Communicating the F-word
The effects on equivalence and visual frames on attitudes towards and
participation in feminist movements
by Vonach, Theresa Student Number: 10239065
Master Thesis Political Communication
Amsterdam Graduate School of Communication
Supervisor: Wojcieszak, Magdalena January 29, 2015
Communicating the F-word: The effects on equivalence and visual frames on attitudes
towards and participation in feminist movements
Abstract
Drawing on message persuasion and social movement theory, this research examines to what degree framing can affect people’s attitudes towards the feminist movement as well as their willingness to participate in collective action. With a factorial experiment (n=249), this study investigates the effects of exposure to a photo of a feminist as either a private person or a public figure as well as the effects of a conflict-oriented vs. a peaceful message frame were tested. The findings of this thesis suggest that the messages feminist organizations or political actors who advocate gender equality disseminate, do affect people’s view on the organization and its cause. Yet, the same frame can affect attitudes and behavioral intentions differently as well as women and men react potentially different to the same frame.
Keywords:Feminism,Gender, Framing Effects, Equivalence Framing, Experimental Methods
Communicating the F-word
The effects on equivalence and visual frames on attitudes towards and participation
Feminism, defined as „the advocacy of women’s rights on the grounds of political, social, and economic equality to men“ (Feminism [Def.1,US English] Oxford Dictionaries), has achieved notable gains in the past decades. Despite its’ efforts to mobilize people globally for the advocation of gender equality, feminism is still a dirty word that few want to be associated with (e.g. Beck, 1998; Budgeon, 2001). Why this is the case has been and still is a troubling question to ponder.
The answer that feminism has lost its importance because gender equality has already been achieved (at least in western democracies), is hardly convincing, considering that across the EU one in three women has suffered from gender based violence, women still earn 16% less than then men per hour of work and only 11% of major political parties are led by a woman (European commission, 2013, 2014). Other explanations focus on discords within the movement, such as the struggle to unify its different viewpoints (e.g. Lugones & Spelman, 1983; Devereux & Devereux, 2005).
Yet, the strong argument has emerged that the lack of wider social recognition of feminism is also based on how the movement has been communicated. In particular, scholars have traced how feminism and its actors have been framed in the media (e.g. Beck, 1998, Bronstein, 2005; Dean, 2010; Lindt & Salo, 2002;) and, to a much lesser degree, how women’s organizations have approached their own communication efforts (Barnett, 2005). Despite this relatively large body of studies that have assessed reoccurring frames
frames affect people’s perception of women’s movement and, perhaps more importantly, their willingness to participate in it. Given that positive attitudes towards feminism are central for a wider societal acceptance of feminism and that feminist movements need public support when it comes to gathering donations or organizing successful demonstrations, for instance, insight into how communication can influence public opinion and spark participation can be
extremely valuable for feminist organizations and political actors who focus on gender equality.
Therefore, this thesis makes a first attempt to fill this research gap by examining how framing of the women’s movement and its’ actors affects attitudes towards feminism as well as intended participation in the movement, while considering whether the effects differ between men and women. More specifically, this thesis draws on two central frames, namely a private versus a public framing as well as peaceful vs conflict framing, identified as relevant in the literature on communicating feminism. This study systematically attends to both visual as well as textual frames, by drawing on a 2 (image, private vs. public portrayal) by 2 (text, peaceful vs. conflict-oriented) experimental design.
Theoretically, this thesis draws on social psychology theories to explain framing effects. It is especially referred to concepts of source similarity and self-referencing as well as collective identity, efficacy and injustice as the social psychological underpinnings of social movement participation (e.g. Van Stekelenburg & Klandermans, 2013; Van Zomeren, Postmes, & Spears, 2008). By doing so, we examine the research question, how private vs. public and conflict vs. peaceful framing can affect attitudes and one’s willingness to participate in the women’s movement and how these effects differ among female and male participants.
Theory Outline
Before reviewing the research that has been done on frames that are recurring in the communication about feminism and the women's movement, the theoretical underpinnings of this thesis have to be clarified. Thus, the main theories that this study draws on are briefly reviewed. First, the conceptual framing model that is applied is outlined. Second, the relevant elements of social psychological patterns in regard to message framing and social movement participation are discussed. Third, an overview is given regarding studies that have assessed how feminism has been framed and the two investigated frames of this study are discussed. By connecting specific findings about framing feminism with the general social-physiological underpinnings that are discussed, the guiding hypotheses are derived.
