Act like a Lady, think like a Man
How coverage on (fe)male politicians affects attitudes towards gender equality
Master’s Thesis
Graduate School of Communication Master’s Programme Communication Science
University of Amsterdam
Author: E. L. van Wulfften Palthe Student Number: 5875692 Date of Completion: 30-01-2015
Abstract
This study attempts to expand the research on gender equality in politics, by focusing on the different representation of male and female politicians in the media. By means of an experimental design, the effects on evaluation of politicians and gender equality are measured, with stimuli portraying a male and female politician with feminine and masculine
messages. The results show that female politicians with masculine traits are overall best evaluated, yet also cause a decrease in perceived gender equality. All in all, this research tries
to bridge the fields of gender, politics and media.
Keywords:
Politics, media, gender, stereotypes, issue capability, gender equality
Acknowledgements:
I would like to thank my supervisor during the writing of my thesis, Magdalena Wojcieszak, who has instigated my craze about women in politics. Her enthusiasm, help and fresh insights
have made this study very challenging, but foremost an inspiring process, which has caused me to consider a professional career in working with (fe)male politicians who are struggling
“Act like a lady, think like a man.” - Steve Harvey
Introduction
The above relationship advice given by the American comedian and author Steve Harvey could also be of use for female politicians. Up until recent years there was hardly any place for women in politics (Braden, 1996; Connell, 2005; Herrnson, Lay & Stokes, 2003). Notably, this traditional understanding of politics as a man’s world still has its realm in society nowadays. Moreover, it appears that women in politics are still seen as novelties, as their gender is repeatedly underlined in the media (Cantrell & Bachmann, 2008). More often than not, female candidates are asked how they will combine motherhood and being a politician at the same time. Apparently, the long-established idea of politics as a masculine field brings gendered expectations (Connell, 2005).
Besides this realm in society due to the former roles of men and women, there are other factors that strengthen these expectations. It is namely the media that have been found to play an important role in magnifying gendered stereotypes. This is a worrisome phenomenon, as gender stereotyped media coverage affects the voters evaluation of politicians, as well as voters attitudes towards gender equality (Connell, 2005;Inglehart & Norris, 2003; Terkildsen & Schnell, 1997).
This thesis draws upon existing literature on the representation of women in politics, and through an experimental design examines the cognitive effects of stereotyped media coverage of politicians on attitudes towards gender equality. The main research question is as follows: To what extent do gender stereotyped campaign messages affect the voter’s
evaluations of politicians and their attitudes towards gender equality? This study then
expands on existing research regarding gendered media coverage of politicians (Braden, 1996; Connel, 2005; Devitt, 2002; Niven & Zilber, 2001; Kahn, 1992) by taking it a step
further and examines how this coverage affects attitudes towards gender equality. Gender, political trust and political ideology are used as independent variables. The evaluation of the politician, perceived issue ownership, attitudes towards gender equality and political
participation are the dependent variables for this research. The results can ultimately offer suggestions for struggling (female) politicians on how to steer clear of media resistance when climbing the political ladder. Furthermore, this study offers a greater understanding into the complexity of women in politics, and provides more insight into how media coverage impacts notions of gender equality.
Theoretical Framework
Historically, there has been exclusion of women in politics. Stereotypically, men used to be the supposed ‘protectors and breadwinners’ whereas women were the ‘caretakers’ (Heldman, Caroll & Olson, 2005; Herrnson et al., 2003; Major & Coleman, 2008). In the Netherlands, it has been possible for women to get elected into Parliament since the year 1917 (even before women’s vote rights in 19221), yet the number of women in government remained drastically low. It was only after World War II that women gained more active roles in the work force, including the political field (Braden, 1996; Kleinberg, 1988). From 1977 to 1991 there has been a slow increase of 20 to 40 women in Dutch Parliament (out of 150 members). Currently, in the year 2015, a bit over one third of Parliament is female (56 out of 150)2.
Although the previous century has seen a vast increase of women in Parliament, the idea of politics as a man’s business still stands foot in contemporary society. Women are still in a pressing minority and the traditional stereotypes with men as breadwinners and women as caretakers still persist. However, women are now increasingly trying to overcome the
long-1Retreived January 25, 2015 from http://www.parlement.com/id/vh8lnhrre0zv/vrouwen_in_de_tweede_kamer. 2Retreived January 25, 2015 from http://www.nieuwsdossier.nl/politiek/stemrecht-voor-vrouwen-in-nederland
standing tradition of politics as a man’s world, and are trying to find ways to fit themselves into this still predominant masculine field of work (Cantrell & Bachmann, 2008).
Media coverage
The problem for women however does not lie in restrictions to get elected in to high office, more so the problem lies in media coverage about female politicians(Cantrell & Bachmann, 2008; Devere & Davies, 2006; Devitt, 2002). It is in particular the framing of female politicians which is problematic. Frames are story patterns that editors and reporters use, in order to structure public opinion or attribute a certain meaning to a story (Gamson, 1992; Terkildsen & Schnell, 1997). For this reason, it is important to research the effects that framing can have on attitudes towards female politicians and gender equality.
Although it is not always visible at first sight, female politicians are structurally gendered through stereotypical frames in television, radio and newspapers. This is not surprising, as the media still often use the traditional frames which are built around the dominance of men (Fountaine & McGregor, 2002). Research has shown that female politicians are taken less seriously and are given less attention in the media (Cantrell & Bachmann, 2008; Devitt, 2002; Kahn, 1992; Niven & Zilber, 2001). Furthermore, the distinction between men and women is often exaggerated by the media, even though most political gender stereotypes have no foundation in actual professional differences (Niven & Zilber, 2001).
