/0146167284103015
Davis, S. N., & Greenstein, T. N. (2009). Gender ideology: Components, predictors, and consequences. Annual Review of Sociology, 35, 87-105. doi:10.1146/annurev-soc-070308-115920
Djerf-Pierre, M. (2007). The Gender of Journalism: The Structure and Logic of the Field in the Twentieth Century. Nordicom review, 28, 81-104.
Donovan, J. (1992). Feminist Theory: The Intellectual Traditions of American
Feminism. New York: Continuum.
Eagly, A. H., & Karau, S. J. (2002). Role congruity theory of prejudice toward female leaders. Psychological Review, 109, 573-598. doi:10.1037/0033-295X.109.3.573 Eastin, M. S. (2001). Credibility assessments of online health information: The effects of
source expertise and knowledge of content. Journal of Computer-Mediated
Communication, 6(4). doi:10.1111/j.1083-6101.2001.tb00126.x
Edström, M. (2017). Is there a nordic way? A Swedish perspective on achievements and problems with gender equality in newsrooms. Medijske studije, 2(3), 64-74. Eisend, M. (2006). Source credibility dimensions in marketing communication—a
generalized solution. Journal of Empirical Generalizations in Marketing Science, 10, 1–33.
Engstrom, E., & Ferri, A.J. (2000). Looking Through a Gendered Lens: Local U.S. Television News Anchors' Perceived Career Barriers, Journal of Broadcasting &
Electronic Media, 44(4), 614-634. doi:10.1207/s15506878jobem4404_6
Everett, J. A. (2013). The 12 item social and economic conservatism scale (SECS). PLoS
One, 8(12), e82131, 1-11. doi:10.1371/journal.pone.0082131
Fairness and Accuracy in Reporting. (2001). Who’s on the news? Study shows network
sources skew white, male, and elite. Retrieved from http://www.fair.org/index.php?
Flanagin, A. J., & Metzger, M. J. (2000). Perceptions of Internet information credibility.
Journalism & Mass Communication Quarterly, 77(3), 515-540.
doi:10.1177/107769900007700304
Gallivan, J. (1991). Gender bias in students' ratings of essays. Journal of Social Behavior &
Personality, 6(1), 119-124.
Gaziano, C., & McGrath, K. (1986). Measuring the concept of credibility. Journalism
Quarterly, 6(3), 451–462. doi:10.1177/107769908606300301
Global Media Monitoring Project. (2015a). GMMP 2015 Reports. Retrieved from http://whomakesthenews.org/gmmp-2015.
Global Media Monitoring Project. (2015b). Global Media Monitoring Project 2015 -
National Report: Germany. Retrieved from http://cdn.agilitycms.com/who-makes
-the-news/Imported/reports_2015/national/Germany.pdf
Golan, G.J. (2010). New Perspectives on Media Credibility Research. American Behavioral
Scientist, 54(1), 3–7. doi:10.1177/0002764210376307
Gross, M. M., & Geffner, R. A. (1980). Are the times changing? An analysis of sex-role prejudice. Sex Roles, 6(5), 713-722.
Harp, D., & Tremayne, M. (2006). The gendered blogosphere: Examining inequality using network and feminist theory. Journalism & Mass Communication Quarterly,
83(2), 247-264. doi:10.1177/107769900608300202
Helwerth, U. (2002). Einschalt-Quote: Zur Geschlechterdemokratie in den Medien [Audience ratings: About gender democracy]. Retrieved from http://
www.womanticker.net/content.php
Hovland, C. I., Janis, I. L., & Kelley, H. H. (1953). Communication and persuasion. New Haven, CT: Yale University Press.
Johnson, T. J., & Kaye, B. K. (1998). Cruising is believing?: Comparing Internet and traditional sources on media credibility measures. Journalism & Mass
Communication Quarterly, 75(2), 325-340. doi:10.1177/107769909807500208
Jost, J. T., Glaser, J., Kruglanski, A.W., & Sulloway, F. (2003). Political conservatism as motivated social cognition. Psychological Bulletin, 129, 339–375.
Katsarou, M. Women & the Leadership Labyrinth: Howard vs Heidi. Retrieved from http://www.leadershippsychologyinstitute.com/women-the-leadership-labyrinth-howard-vs-heidi/
Kalra, P. (2018). Curbing Gender Bias in Journalistic Work: The Role of Awareness (Unpublished Graduate Thesis). University of Amsterdam.