Theoretical Models of Framing Effects
Framing and its’ effects are one of the most extensively studied concepts in (political) communication (e.g. Chong & Druckman, 2007; Entman, 1993; Scheufele, 2000). Generally speaking, framing is based on the assumption that how an issue is characterized leads to different understandings among the audience; and its effects thus refer to behavior or
attitudinal outcomes that are due to differences in how an issue is presented (e.g.Scheufele & Tewskbury, 2007; Scheufele & Iyengar, 2012). Yet, framing can be defined variously and thus the conceptual definition that this thesis is drawing on has to be clarified.
Importantly, a distinction is made between emphasis and equivalence framing, the latter being the concept that is applied in this study. Emphasis framing focuses on the selection of certain issues and views and thus conveys differing perspectives on an issue (e.g.Entman, 1993), while equivalence framing refers to presenting information in logically
equal but differently phrased ways, e.g. „a 70% chance of losing“ versus „a 30% chance of winning“i(Kahneman, 2002; Scheufele & Iyengar, 2012).
While emphasis frames may produce spurious factors such as varying strength of arguments, equivalence framing allows to create directly comparable stimuli by altering the manipulations along specific dimensions, and thus directly links attitudinal or behavioral changes to the framing effect per se and not to some other content differences in the tested message (Scheufele & Iyengar, 2012).
Furthermore, equivalence framing is advantageous, as it is applicable to verbal messages as well as visual cues (Scheufele & Iyengar, 2012), both of which are used in this study. Just like text, visuals can be framing devices as they licable to verbal messages as well event graphically“ (Rodriguez & Dimitrova, 2011, p.51). The inclusion of visual material has been argued to be a valuable asset when studying framing effects (Coleman, 2010; Scheufele & Iyengar, 2012), as visuals potentially have a strong persuasive power and can thus add to the strength of the stimuli (e.g. Rodriguez & Dimitrova, 2011). In short, equivalence framing offers both methodological as well as operational advantages for experimental studies that are highly beneficial when focusing on how differently framed messages as well as pictures affect people’s attitudes and behavioral intentions.
The Social Psychology of Framing Effects
The choice to conceptualize frames as equivalence frames has further implications on how potential effects should be interpreted. Equivalence frames have been argued to cause schema-dependent effects, in a way that a message frame is more likely to be successful when it resonates to an individual’s mental schema that is activated when processing information (Scheufele & Iyengar, 2012). Social psychology as well as communication scholars, have
found that one of these schemas refers to the similarity between the source and the receiver, meaning that individuals are more likely to accept a certain message as convincing if they perceive the source of that message to be like themselves (e.g. Burnkrant & Unnava, 1995, Mackie & Worth, 1990; Rogers & Bhowmik, 1970; Wang,Walther, Pingree, & Hawkins, 2008). In other words, when a message is presented by a source similar to the receiver, the source is more likely to be perceived as an in-group member, as the receiver puts more weight on the source’s opinions, which in turn make the message more convincing and facilitates elaboration (e.g. Mackie & Worth, 1990; Rogers & Bhowmik, 1970; Wang,Walther, Pingree, & Hawkins, 2008). This is closely linked to self-referencing, which refers to the process that an individual is more likely to relate incoming information to self-structure or parts of one own’s memory the more linkage points he/she finds with the source, which in turn enhances elaboration and can facilitate attitudinal changes (Burnkrant & Unnava, 1995). This means that when a person finds linkage points with the source, e.g., similar age, he or she will elaborate the source’s message more deeply.
Yet, this thesis is not only concerned how frames can affect attitudes, but also how willingness to participate in a social movement, in particular the women’s movement, can be sparked. Social movement theory suggests that there are three main social psychological factors that are needed to be addressed in order to trigger collective action (Van Stekelenburg & Klandermans, 2013; Van Zomeren, Postmes and Spears, 2008).
First, frames should be more successful when collective identity is stressed. Collective identity refers to the feeling of belonging to a group that is defined by a socially relevant dimension (e.g. gender, social class, etc.), and it is strengthened when people are aware of the similarities they have with their group members (i.e. the protesters) (Van Stekelenburg &
Klandermans, 2013) and the organizers of the movement (i.e. leaders of a social movement organization) (e.g. Simon & Klandermans, 2001, Stürmer & Simon, 2004).