These stereotypical frames however are not solely based on the editorial preference of the media (Terkildsen & Schnell, 1997) as journalists can choose from an array of
perspectives provided by the politician on which the frame is created (Gamson, 1992). Thus although the mass media are very powerful in shaping public perceptions, political institutions still hold substantial power over the political representation (Gallagher, 2001; Mazzoleni &
Schulz, 1999). For this reason, the relationship between politicians and media can be described as a ‘tango’, in which politicians and media constantly interact when it comes to who leads the coverage (Strömbäck & Nord, 2006). However, it does not seem that journalists bias the coverage purposely; rather the coverage shows that the gendering of politicians is deeply rooted in contemporary news media (Major & Coleman, 2008).
Stereotypical frames
An analysis of news coverage has shown that male candidates get more coverage on stereotypical masculine issues such as economy and defense (Heldman, Caroll & Olson, 2005; Herrnson et al., 2003; Major & Coleman, 2008), whereasfemale candidates get more coverage on stereotypical feminine issues such as abortion, education and healthcare (Jalalzai, 2006; Kahn, 1994; Major & Coleman, 2008). More often than not, news media tend to focus more on the personal traits and appearance of a female politician, in compared to their male counterparts. Also, men are portrayed as being more competent and knowledgeable on issues and they receive more attention to their accomplishments and political stance (Braden, 1996; Devitt, 2002). From the public’s perspective, women are then more likely to be seen as bench warmers, rather than integral part of government (Braden, 1996).
Regarding stereotypical masculine and feminine traits, the differentiation can be made as follows. Men are more characterized by individualism, toughness, assertiveness and
strength (Devere and Davies, 2002; Hayes, 2005), and their work is characterized by controversy and conflict (Braden, 1996). Women on the other hand portray images of compassion and empathy (Hayes, 2005). Also women portray more warmth (Huddy and Terkildsen, 1993), and their work ethic is characterized by compromise and collaboration (Braden, 1996). For the experiment conducted in this study the stereotypical masculine issues
and traits are then operationalized into a ‘masculine message’, and the stereotypical feminine issues and traits are then operationalized into a ‘feminine message’.
This gendering of politicians in the media is problematic, as frames do not only shape considerations voters use to evaluate policies, these frames are also capable of shifting attitudes towards the politician (Iyengar, 1994; Terkildsen & Schnell, 1997). This is an important notice, as voters use character traits and issues of candidates to make political decisions. It is especially problematic, since a politician’s media presentation has become increasingly important in the last decades (Iyengar, Norpoth & Hahn, 2004). Politicians need favorable news coverage of their messages in order to gain support for their policies (Kittilson & Fridkin, 2008), especially because there is hardly any personal contact with politicians anymore (Hitchon, Chang & Harris, 1997). For this reason, more than ever, politicians need to carefully consider their media strategies as voters mainly cast their vote based on the
information they retrieve from the media (Scharrer, 2013; Terkildsen & Schnell, 1997).
Hypotheses
Stimuli and evaluation. The stereotypical media coverage is obstructing the route of women’s leadership, as the imagery is inconsistent with the image female politicians need to project in order to successfully compete with their male counterparts (Devitt, 1999; Huddy & Terkildsen, 1993). This is exemplified by the disadvantage caused for women as voters generally prefer masculine over feminine traits in their political leaders (Hayes, 2011; Huddy & Terkildsen, 1993a, 1993b; McGinley, 2009). Also masculine issues are seen to be more important than feminine issues (Rosenwasser & Seale, 1988; Kahn, 1992). For this reason it is interesting to examine how politicians are evaluated, after being exposed to politicians with stereotypical masculine and feminine messages. Taking the above theory into account, it is
hypothesized that politicians with a more masculine message are overall better evaluated than politicians with a feminine message (H1a).
To elaborate a bit more on the evaluation, research has also shown that women are perceived to be more trustworthy and honest (Dolan, 2005). Therefore it is hypothesized that female politicians are evaluated higher on trustworthiness and honesty than the male
politicians (H1b). Yet on the other hand, men are portrayed as better leaders (Braden, 1996; Dolan, 2005; Niven and Zilber, 2001), thus it is hypothesized that male politicians are better evaluated on leadership than female politicians (H1c). This is then useful for politicians to know when focusing on certain traits.
Based on these stereotypical expectations, it is also possible to make a prediction on what traits work best for male and female politicians. On the one hand, research has shown that women have encountered much resistance when they break out of their stereotypical roles (Devere & Davies, 2006; Devitt, 2002). Yet on the other hand, research has shown that
women who portray a strong ‘masculine’ message, gain wider support amongst both men and women (Eagly & Mladinic, 1994). Thus taking into account the generally high likeability of women (Devere & Davies, 2006), and the preference for masculine traits, it is hypothesized that a female politician with a masculine message is overall best evaluated (H1d).
Political ideology and evaluation. It should be noted however that there are several factors that can have an influence on how politicians are evaluated. One of these factors is political ideology. This is a factor that political parties bear this in mind when presenting a top candidate for elections (Hayes, 2011). The political ideology variable for this study can be divided into two categories. The first category is the economical ideology which relates to welfare distribution (left to right) and the second category is social ideology which relates to social values (liberal to conservative) (Jost, Federico & Napier, 2009). It is theorized that
women are more frequently elected by liberals than by conservatives (McGinley, 2009; Sanbonmatsu, 2004). Following up on this theory, it is hypothesized that liberals evaluate the female politician higher than conservatives (H2a). This could be a reason for politicians to make strategic decisions on how to best present female candidates for voters with different ideologies (Dolan, 1998).