Kiecolt, K. J., & Acock, A. C. (1988). The long-term effects of family structure on gender-role attitudes. Journal of Marriage and the Family, 50,709-717.
Knobloch-Westerwick, S., & Glynn, C. J. (2013). The Matilda effect --Role congruity effects on scholarly communication: A citation analysis of Communication Research and
Journal of Communication articles. Communication Research, 40(1), 3-26. doi:10.1177/0093650211418339
Knobloch-Westerwick, S., Glynn, C. J., & Huge, M. (2013). The Matilda effect in science communication: an experiment on gender bias in publication quality perceptions and collaboration interest. Science Communication, 35(5), 603-625.
doi:10.1177/1075547012472684
Levenson, H., Burford, B., Bonno, B., & David, L. (1975). Are women still prejudice against women? A replication and extension of Goldberg’s study. Journal of Psychology,
89, 67-71. doi:10.1080/00223980.1975.9923907
Locksley, A., Borgida, E., Brekke, N., & Hepburn, C.(1980). Sex stereotypes and social judgement of individuals. Journal of Experimental Social Psychology, 18, 23-42. doi:10.1037/0022-3514.39.5.821
Lottes, I. L., & Kuriloff, P. J. (1992). The effects of gender, race, religion, and political orientation on the sex role attitudes of college freshmen. Adolescence, 27(107), 675. Lumby, C. (1994). Feminism and the media: The biggest fantasy of all. Media Information
Australia, 72, 49–54. doi:10.1177/1329878X9407200108
Malka, A., Lelkes, Y., & Soto, C. J. (2017). Are cultural and economic conservatism positively correlated? A large-scale cross-national test. British Journal of Political
Science, 1-25. doi:10.1017/S0007123417000072
Meyer, P. (1988). Defining and measuring credibility of newspapers: Developing an index.
Journalism Quarterly, 65, 567–574. doi:10.1177/107769908806500301
Miller, G. R., & McReynolds, M. (1973). Male chauvinism and source competence: A research note. Speech Monographs, 40(2), 154-155.
(2012). Science faculty’s subtle gender biases favor male students. Proceedings of
the National Academy of Sciences, 109(41), 16474-16479.
North, L. (2016). The gender of “soft” and “hard” news, Journalism Studies, 17(3), 356-373. doi:10.1080/1461670X.2014.987551
Nosek, B. A., Banaji, M. R., & Greenwald, A. G. (2002). Math= male, me= female, therefore math≠ me. Journal of personality and social psychology, 83(1), 44-59.
Nosek, B. A., Smyth, F. L., Sriram, N., Lindner, N. M., Devos, T., Ayala, A., & Greenwald, A.G. (2009). National differences in gender-science stereotypes predict national sex differences in science and math achievement. Proceedings of the National
Academy of Sciences, 106, 10593-10596. doi:10.1073/pnas.0809921106
Ohanian, R. (1990). Construction and validation of a scale to measure celebrity endorsers’ perceived expertise, trustworthiness, and attractiveness. Journal of
Advertising, 19(3), 39–52. doi:10.1080/00913367.1990.10673191
Oláh, L. S., Kotowska, I. E., & Richter, R. (2018). The new roles of men and women and implications for families and societies. In A Demographic Perspective on Gender,
Family and Health in Europe (pp. 41-64). Springer, Cham.
Orwant, J.E., & Cantor, M. (1977). How sex stereotyping affects perceptions of news preferences. Journalism Quarterly, 54(1), 99 108, 139. doi:
10.1177/107769907705400114
Paludi, M. A., & Bauer, W. D. (1983). Goldberg revisited: What's in an author's name. Sex
Roles, 9(3), 387-396.
Paludi, M. A., & Strayer, L. A. (1985). What's in an author's name? Differential evaluations of performance as a function of author's name. Sex Roles, 12(3/4), 353-361. Parker, E. (2004). Does a news anchor’s gender influence audience evaluations of
the anchor? Paper presented at the Association for Education in Journalism and
Mass Communication, Toronto, Canada.