Second, when targeting groups that are not directly linked to the disadvantaged group (i.e. men in the women’s movement), mobilization through collective identity is challenging (Ellemers, 1993). Here, social movement scholars suggest an instrumental pathway, in a way that low-identifiers can become active participants in a movement if they perceive it as more efficacious (Simon, Loewy, Stürmer, Weber, Freytag, Habig, & Spahlinger, 1998;Van
Zomeren, Postmes and Spears, 2008). Here, efficacy does not refer to group efficacy, which means the sense that people feel themselves and one’s group members capable of
transforming the status quo and empowered to trigger social change (Drury & Reicher, 2005). Rather, it is referred to as political efficacy meaning that political actions are perceived to have an impact (Van Stekelenburg & Klandermans, 2013).
Third, the feeling of injustice is discussed in literature as one of the main components of social movement communication (e.g. Gamson, 1995, Gamson & Wolfsfeld,1993).
Injustice frames, which focus on conflict with political opponents, emphasize an unjust authority and are emotionally laden and anger–focused (Gamson, 1995). Social movements apply this strategy to make the movement’s issue seem more important and encourage participation in protest and rebellion as it „focusses on the kind of righteous anger that puts fire in the belly and iron in the soul“ (Gamson, 1995, p.91) – which arguably sparks
participation. Again, the feeling of injustice is not perceived equally among all groups, but the emotions that arise from the injustice frame are triggered especially, when it is one’s own group that is disadvantaged. When such group-based emotions arise, individuals’ willingness
to participate in collective action is enhanced (e.g. Frijda, 1986; Van Zomeren, Postmes, & Spears, 2008).
The Framing of Feminism
Communication scholars have extensively studied how the women’s movement and feminists have been framed in the media and have overall come to similar conclusions (e.g. Lind & Salo, 2002; Dean, 2010; Beck, 1998, Schreiber,2010; Mendes, 2011): Although feminism has gained more recognition, it remains a marginalized issue that is often
confronted with opposition (e.g. Beck,1998, Bronstein, 2005, Mendes, 2011) and feminists are often described as different from the average citizen and associated with reoccurring stereotypes (Lind & Salo, 2002; Mendes, 2011).
However, frames are not only created in news media, but also by political actors, who strategically form frames in order to influence public opinion (e.g. Chong & Druckman, 2007; Nelson, Oxley & Clawson, 1997). Yet, with regard to feminism, communication research has not closely attended to the feminist movement’s own attempts to strategically shape its messages. In a notable exception, Barnett (2005) analyzed a feminist organizations' press releases and discovered that the organization portrayed itself as a protector of women’s rights. In addition, the press releases suggested that the given organization was committed to uniting and working with other organizations to promote human rights. On the flip side, however, the organization was especially harsh on its opponents and their actions (Barnett, 2005).
This present study draws on both, studies about the framing of feminism in the media and by feminist organizations themselves. It closely attends to two dominant frames identified in extant literature, namely private vs. public and conflict vs. peaceful framing.
Private vs. Public Framing. The first frame pertains to whether a feminist is portrayed as a
personalized and connected to the private sphere, but are more likely to be portrayed as public figures who fulfill political obligations (Lind & Salo, Barnett, 2005). First, Lind and Salo (2002) found that in the news media, feminists are rarely associated with topics regular women are often connected to, including appearance, family and motherhood, personal qualities, leisure activities, or day-to-day work. To the contrary, feminists are almost exclusively connected to politics, government and the media (Lind & Salo, 2002). Second, similar tendencies were found in the communication strategy of women’s organizations. Barnett’s (2005) investigation of press releases of the feminist organization NOW suggests that the organization frames its feminine role as the „maternal protector“ in a sense that they stay watchful upon government institutions and will fight for equality for their daughters. Thus, the issue of motherhood was detached from privacy and was associated with actions within the public sphere, while personal issues and private sights remained unmentioned.
In this thesis the private versus public framing of feminists is operationalized as a visual cue – one picture shows a female feminist in a private setting and the other depicting her as a public figure. It is argued that attitudes towards the feminist movement and its actors as well as intended participation in the movement are affected by these differences in visual presentation, yet the frames are expected to affect men and women in different ways.