Trust and evaluation. The second factor that could influence the evaluation is political trust. Research has pointed out that emphasizing feminine issues, such as education and healthcare, can actually be beneficial if voters are looking for a change from the status quo (Dolan, 1998, Herrnson et al., 2003; Paolino, 1995). This preference for female
representatives is mostly present for voters with low trust in political institutions (High & Pippert, 1998). Hence for this group, the female candidate would actually be better evaluated than the male candidate. Taking this theory into account, it is hypothesized that voters with low trust in politics will evaluate the female politician higher than the male politician (H2b). Political trust then serves as the independent variable and evaluation as the dependent.
Gender and evaluation. Research has also shown than female voters are actually more likely to be supportive of women in politics than men (Dolan, 1998; Williams, 1994).
Therefore, it is hypothesized that women evaluate a female candidate with higher than men do (H2c). Gender serves here as the independent and the evaluation as the dependent variable.
Political participation. Furthermore, research shows that female voters who are represented by a female politician, actually show higher political participation than female voters who are represented by a male politician (Dolan, 1998, Herrnson et al., 2003; High & Pippert, 1998; Paolino, 1995). Based on this notion, it is hypothesized that female voters are more inclined to cast their vote in elections after being exposed to the female politician in compared to the male politician (H3). Gender serves here as the independent, and intended voting behavior server as the dependent variable. The latter three hypotheses can thus test
how female politicians can use their femininity as an advantage, by targeting women, and people with low trust or political participation.
Issue capability. Apart from the evaluations of the politician, it is also tested how voters perceive the politicians issue capability, as news media tend to portray men and women as owning certain political issues. In this manner the concept of gender ownership is
developed (Herrnson et al., 2003). It is the notion that men and women are seen as better handling certain political issues. This capability can be described as how competent, qualified and experienced the candidate appears (Louden & McCauliff, 2004). Taking this theory into account, it is hypothesized that male politicians are seen as more capable of handling
stereotypical issues of economy and defense than female politicians (H4a). Adding to this, it is hypothesized that female politicians are seen as more capable of handling the stereotypical topics of healthcare, education and equal opportunities for men and women than male
politicians (H4b). Issue capability thus serves as the dependent variable. Furthermore, it is hypothesized that women rate female politicians higher on capability than men do (H4c).
Gender equality. Besidesthe effects that gendered stereotypes could have on evaluations of the politicians, intended voting behavior and perceptions of issue capability, the exposure to either a stereotypical male of female politician could also have an influence on attitudes towards political gender equality. In this study, gender equality is divided into two categories. The first category is the notion that women are just as capable as men on handling political issues. The second category withholds gender equality in government (Kabeer, 2005). In the latter category the equal division of menand women in government is addressed.
Although opinions on gender equity have liberalized, it is not sure in what direction this is going (Bolzendahl & Myers, 2004). Furthermore, being exposed to a female politician who portrays stereotypical masculine ‘strong’ and ‘important’ issues could increase support
for gender equality because it blurs the traditional division between men and women (Eagly & Mladinic, 1994). Therefore it is hypothesized that being exposed to a female politician with a masculine message generates the most positive attitudes towards gender equality (in
capability) (H5a). The second category of gender equality also functions as a dependent variable, by hypothesizing that being exposed to a female politician with a masculine message generates the most positive attitudes towards gender equality in government (eg. equal
division of men and women) (H5b).
Political ideology and gender. Furthermore, it is interesting to add political ideology as an independent variable, as conservatives are generally less inclined to be favorable of an equal division of men and women in government (Hogan, 2010). However, this attitude could change after being exposed to a more masculine female (Herrnson, Lay & Stokes, 2003). Henceit is hypothesized that support among conservatives for a more equal division of men and women in government is higher, after being exposed to a female politician with a
masculine message instead of a female with a feminine message (h5c).The above hypotheses thus test what the effects are on attitudes towards gender equality, after being exposed to a male or female politician, portraying a masculine or feminine message.
Methodology
To test the hypotheses a 2 (gender: male versus female) x 2 (message: masculine versus feminine) factorial experiment was designed and conducted on a sample of Dutch participants above 18. The experiment was performed using Qualtrics, a platform where surveys can be created and distributed. An online survey was administered during the first two weeks of December 2015. Overall, out of 279 who people started the survey, 86% finished it, resulting in the final sample of N=239. The participants were recruited through the network of the author, and during this process it was made sure that participants would be of a variety of
gender, age and education to make the sample as representative as possible. This was assured by monitoring the responses of the survey and accordingly adapting the distribution of the survey to this. The questionnaire was distributed among different socio-demographic groups, such as students, a variety of Dutch companies and online networks (eg. LinkedIn).
Participants
The respondents in the final sample were 53% female, with the average age of 39 years (SD=15.9). Furthermore 50% of the participants own a University degree and 30% a HBO degree (higher education). Almost 48% of the sample was employed either full-time or part-time, 22% were students and 16% was self-employed. See appendix 1 for all socio-demographic characteristics.
Procedure
Each participant was randomly assigned to one out of four conditions. Each of the conditions got an almost equal amount of participants (Male politician with a masculine
message: 25.1%; Male politician with a feminine message; 24.7%; Female politician with a feminine message: 26.4%; Female politician with a masculine message: 23.8%). A
comparison of the four stimuli groups shows that there are no significant differences between the groups regarding gender (x2(3)=.23, p=.828), age (F(3)=.30, p=.823) or education
(x2(3)=6.22, p=.294). This points to successful randomization. After exposure to one of the stimuli, participants filled out a short questionnaire.