Powell, F. C., & Wanzenried, J. W. (1995). Do current measures of dimensions of source credibility produce stable outcomes in replicated tests? Perceptual & Motor Skills,
81(2), 675–87. doi:10.1177/003151259508100263
Reysen, S. (2005). Construction of a new scale: The Reysen Likability Scale. Social
Behavior and Personality, 33(2), 201-208. doi:10.2224/sbp.2005.33.2.201
Rimmer, T., & Weaver, D. (1987). Different questions, different answers? Media use and media credibility. Journalism Quarterly, 64(1), 28-44. doi:
10.1177/107769908706400104
Ritter, B. A., & Yoder, J. D. (2004). Gender differences in leader emergence persist even for dominant women: An updated confirmation of role congruity theory. Psychology of
Women Quarterly, 28(3), 187-193. doi:10.1111/j.1471-6402.2004.00135.x
Ross, K., Boyle, K., Carter, C., & Ging, D. (2018). Women, men and news: it’s life, Jim, but not as we know it. Journalism Studies, 19(6), 824-845. doi:10.1080/1461670X. 2016.1222884
Sebba, A. (1994). Battling for News: Rise of the woman reporter. London: Hodder & Stoughton.
Shaw, D. L., Cole, L. P., Moore, R. L., & Cole, R. R. (1981). Men versus women in bylines.
Journalism Quarterly, 58(1), 103–106. doi:10.1177/107769908105800118
Smyth, J. (2008). Transcending Traditional Gender Boundaries: Defining Gender Roles Through Public and Private Spheres. Elements, 4(1), 28-34. doi:10.6017/
eurj.v4i1.9010
personality attributes and sex-role attitudes across age groups. Developmental
Psychology, 15, 583-584. doi:10.1037/h0078091
Steiner, L. (2012). Failed theories: Explaining gender difference in journalism. Review of
Communication, 12(3), 201-223. doi:10.1080/15358593.2012.666559
Strong, C. & Hannis, G. (2007). The visibility of female journalists at Australian and New Zealand newspapers: the good news and the bad news. Australian Journalism Review,
29(1), 115-126.
Surawski, M. K., & Ossoff, E. P. (2006). The effects of physical and vocal attractiveness on impression formation of politicians. Current Psychology, 25(1), 15-27.
Swim, J., Borgida, E., Maruyama, G., & Myers, D. G. (1989). Joan McKay versus John McKay: Do gender stereotypes bias evaluations?. Psychological Bulletin, 105(3), 409.
Thorson, E., Meyer, H., Denton, F., & Smith, J. (2008). Prediction of print and online newspaper readership from indices of news quality: Annual meeting of the
International Communication Association, Montreal, Canada: TBA.
Tosi, H. L., & Einbender, S. W. (1985). The effects of the type and amount of information in sex discrimination research: A meta-analysis. Academy of Management Journal,
268, 712-723. doi:10.5465/256127
Twenge, J. M. (1997). Attitudes toward women, 1970–1995: A meta-analysis. Psychology
of Women Quarterly, 21(1), 35-51. doi:10.1111/j.1471-6402.1997.tb00099.x
Usher, N., Holcomb, J., & Littman, J. (2018). Twitter makes it worse: Political journalists, gendered echo chambers, and the amplification of gender bias. The International
Journal of Press/Politics, 23(3), 324-344. doi:10.1177/1940161218781254
among women. Aggressive behavior, 37(6), 569-577. doi:10.1002/ab.20413 Weibel, D., Wissmath B., & Groner, R.(2008). How gender and age affect newscasters’
credibility—An investigation in switzerland, Journal of Broadcasting & Electronic
Media, 52(3), 466-484. doi:10.1080/08838150802205801
West, M. D. (1994). Validating a scale for the measurement of credibility: A covariance structure modeling approach. Journalism Quarterly, 71, 159–168.
doi:10.1177/107769909407100115
White, H. A., & Andsager, J. L. (1991). Newspaper column readers' gender bias: Perceived interest and credibility. Journalism Quarterly, 68(4), 709-718.
doi:10.1080/08824091003737869
Willetts-Bloom, M. C., & Nock, S. L. (1994). The influence of maternal employment on gender role attitudes of men and women. Sex Roles, 30, 371-389.
doi:10.1007/BF01420599
Women’s Media Center. (2017). The Status of Women in the US Media: 2017. Retrieved from http://www.womensmediacenter.com/assets/site/reports/10c550d19ef9f368 8f_mlbres2jd.pdf.
Appendix Stimulus Material
Figure 3. Stimulus 1: Tesla article with a male author (Role-congruent)
Figure 5. Stimulus 3: H&M article with a male author (Role-incongruent)