First, an overall difference between female and male participants is predicted. Former research has shown that women, on average, are more interested in ideas of feminism than men (e.g. Bolzendahl & Myers, 2004; Davis & Robison, 1991). Thus, women are more likely to be issue publics, that is a part of the general public who is more concerned about a specific topic and is therefore more attentive when exposed to information about issues related to that topic (e.g. Kim, 2009; Price & Zaller, 1993). In line with theories of self-referencing (e.g. Burnkrant & Unnava, 1995) research suggests that messages are more persuasive and exert
stronger effect when the source and the recipient are of the same sex (Bochner, 1994; Peck & Loken, 2004; White & Andsager, 1991). Since this study focuses on female (and not male) sources, female participants should be more likely to self-reference than men. This leads to the following expectation:
H1a: Women will have more positive attitudes towards the women’s movement, feminists and the feminist organization than men(regardless of the visual frame). Second, this difference between women and men is expected to interact with potential effects of private vs. public framing. Specifically, the private frame that presents a feminist more like an average women (rather than a public figure) should enhance self-referencing and enable effects of source similarity among women: Portraying feminists in a way that regular women are often portrayed and also emphasizing their ties to the private sphere (via such visual cues as a living room, a children’s drawing and casual, feminine clothing) may resonate with women’s own lives and thus makes feminists seem more similar to an average female and feminism seem more directly applicable to daily life (Lind & Salo, 2002; Roy, Weibust, & Miller, 2007). For these reasons, the private picture rather than the public picture should lead to a more positive attitude towards feminism and feminist actors among female participants.
In contrast, the public and governmental character of the source in the public condition is expected to contain processes of source-similarity, as the cues given in the public picture (e.g. government setting, official, „masculine“ clothing) are typically associated with feminists as being not „like me“, which was found to be a dominant negative stereotype people have about feminists (Twenge & Zucker, 1999, p.603). Also, portraying feminists as different to regular women (with the presumption that not many females hold public functions or are often involved in governmental affairs) may „reinforce the perception that feminism is
neither relevant nor particularly applicable to the bulk of daily life of the majority of citizens“ (Lind & Salo, 2002, p.224). This review leads to the following hypothesis:
H1b: The difference in attitude between men and women will increase in the private condition, as women will have more positive attitudes towards the women’s
movement, feminists and the feminist organization especially when exposed to a feminist in a private rather than public setting.
Looking beyond attitudinal outcomes, this thesis is also interested in behavioral intensions. In social movements, like feminism, it is not only important to obtain support for the movement’s views, but also to mobilize collective action, as an actively participating public is vital for any social movement’s survival (e.g. Klandermans,1984).
Again, it can be assumed that women show a higher willingness to participate in the women’s movement when compared to men, due to a higher likeliness of self-referencing and the women-focused topic. As discussed above, women are more likely to be issue publics (e.g. Kim, 2009; Price & Zaller, 1993), because feminism in general and the stimuli of this study in particular focus on the suppression of women (rather than men), which in turn should lead to a higher willingness to participate among women in general.
H 2a. Women will be more willing to participate in the women’s movement than men (regardless of the visual frame).
Moreover, women and men are expected to respond differently to the private and the public frame. First it is expected that women respond positively to the private picture, as the picture potentially stresses collective identity between the movement’s organizers and potential participants (e.g. Simon & Klandermans, 2001, Stürmer & Simon, 2004). Women are expected to identify more with the president of the feminist organization when she is depicted in a private setting, since it portrays those who actively contribute to the movement
as women „like me“, rather than abstract politicians (Lind & Salo, 2002). To the contrary, the public portrayal of a feminist may convey the message that feminists „are involved in types of things (…) that most of us will never become active in“ (Lind & Salo, 2002, p. 224) and underlines the discrepancy between female participants and feminist activists.
However, generally speaking, it is challenging to motivate men to participate in the women’s movement via collective identity, as men are not directly associated with the women’s right movement, do not constitute an issue public for this cause, and do not form part in the largely disadvantaged group that feminism represents. As a result, it may be harder for men to adopt collective identity (Ellemers, 1993) and – by choosing a female (not male) president of a feminist organization – this study does not counteract these challenges.
It is possible that men, as low identifiers and thus with a low collective identity, can be mobilized for feminism via political efficacy (Van Zomeren, Postmes, & Spears, 2008). In particular, by portraying a president of a feminist organization as a public figure who is connected to the political sphere, the picture may convey higher levels of the movement’s political success, as it depicts the feminist as someone who is more likely to be perceived as competent to successfully influence those in power. This review leads to the following expectation:
H 2b: The difference in participation between men and women will be canceled out in the public condition. Women will be especially likely to participate in the women’s movement after exposure to the private picture, while males will be more willing to participate after exposure to the public picture.
Conflict vs. Peaceful Frame. The second frame (conflict vs. peaceful) refers to the tone in
which opponents of feminism are addressed and is operationalized as a textual message. The strategy to harshly attack opponents, mostly labeled as injustice frame, has been found to be a
reoccurring theme in social movement framing (e.g. Benford & Snow, 2000; Gamson, 1982) and was also found to be present in communication strategies of feminist organizations (Barnett, 2005). Here, opponents, which can be anyone who contradicts the paradigms of gender equality, have been casted as „contemptuous, irrational, dishonest and
radical“ (Barnett, 2009, p.353).