Stimuli
For this study four fictional stimuli were created, with a male and female politician both portraying a stereotypical masculine and stereotypical feminine campaign message. The stimuli were created as if they were an article in print media. Although coverage for political candidates is also widely broadcast on television, newspaper coverage influences voter’s
knowledge on politicians more than television does (Bystrom, 2004). For this reason, the stimulus is created as if it was a newspaper article, and this is also what the participants were told when seeing the stimulus. See appendix 2 for stimuli with photos and messages.
Gender of the politician. The first factor manipulated in this experiment is the gender of the politician (male or female). To create the stimuli, two photos of a male and female were selected from an online image database. To eliminate the risk that the results were influenced by the appearance of the politician, not by gender per se, a pre-test was performed prior to the actual survey. Eight pictures (4 men and 4 women) were shown in the pre-test and
respondents (N=47) had to judge each of them on likeability, competence, confidence,
trustworthiness, attractiveness, knowledge and leadership on a scale from not at all (1) to very
much (5). Also it was checked whether the respondent recognized the person in the picture, and if so this picture got eliminated from the analysis. A comparison of means of the total evaluation disclosed one man and one woman who scored almost equal on the evaluation (man: M=2.95, SD=.79, woman: M=2.94, SD=.61). Thus these two pictures were chosen for the actual experiment.
Message. The second factor manipulated in this study was the political message of the candidate (masculine or feminine). The stereotypical male text portrays issues of economy, defense and international relations. The traits that are written are assertiveness, efficiency, decision making qualities and short-term planning. The portrayed work ethics are controversy and conflict (Braden, 1996; Devere & Davies, 2002; Dolan, 2005).
The stereotypical feminine message presents issues of healthcare, education and creating jobs for the youth. The female traits that are presented show are warmth, honesty, listening, empathy and long-term planning. Furthermore, the work ethics portray compromise and collaboration (Braden, 1996; Devere & Davies, 2002; Dolan, 2005).
It was made sure that both the masculine and feminine text was of same length (100 words) and presented the same amount of issues (3) and traits (6). Besides to this, to enhance treatment equivalence, all stimuli were similar in lay-out and design.
Measures
After exposure to the stimulus, participants were asked to state their attitudes towards several concepts, which function as the dependent variables. See appendix 1 for all variables.
Evaluation. Firstly, participants were asked to evaluate the politician on a 7-point scale from not at all (1) to very much (7). All the specific evaluations (authenticity,
trustworthiness, competence, leadership, honesty, confidence and integrity) were grouped into
one variable called evaluation (α=.85, M=4.29, SD=.92).
Issue capability. Secondly, perceptions towards capability of handling issues (such as economy or healthcare) for men and women was measured with seven items (eg.: ‘How capable are men on handling the issue of economy/how capable are women on handling the issue of economy’). This resulted in two variables being capability men (α=.94, M=4.05, SD=.71) and capability women (α=.95, M=4.13, SD=.69).
Gender equality. Thirdly, participants were asked to rate on a scale from strongly disagree (1) to strongly agree (7) how much they agreed with four statements regarding gender equality (eg.: ‘Women are just as capable as men to be successful in politics’). The four items together then account for the variable of gender equality (α=.74, M=6.06, SD=.93). On the same scale, participants were asked to indicate their agreement towards more women in government (eg.: ‘The introduction of gender quotas is important for the equal
representation of men and women in government’). Four of these items form the variable
Intended voting behavior. Furthermore, intended political participation was measured with three items (eg. the likeliness to share political information with friends or family), yet the Cronbach’s α score is too low (resp. α=.66 and α=.50) to use this variable for valid conclusions. Yet it is still valuable to include the item of intended voting behavior (M=6.72, SD= .89), as this is generally the most known form of political participation. Analysis shows that the item is normally distributed.
Moderators/independent variables
Apart from the dependent variables, several factors were used as moderators or independent variables, apart from the socio-demographic variables.
Political trust. The first moderating variable is political trust. This is trust towards both politicians and institutions, and was measured on a scale from strongly disagree (1) to strongly agree (7) with statements such as ‘I have trust in Dutch politicians’. The items together form the variable of political trust (α=.89, M=4.10, SD=1.32).
Political ideology. Furthermore, political ideology was measured both economically and socially. The average political ideology of the participants regarding economy is
somewhat center-right (M=4.18, SD=1.35) (measured on a 7-point scale from left (1) to right (7). The average political ideology of the participants regarding social issues is somewhat liberal (M=2.71, SD=1.23) (measured from liberal (1) to conservative (7).
Results
First it was tested what main effects the stimuli have on attitudes towards evaluation,
gender equality and issue capability. All ANOVA’s controlled for gender, age and education,
as women, as well as highly educated and young people are usually more supportive of gender equality (Bolzendahl & Myers, 2004). The models found two main effects. Firstly,
there is a significant difference for evaluation after exposure to the four conditions
(F(3)=4.48, p<.01). The female politician with the masculine message is significantly best evaluated (F(3)=4.48, p<.01). A post hoc Tukey test shows that the female politician with the masculine text (M=4.56, SD=.92) is even higher evaluated than the male politician with the masculine text (Mdifference= -.54) and the male politician with the feminine text (Mdifference= .39). This means H1d can be accepted. See table 1 and 3.
The second main effect of the stimuli is on attitudes towards gender equality. The model shows that there is a significant difference in the attitude towards gender equality after exposure to the stimuli (F(3)=4.60, p<.01). A post-hoc Tukey test shows that the participants attitude towards gender equality is lower after seeing the female politician (M=5.71,
SD=1.17) compared to the male politician with the masculine text (Mdifference= .49) and the male politician with the feminine text (Mdifference= .56). This means that after being exposed to the female politician with the masculine text, the participants actually think women are less equally capable as men to work in politics. Hence the prediction with H5a that attitudes towards gender equality would positively increase after being exposed to the female politician with the masculine text should be rejected. Yet there was a significant effect in the opposite direction. See table 2 and 3.
Male politician + Masculine message Male politician + Feminine message Female politician + feminine message Female politician + masculine message Man 4,03 4,03 4,24 4,52 Woman 4 4,29 4,65 4,59 Total 4,02 4,16 4,45 4,56 3,8 3,94 4,1 4,2 4,3 4,4 4,5 4,6 4,7 M ea ns
Furthermore there was no support for H5b, which predicted that seeing the female politician with the masculine text would increase support for gender equality in government, yet there were no significant results (F(3)=.034, p= .99). Also the stimuli had no significant effect on the perceived capability of men and women (F(3)=1.10, p= .35). See appendix 3 for the correlations table. This means both H4a and H5b were rejected, as the perceived issue capability did not change after being exposed to the stimuli. See table 3 below for the results of the ANOVA’s.
Table 3. Summary table of means and F-values for dependent variables Stimuli Man + Masculine text (N=59) Man + Feminine text (N=58) Woman + Feminine text (N=62) Woman + Masculine text (N=57) F Evaluation 4.02 (.77) 4.17(.90) 4.44 (.97) 4.56 (.92) 4.48** Gender equality 6.23 (.82) 6.27 (.76) 6.04 (.84) 5.71 (1.17) 4.60** Gender equality government 3.80 (1.33) 3.75 (1.35) 3.75 (1.31) 3.75 (1.51) .034 Capability Men 4.06 (.79) 4.13 (.65) 3.91 (.72) 5.12 (.65) 1.10 Capability Women 4.13 (.79) 4.18 (.66) 4.04 (.70) 4.17 (.64) .50
Note. *P<0.05, **P<0.01. Higher scores indicate higher evaluations of the politician and more positive attitudes towards
gender equality and capability. Standard deviation found within parentheses.
Male politician +
Masculine message Feminine messageMale politician + Female politician +feminine message masculine messageFemale politician +
Man 6,2 6,13 5,92 5,57 Woman 6,25 6,38 6,17 5,82 Total 6,23 6,27 6,04 5,71 5,3 5,5 5,7 5,9 6,1 6,3 6,5 M ea ns
Evaluations
Apart from these main effects, the results are given for all the hypotheses with
evaluation as dependent variable. To start, H1a predicts that those who were exposed to the
stimuli with the masculine message will evaluate the politician higher than those who were exposed to the stimuli with the feminine message. Yet the results show that there is no significant difference in evaluation between the two message conditions (F(1)=.052, p= .82). The masculine message (M=4.28, SD=.89) received only slightly lower evaluations than the feminine message (M=4.31, SD=.95). Thus there is no support for H1a.
It could be however that the female politician is significantly better evaluated generally, as females are generally perceived as more trustworthy and honest (H1b). Yet on the other hand, male politicians are perceived as better leaders (H1c). The models indeed found a significant difference between the man and the woman regarding their perceived trustworthiness and honesty (F(1)=16.07), p<.01). The female politician (M=5.65, SD=1.16) scored higher on trustworthiness and honesty than the male politician (Mdifference= .60) hence H1b can be accepted. Regarding their leadership qualities however, there is no significant difference (F(1)=3.58) although the female politician scores slightly higher (M=4.23, SD=1.23) than the male politician (M=3.92, SD=1.23) thus there is no support for H1c. As already presented earlier, the overall evaluation is best for the female politician with the masculine message (H1d).
Political ideology and evaluation. Social political ideology should also influence how the politicians are evaluated. Therefore it was predicted that female politicians will find more support among liberals than conservatives (H2a). Because the ideology was measured on a scale, and the photo variable only contains 2 conditions (male or female politician), a linear regression analysis is performed. Again, the models controlled for demographic variables. Table 4 below shows the effect of the interaction with political ideology and photos on the
evaluation. The model shows that, except for the photos, the interaction with political ideology is not a predictor for the evaluation of the male or female politician, thus H2a was rejected.
Table 4. Linear regression analysis. A prediction model for the evaluation of the politician
Model 1 Model 2
Evaluation male/female politician B Std. Error Beta B Std. Error Beta
Constant 3.87
Photo man/woman .392* 1.12 .216 - .040 .094 - .103
Photo man/woman * Pol. Ideology - .063 -.047 - .086 -.062 .082 -.631 Photo man/woman * Pol. Trust - .070 - .047 - .095 - .068 .092 - .782
R2= .03
Note. *P< 0.05, **P<0.01.
Trust and evaluation. Another factor that should influence the evaluation is political trust. H2b states that the stimulus with the female photo will be better evaluated than the male photo by voters with low political trust, as female politicians are a desired change of the status quo. The results show however that there no significant difference for the interaction with political trust in predicting the evaluation of the politician. Thus H2b was rejected.
Gender and evaluation. Furthermore is was also predicted that women are actually more likely to be supportive of a female politician who stresses feminine issues than a male politician (H2c), thus gender is added as an additional factor. Results show that there is indeed a significant difference in how men and women evaluate the female politician with the feminine text differently (F(1)=7.23, p=<.01). Women evaluate the female politician with the feminine text (M=4.65, SD=.16) significantly higher than men do (M=4.24, SD=.16). Thus H2c can be accepted. See table 1 in the evaluations section.