However, it seems that this frame evokes a negative perception amongst journalists and arguably the public. Various scholars (e.g. Beck, 1998; Bradley, 2004; Bronstein, 2005; Lind & Salo, 2002) found that feminists themselves are often assigned with negative attributes, when being portrayed as radical, outraged man-haters in the media and aggressiveness and radicalism are the negative stereotypes associated with feminist
movements and its actors (Berryman-Fink & Verderber, 1985). Previous research furthermore suggests that a focus on conflict can have negative effects on people’s attitudes ( Arpan, Baker, Lee, Jung, Lorusso, & Smith, 2006) and thus ,potentially evokes negative stereotypes as it could be perceived as aggressive and stubborn (Roy, Weibust, & Miller, 2007), rather than protective, democratic and just (Barnett, 2009).
To the contrary, exposure to the peaceful frame, one that emphasizes collaboration and willingness to compromise, should challenge negative stereotypes, and thus have a positive effect on participants’ attitudes towards feminism and feminist actors. In fact it has been shown that exposure to positive stereotypes – rather than negative stereotypes that tap radicalism, obstinacy and conflict-orientation – leads to a higher self-identification as a feminist (Roy, Weibust & Miller, 2007). On the basis of this review, the following is predicted:
H3a: Exposure to the peaceful framing will result in more positive attitudes towards the women’s movement, feminist activists and feminist organizations, than exposure to conflict framing regardless of the gender of the recipient.
Exposure to the peaceful frame, that challenges feminist stereotypes, is expected to be especially effective among women. Studies that have investigated the effects of positive vs. negative stereotypes regarding feminists have shown that alongside a higher
self-identification with feminism, positive stereotypes also enhance lead to a higher perceived ability to meet individual challenges among women (Roy, Weibust & Miller, 2007). Being confronted with non-stereotypical information, as presented in the peaceful condition, could thus enhance positive self-referred feelings among women (more than among men), which in turn should strengthen the positive effects on attitudes towards feminism and feminist actors among women. Thus, this review leads to the following prediction:
H3b: The positive effect of the peaceful condition on attitudes towards the women’s movement and its actors will be more evident among female participants than among male participants.
As mentioned before, this thesis is also concerned about how different frames affect people’s willingness to participate in the movement. Despite its’ potential negative effect on attitudes towards feminism and its actors, the emotional character of the conflict frame may trigger what is needed to spark collective action (Gamson, 1982). As previously outlined, the focus on conflict with an unjust authority addresses emotions of anger and compassion which scholars have found to be essential to mobilize individuals to take part in collective actions (Gamson, 1982; Klandermans, 1984). Therefore, we predict the following:
H4a: Exposure to conflict framing will result in a higher willingness to participate in the women’s movement than peaceful framing for both men and women.
Furthermore, it is expected that the effect of conflict vs. peaceful framing interacts with gender, in a way that it is especially pronounced among women. As discussed, social movement theory suggests that the emotions that arise from the injustice frame are triggered especially, when it is one’s own group that is disadvantaged, and that when such group-based emotions arise, individuals are more likely to participate in collective actions (e.g. Frijda, 1986; Van Zomeren, Postmes, & Spears, 2008). Despite the fact that gender inequality also has negative consequences for males, feminism in general, and the stimulus material of this study in particular, focus inequality as primarily related to women, which is why it can be assumed that group-based anger is more likely to occur and spark collective action especially among female participants. On the basis of this review the following is expected:
H4b: The framing effect will interact with gender, in a way that the more positive effect of the conflict frame, i.e. intended participation, will be especially pronounced among female participants.
Interaction between the two frames. Lastly, it can also be expected that the two factors,
private vs. public and peaceful vs. conflict framing, should interact. Since the private picture as well as the peaceful text are expected to have a more positive influence on attitudes towards the women’s movement and its’ actors than the public picture and the peaceful text, respectively, an overall effect of all the four conditions can be expected. Furthermore, both effects were hypothesized to be emphasized among women, which leads to the following prediction.
H5: Participants in the private-peaceful condition will have a more positive attitude towards the women’s movement, the feminist organization and feminists than participants in the public-conflict condition and this effect will be especially pronounced among female participants.
As discussed earlier, the effects of conflict and peaceful framing may be distinct when it comes to mobilizing action. Yet, the private picture was hypothesized to have a positive effect on women’s wiliness to participate, while public framing should increase men’s readiness for collective action. Thus, the following is presumed:
H6: Women in the private-conflict condition will be more willing to participate in the women’s movement than women in the public-peaceful condition, while men will be more willing to participate when exposed to the public-conflict condition.