Political participation
Previous studies have also shown, that women who are represented by a female politician, show higher political participation than when they are represented by a male
politician (H3). For this hypothesis the intended voting behavior is the dependent variable. Results of an ANOVA show however that there is no significant difference for intended voting behavior after exposure to the stimuli (F(3)= .185), p=.91). Also the interaction with gender shows no significant results (F(1)= 2.06, p= .107). Thus seeing a female politician has no effect on intended voting behavior for women, which means H3 was rejected.
Issue capability
The correlation table in appendix 3 shows that there is no significant direction in capability of male and female politicians for handling economy, defense, healthcare, education and equal opportunities for men and women.
Gender and issue capability. It is still interesting to see however whether the participants’ gender is of any influence towards the perceived capability, therefore this is tested with an ANOVA. Results show that there is a significant difference in how men and women perceive men’s capability on issues (F(1)=4.88, p<.05). It turns out that men rate the male politicians capability lower (M=3.95, SD=.70) than women (Mdifference= .20).
Furthermore, there is also a significant difference between men and women for the perceived capability of female politicians (F(1)=10.905), p<.01). Again, men rate the female politician lower (M=3.97, SD=.70) than women (Mdifference= .48). Hence H4c is accepted. See appendix 4 for graph.
Gender equality
Table 3 already showed that there was a significant main effect on gender equality after being exposed to the female politician with the masculine message. Yet there was no significant effect for the attitudes towards gender equality in government. However, it could be that political ideology can influence this. Therefore it was theorized that conservatives are less inclined to favor an equal division of men and women in government (Hogan, 2010). Being exposed to the female with the masculine message could however enhance the notion
of gender equality for conservatives (H5c). Results of the ANOVA show however that there is no significant increase in the support for gender equality in government for conservatives, after having seen the female politician with the masculine text (F(15)=1.57, p=.083). Thus there is no support for H5c. See table 4 below for means.
Table 4. Summary table of means for attitudes towards gender equality in government Stimuli Man + Masculine text (N=60) Man + Feminine text (N=58) Woman + Feminine text (N=63) Woman + Masculine text (N=56) Extremely Liberal 4.70 (1.19) 2.86 (1.36) 3.19 (.92) 5.88 (.88) Liberal 3.77 (1.21) 3.63 (1.45) 3.67 (1.40) 4.12 (1.58) Slightly Liberal 4.10 (1.21) 4.35 (1.48) 4.05 (.78) 3.46 (1.35) Moderate 2.37 (1.23) 3.75 (.35) 3.90 (1.69) 2.53 (1.01) Slightly Conservative 4.25 (2.05) 3.93 (.80) 4.75 (.81) 3.42 (.29) Conservative 2.75 (2.47) 3.25 (.75) 2.58 (1.46) 4.13 (1.94)
Note. *P<0.05, **P<0.01. Higher scores indicate more support for gender equality in government. Standard deviation
found within parentheses.
Conclusion and discussion
This was an experimental study, designed to explore how male and female politicians can best present themselves, in order to gain favorable media coverage. Moreover it has attempted to show, how stereotyped media coverage on politicians influences the notion on gender equality. The goal of this study was to advance research on media representation of men and women, in which there was an overlap between gender, politics and media. By doing so, several concepts were touched upon, such as political ideology and trust, which in turn would have an effect on the dependent variables of evaluation, issue capability and gender equality.
Overall, there were some interesting significant effects of the four stimuli that were shown to the participants. Firstly, the female politician portraying the masculine message was by far best evaluated among the stimuli. This is surprising, taking into account the traditional
notion of politics as a predominant masculine field, yet the combination of the female
politician with the masculine message might combine ‘best of both worlds’. This is supported by the other significant results which show that women score higher on honesty and
trustworthiness, whereas men score higher on leadership. Furthermore, there was a significant effect for gender and evaluation, as women evaluate the female politician with the feminine message much higher than men do. Therefore, it could be useful for female politicians to target female voters, by stressing feminine issues. Trust and political ideology however had no effect on the evaluation.
Moreover, there was a significant effect on gender equality. Although predicted differently, being exposed to the female politician with the masculine message actually created a significant decrease in attitudes towards gender equality. It means that people think men and women are actually less equal to be successful in politics. This is a remarkable outcome, as it would appear masculine females would actually blur the lines between the traditional stereotypes of men and women. A cautious explanation for this effect could be however that people are caught by surprise when seeing a very masculine female, and as a consequence are reminded of the conventional roles men and women supposedly have in society.
Furthermore, there was no difference found in the perceived issue capability of men and women. Hence this implies that people do not see any difference in the professional capability for male or female politicians to handle certain topics. This actually strengthens the notion that most of the factors that hinder women to climb the political ladder are external. Apart from the fact that many women have to give labor at a certain age, which could cause obstacles for a career into high office, there are many other factors which are to be
An extensive analysis has shown that both men and women are often stereotyped in the media, which causes for unwanted media representations.
However, as both journalists and politicians are responsible for the media coverage, politicians should be aware of the effects their media presence has on the voters. This study can thus provide guidance for (fe)male politicians in how to best present themselves in order to gain advantageous media coverage. For example, as it has turned out, female voters are especially fond of female politicians stressing feminine issues, thus this is useful knowledge to bear in mind during a campaign. Also strength is very positively evaluated, and thus could be a trait to emphasize. It then depends on what goal the politician has in mind, and what kind of coverage serves that goal.
Due to the academic standards of this research, it was important that the stimulus material was very similar in design and content. Also the stereotypes were very strong, and would actually require much more nuance in an actual campaign. These strong messages caused for valid results in this study, yet it should be noted that in the ‘real’ political arena there are many more factors that can influence the evaluation of a politician; such as party identification or political issues that are important at that time.
To conclude, the advice to ‘act like a lady and think like a man’, could indeed be very useful for struggling female politicians. Yet as always, this is heavily dependent on context, and thus the results should be used as a tool rather than an objective, in order to carefully lay out a suitable media strategy.
Reference List
Bolzendahl, C. I., & Myers, D. J. (2004). Feminist attitudes and support for gender equality: Opinion change in women and men, 1974–1998. Social Forces, 83(2), 759-789.
Braden, M. (1996). Women politicians and the media. University Press of Kentucky. Cantrell, T. H., & Bachmann, I. (2008). Who is the lady in the window? A comparison of
international and national press coverage of first female government heads. Journalism
Studies, 9(3), 429-446.
Connell, R. W., & Connell, R. (2005). Masculinities. Univeristy of California Press.
Connell, R. W. (2005). Change among the gatekeepers: Men, masculinities, and gender equality in the global arena. Signs, 30(3), 1801-1825.
Devere, H., & Davies, S. G. (2006). The Don and Helen New Zealand Election 2005: A media a-gender?. Pacific journalism review, 12(1).
Devitt, J. (2002). Framing gender on the campaign trail: Female gubernatorial candidates and the press. Journalism & Mass Communication Quarterly, 79(2), 445-463.
Eagly, A. H., & Mladinic, A. (1994). Are people prejudiced against women? Some answers from research on attitudes, gender stereotypes, and judgments of competence. European review of
social psychology, 5(1), 1-35.
Fountaine, S., & McGregor, J. (2002). Reconstructing gender for the 21st century: News media framing of political women in New Zealand. Australian New Zealand Communication
Association Annual Conference, 10-12.
Funk, E. (2013). Ms. Representation: Gender Bias and its Effect on Female Political Ambition.
The Elizabeth Cady Stanton Student Research Conference Proceedings.
Gamson, W. A. (1992). Talking politics. New York: Cambridge University Press.
Gallagher, M. (2001). The push and pull of action and research in feminist media studies.
Feminist media studies, 1(1), 11-15.
Harvey, Steve. Act like a lady, think like a man. Amistad Press, 2011.
Hayes, D. (2011). When gender and party collide: Stereotyping in candidate trait attribution. Politics & Gender, 7(2), 133–165.
Heldman, C., Carroll, S., and Olson, S. (2005). “She brought only a skirt”: Print media coverage of Elizabeth Dole’s bid for the Republican presidential nomination. Political Communication, 22(3), 315–335.
Herrnson, P. S., Lay, J. C., & Stokes, A. K. (2003). Women running “as women”: candidate gender, campaign issues, and voter‐targeting strategies. Journal of Politics, 65(1), 244-255.
High-Pippert, A., & Comer, J. (1998). Female empowerment: The influence of women representing women. Women & Politics, 19(4), 53-66.
Hitchon, J. C., Chang, C., & Harris, R. (1997). Should women emote? Perceptual bias and opinion change in response to political ads for candidates of different genders. Political
Communication, 14(1), 49-69.
Hogan, R. E. (2010). Candidate Gender and Voter Support in State Legislative Elections.
Journal of Women, Politics & Policy, 31(1), 44-66.
Huddy, L. and Terkildsen, N. (1993a). Gender stereotypes and the perception of male and female candidates. American Journal of Political Science, 37(1), 119–147.
Huddy, L. and Terkildsen, N. (1993b). The consequences of gender stereotypes for women candidates at different levels and types of office. Political Research Quarterly, 46(3), 503-525.
Inglehart, R., & Norris, P. (2003). Rising tide: Gender equality and cultural change around the
world. Cambridge University Press.
Iyengar, S., Norpoth, H., & Hahn, K. S. (2004). Consumer demand for election news: The horserace sells. Journal of Politics, 66(1), 157-175.
Iyengar, S. (1994). Is anyone responsible? How television frames political issues. University of Chicago Press.
Jalalzai, F. (2006). Women Candidates and the Media: 1992‐2000 Elections. Politics & Policy,
34(3), 606-633.
Jost, J. T., Federico, C. M., & Napier, J. L. (2009). Political ideology: Its structure, functions, and elective affinities. Annual review of psychology, 60, 307-337.
Kabeer, N. (2005). Gender equality and women's empowerment: A critical analysis of the third millennium development goal 1. Gender & Development, 13(1), 13-24.
Kahn, K. F. (1992). Does being male help? An investigation of the effects of candidate gender and campaign coverage on evaluations of US Senate candidates. The Journal of Politics,
54(2), 497-517.
Kittilson, M. C., & Fridkin, K. (2008). Gender, candidate portrayals and election campaigns: A comparative perspective. Politics & Gender, 4(03), 371-392.
Kleinberg, S. J. (1988). Retrieving women's history: changing perceptions of the role of women
in politics and society. Berg Publishers.
Louden, A. and McCauliff, K. (2004). The “authentic candidate”: Extending candidate image assessment. Presidential candidate images, 85–103.
Major, L. H., & Coleman, R. (2008). The intersection of race and gender in election coverage: What happens when the candidates don't fit the stereotypes?. The Howard Journal of
Communications, 19(4), 315-333.
Mazzoleni, G., & Schulz, W. (1999). " Mediatization" of politics: A challenge for democracy?.
Political Communication, 16(3), 247-261.