Methodology
Context
The study was carried out in Austria, which represents and interesting context for the
purpose of this study: Austrian women experience among the highest rates of inequality in the EU, which become especially evident in the professional field. The gender pay gap is the highest in the European Union and Austrian women earn 25.5% less than men, the workforce is highly segregated and Austrian women are extremely underrepresented in decision-making positions (European Commission, 2012). Yet, women are participating in Austrian labor force almost as much as men – 45% of the female population compared to 57% of the male
population are part of labour-force– (Statistik Austria, 2007). Careers and professionalism have become predominant in Austrian women’s gender-role definition, while the traditional family role of women is widely rejected (Pongrácz, 2006). Thus, women seem to strive for a role in society that is equal to men, and yet, are facing above-average levels of inequality in their day-to-day lives. One of the reasons for this could be that feminist organizations do not find enough support to actually impact governmental decisions and policies. Therefore,
findings of this study could be especially meaningful to Austrian feminist organizations as it could help to attract more public support.
Design and Procedure
To investigate how different textual and visual frames can affect people’s attitudes
towards the women’s movement and its actors as well as their willingness to engage in collective action, an online experiment with a 2 (private vs. public) by 2 (conflict vs. peaceful) between-subjects factorial design was conducted. There were four different experimental conditions, to which participants were randomly assigned.
Between November 25 and December 08, 2014, respondents were contacted via email or private messages on Facebook. Despite methods of convenience sampling were applied, we aimed at recruiting a diverse sample in regard to age and education by contacting different organizations from different occupational fields, new high school alumni, students,
pensioners, etc. The message contained the information that the online study was part of a Master Thesis project at University of Amsterdam and briefly described the procedure. Participants were asked to consider participating and were informed about the approximate duration of the study. The email contained a link to the online study. When clicking on the link, respondents were redirected to a page that contained information about the anonymity of the participants, the general procedure of the study as well as contact details for any inquiries or complaints. To continue with the study, respondents were asked to declare that they
voluntarily agreed to participate, are 18 years or older and to acknowledge to have read and understood the information stated beforehand.
Once respondents had agreed to participate, they were randomly assigned to four different experimental conditions. Due to unknown reasons, 28.4% (N=71) of the participants
who completed the survey were in the publice influence on attitudes towards the w in the pubic-conflict condition, 21.6% (N=54) were in the private-peaceful condition and 23.6% (N=59) saw the private picture and read the conflict frame. A comparison between these groups revealed no differences in age, F(3,243) = .14, p = .936, gender, 𝛘2 = 1.05, p = .790,
education, 𝛘2 = 3.25, p = .354. Therefore, randomization was successful despite slightly
unbalanced cell sizes.
The website that was shown in the stimuli contained a short mission statement and a picture of the female president of the organization. Text and picture varied depending on the condition, whereas the setup of the screenshot was the same in each condition. After exposure to the mock website, respondents were asked to complete a post-hoc test.
Respondents
Overall, 249 adults competed the study. The respondents did not receive any compensation. Participants were between 18 and 78 with an average age of M=30.59,
SD=12.40. Sixty percent of the participants were female and most participants had completed rather high education: Twenty-six percent had finished their Bachelor’s degree, 25.6% had completed a Master’s degree and 24% graduated from high school (AHS). When compared to the actual Austrian population it becomes apparent that the young, the well-educated and women are over represented in this sample (For a more detailed description of the participants please see APPENDIX A).
Stimulus Material
The stimulus was especially constructed for this experiment and designed in a way that it looked like a screenshot from a website of a mock women’s right organization
„Österreichische Organisation for Feminismus und Frauenrecht – ÖOF“ [Austrian
Organization of Feminism and Women’s Right – abr. ÖOF]. The screenshot was designed with the exact same layout in all conditions in order to guarantee stimulus equivalence and a logo, the same across conditions, was added to enhance credibility and ecological validity. Choosing this treatment over presenting information as e.g. a newspaper article, also added to the ecological validity of this study: Feminism is an issue that is highly marginalized in the Austrian mainstream news which makes it difficult for feminist organizations to reach their audiences via that channel (Geiger, 2002). To the contrary, online communication initiated by feminist organizations has been found to be a more successful way to reach a broader
audience to advocate gender equality (e.g. Drüeke & Winker, 2005), which makes exposure to such content in the real world more likely.
Private vs. public framing. The first factor was operationalized by applying image framing.