McGinley, A. (2009). Hillary Clinton, Sarah Palin, and Michelle Obama: Performing gender, race, and class on the campaign trail. Denver University Law Review, 86, 709–725.
Niven, D., & Zilber, J. (2001). Do women and men in congress cultivate different images? Evidence from congressional web sites. Political Communication, 18(4), 395-405.
Paolino, P. (1995). Group-salient issues and group representation: Support for women candidates in the 1992 senate elections. American Journal of Political Science, 294–313.
Rosenwasser, S. and Seale, J. (1988). Attitudes toward a hypothetical male or female presidential candidate: A research note. Political Psychology, 9(4):591–598.
Sanbonmatsu, K. (2004). Democrats, republicans, and the politics of women's place. Scholastic. Scharrer, E. L. (2013). 15 Representations of Gender in the Media. The Oxford Handbook of
Media Psychology, 267.
Strömbäck, J., & Nord, L. W. (2006). Do politicians lead the tango? A study of the relationship between Swedish journalists and their political sources in the context of election campaigns. European Journal of Communication, 21(2), 147-164.
Terkildsen, N., & Schnell, F. (1997). How media frames move public opinion: An analysis of the women's movement. Political Research Quarterly, 50(4), 879-900.
Ulbig, S. G., & Funk, C. L. (1999). Conflict avoidance and political participation. Political
Appendix 1. Descriptives Sample and Variables Variable Cronbach's α M (%) SD Gender (Female)A 52.7% Age 39.3 15.9 Education (University)B 50.2% Occupation (Employed)C 47.7%
Political Ideology, economically: left vs rightD 4.18 1.35
Political Ideology, socially: liberal vs conservativeE 2.71 1.23
Political EngagementF
I am interested in Dutch politics
I am knowledgeable about Dutch politics
.86 4.67 1.42
Political TrustG
I have trust in Dutch politicians
I have trust in Dutch political institutions
.89 4.10 1.32
Importance “male” topicsH
Economy, defense, job opportunities, foreign affairs .66 5.75 .94 Importance “female” topicsH
Healthcare, education, gender equality .66 5.90 .99
Evaluation of the politicianI
Authenticity, trustworthiness, competence, leadership, honesty, confidence, integrity
.85 4.29 .92
Gender equality governmentJ
With more women in Parliament, problem solving is more cooperative and based on consensus.
The introduction of gender quotas is important for the equal representation of women and men in Parliament.
There should be an equal number of men and women on decision-making boards in government.
Having more women in Parliament increases the trust in politicians.
.82 3.76 1.37
Gender equalityJ
Men and women are equally qualified to make decisions concerning public affairs.
Men and women can represent the interests of female voters equally well. Men and women can serve the role of a leader equally well.
Men and women are both equally capable of being successful in politics.
.74 6.06 .93
Capability handling topics menK
Economy, defense, job opportunities, foreign affairs, healthcare, education, gender equality
.94 4.05 .71
Capability handling topics women K
Economy, defense, job opportunities, foreign affairs, healthcare, education, gender equality
.95 4.13 .69
Intended voting behavior L 6.72 .89
Note (N=239). A: Male= 47.3%. B: HBO= 30.1%, VWO= 7.1%. C: Student/intern= 21.8%, Self-employed= 16.3%. D: Political Ideology: Economy left vs right scale: 1= Extreme left to 7= Extreme right. E: Political Ideology: Social liberal vs conservative: 0= Extremely liberal to 7= Extremely conservative. F: Political Engagement scale: 1= Strongly disagree to 7= Strongly agree. G: Trust in politics scale: 1= Strongly disagree to 7= Strongly agree. H: Importance scales: 1= Not at all important to 7= Extremely important. I: Evaluation scale: 1=not at all to 7=very much. J: Equality scale: 1= Strongly disagree to 7= Strongly agree. K: Capability scale: 1= Not at all capable to 5= Very capable. L: Voting scale: 1=very unlikely to 7= very likely.
Appendix 2. The stimuli.
From left to right: Man with masculine text, man with feminine text, woman with feminine text, woman with masculine text.
Appendix 3. Correlation coefficients capability of handling issues Mean SD 1 2 3 4 5 6 7 8 9 10 11 12 13 1 Man – Healthcare 3.95 .86 2. Woman – Healthcare 4.25 .70 .66 3. Man – Education 4.09 .77 .84 .77 4. Woman – Education 4.19 .75 .66 .82 .77 5. Man - Economy 4.18 .77 .70 .79 .83 .79 6. Woman - Economy 4.05 .83 .72 .68 .81 .74 .79 7. Man – Defense 4.16 .77 .63 .71 .70 .68 .79 .59 8. Woman – Defense 3.88 .93 .68 .71 .68 .69 .62 .72 .61 9. Man – Employment 4.04 .84 .80 .71 .83 .69 .81 .76 .71 .65 10. Woman – Employment 4.15 .79 .68 .80 .79 .78 .80 .77 .73 .69 .84 11. Man – Gender issues 3.70 .99 .61 .56 .63 .61 .59 .65 .55 .59 .63 .58 12. Woman – Gender issues 4.23 .80 .54 .71 .65 .70 .69 .62 .66 .57 .65 .73 .51 13. Man – International relations 4.24 .73 .64 .80 .75 .80 .83 .68 .73 .62 .73 .71 .60 .69 14. Woman – International relations 4.14 .79 .66 .74 .74 .79 .75 .80 .64 .69 .74 .77 .62 .67 .76
Capability Men Capability Women Men 3,95 3,97 Women 4,15 4,27 3,7 3,8 3,9 4 4,1 4,2 4,3 M ea ns