Depending on the condition a participant was assigned to a different picture of the same woman. The pictures, which were specifically taken for the purposes of this study, differ in terms of background and clothing, categories important in terms of framing feminists as either private or public figures (Lind and Salo, 2002). In the private condition, the president of the feminist organization is wearing causal clothes, sitting in a living room, and there is a framed children’s drawing hanging on the wall in the background. These attributes portray the woman in a personalized way, by emphasizing her connection to the private sphere.
In the public condition, the woman is standing in front of the Austrian parliament, wearing office clothes, and there is an Austrian flag in the background. These attributes point to public sphere activities, such as politics and government (Barnett, 2005; Lind & Salo,
2002). In both conditions, a label was added to the picture that said „Anna Schubert, president of ÖOF“ (See APPENDIX B for photos).
A pilot test was conducted to test whether the woman is perceived equally likable, pleasant, trustworthy and attractive in order to exclude different perceptions of the woman as a confounding factor. It was furthermore tested whether participants were able to identify the setting. Results showed that the two pictures are comparable in terms of the various
characteristics, and that a majority of respondents was able to identify the setting as private or public, respectively (See APPENDIX C for more detailed results)
Conflict vs. Peaceful Framing. The second factor was constructed on the basis of textual
equivalence framing: Two versions of brief mission statements were written with the exact same arguments and the same number of words (word count: 148). This statement was presented under the heading “About us. Mission and Goals“ and at the very end of the text an appeal was made that asked readers to participate in the women’s movement. The arguments which were presented in the text were all based on facts and the texts were written in the style of a mission statement similar to real women’s organizations. All sentences in the text varied in at least one expression and in sum 19 expressions were altered to construct the conflict versus soft condition.
The conflict frame picked up an aggressive tone towards the opponent and expressions were altered that described the opponents, their actions and the consequences of such actions, as well as strategies to counteract, e.g. To win this battle and to revolt against irrational decisions, which result in contemptuous actions (…)“. The peaceful frame used softer and more neutral expressions to do so, e.g. „To attain our goal and to be able to do something against unreasonable decisions, which result in unfair actions (…)“ (see APPENDIX B for
full text). An independent sample t-test found that participants who read the harsh text
perceived the text to be more aggressive, conflict-orientated and radical (M= 6.02, SD= 2.16) than those who read the soft text (M= 4.41, SD= 1.75), Mdifferernce = -1.61, CI 95% [-2.1,-1.11], t (246) = -6.43, p < .001. Thus, the manipulation was successful.
Dependent Measures 1
Attitudes towards Women’s Movements. Participants were asked to indicate their
agreement or disagreement with a number of statements from Fassinger’s (1997) attitudes toward feminism and women’s movement scale (FWM), reaching from 1(=strongly agree) to 7 (=strongly disagree). Six items were retrieved from this scale (e.g. „The women’s movement made important gains in equal rights for women“). After recoding the negatively phrased items, a principle component analysis showed that the six items form a single uni-dimensional scale (Eigenvalue = 3.22, 53.62% of the variance explained). All items correlated positively with the first component and the higher the score on the scale the more positive the attitude towards the women’s movement (M= 5.02, SD= 1.03). The reliability of the scale was good, Cronbach’s α= .82.
Attitudes towards the feminist organization. The same six items from the FWM scale were
adapted and connected to the specific organization that was presented in the stimulus. Three statements were added that comprised the organization’s trustworthiness, solution orientation and credibility. Again, negatively phrased items were recoded. The nine items form a single uni-dimensional scale ( Eigenvalue= 4.37, 48.56% variance explained). All items correlated positively with this component and a higher score on the scale reflects a more positive attitude
See APPENDIX D for question wording and M and SD for separate items
towards the feminist organization (M= 4.59, SD= .97). The reliability of the scale was found to be good, Cronbach’s α= .87. Additionally, participants indicated their feelings towards the organization on a feeling thermometer ranking from 0 (=very negative feelings) to 100 (=very positive feelings) (M= 57.78, SD= 23.57).
Evaluation of Feminists. Respondent were asked to evaluate „the average feminist“ using a
10-point bipolar scale. Six items tapped attractiveness (not concerned with appearance–very concerned with appearance), gender (masculine–feminine), political attitude (radical– moderate; conflict oriented–consensus oriented) and attitudes towards men (hate men–like men) (e.g. Reid& Purcell’s, 2004). The six items formed a single uni-dimensional scale (Eigenvalue= 3.36, 56.08% of the variance explained). All six items correlated positively with this component. The higher participants scored on the scale, the less stereotypical they
evaluated the average feminist, i.e. a higher score points to a description of average feminists as consensus oriented, moderate, feminine, etc. rather than conflict oriented, radical and masculine (M= 5.01, SD= 1.52). For this scale the reliability was good, Cronbach’s α= .84. Furthermore, participants were asked to describe their feelings about feminist activists, by indicating a number between 0 (=very negative feelings) and 100 (=very positive feelings), (M= 52.68, SD= 26.18).
Willingness to Participate in the Women’s Movement. Online as well as direct participation
was measured. Respondents were asked how likely they were to participate in each of the nine activities related to the women’s movement within six month on a seven-point scale from very unlikely (=1) to very likely (=7). The items that concerned online participation were adapted from ANES and the Youth & Participatory Politics Survey Project (Cohen & Kahne, 2012), e.g. „Use the information on ÖOF’s website to convince other people of your standpoint in a
chat or a conversation“. Furthermore direct participatory actions such as „attending events organized by a women’s center“, or „sign an email or paper petition for a women’s rights issue“ were included (Liss, Crawford, & Popp, 2004). The principle component analysis confirmed that the nine items formed two uni-dimensional scales: There were two
components with an eigenvalue above 1. The first component had an eigenvalue of 4.98 and explained 39% of the variance, the second with an eigenvalue of 1.10 explained 12.24% of the variance. The six items that dealt with direct participation in the women’s movement correlated positively with the first component (M=3.41, SD=1.39). This scale’s reliability was good, Cronbach’s α= .88. The three items which referred to online participation correlated positively with the second factor. The reliability of the scale that measured online
participation (M=2.52, SD=1.41) was also good, Cronbach’s α= .83. For both scales a higher score means a higher willingness to participate.
Gender. We asked participants to indicate whether they were male or female.
Control Measures. Participants also indicated to what degree they have experienced
discrimination based on their gender (Liss, Crawford, & Popp, 2004) on a seven-point frequency scale, reaching from 1(=never) to 7(=all the time), (M=2.70, SD=1.29).
Additionally, participants’ feminist network was assessed by asking „how many of the people you are in regular contact with would consider themselves as feminist?“, reaching from 1(=none) to 6 (=all) (M=3.15, SD=1.13). In addition, political knowledge was measured with three true or false questions about Austrian politics (M=2.06, SD=1.0) and political ideology was measured on a 0 (=left) to 10 (=right) scale (M=4.49, SD= 1.87)(Van Egmond, Sapir, van der Brug, Hobolt, & Franklin, 2010).
Results
Analytical Procedure
To test the proposed hypotheses, multivariate analysis of covariance was used. The analyses were spilt in three main parts. To test the particular effects of the manipulated texts and the manipulated pictures, respectively, two dichotomous variables were computed: One variable which differentiated between the two visual frames (public vs. private) and one which differentiated between the two textual frames (conflict vs. peaceful).
First, a MANCOVA tested the effects of the visual framing condition as well as its interaction effect with gender on the dependent variables. Second, a multivariate analysis of covariance was conducted to test the individual effects of peaceful versus conflict framing , as well as its interaction with gender. Finally, the overall effects of two the conditions were tested to assess potential interaction between the visual and the textual frames, also using a MANCOVA. Whenever an interaction effect was found to be significant, a follow-up post-hoc test was conducted that used a computed interaction variable (e.g. gender*picture condition).
Feminist network and experienced discrimination were included as covariates in all models (main and follow-up), since these variables have been found to be strong predictors of feminist attitudes and participation in the women’s movement (Liss, Crawford, & Popp, 2004). As framing effects of political messages have been found to be influenced by individual differences such as political sophistication and the firmness of one’s political orientation (e.g. Zaller, 1996; de Vreese, 2009) political ideology and political knowledge were also included as covariates. Spurious effects that are due to differences in education and age are not expected since it was shown that these variables were equally distributed between the experimental conditions. Since all reported effects are based on models which used the
described covariates, the reported means and standard errors are estimated marginals that were calculated while holding the included covariates constant. 2
Hypotheses testing
Visual Frames (private vs. public). The first hypothesis (H1a) predicted that women will
have more positive attitudes towards the women’s movement, feminists and the feminist organization (regardless of the visual frame). Table 1 shows that women had, indeed, a more positive attitude towards the organization, reported more positive feelings towards the feminist organization and evaluated feminists as less stereotypical than men did. However, female and male participants did not differ in regard to their attitude towards the women’s movement and their feelings towards feminist activists. Therefore, we can only partially accept H1a.
To test H1B, an interaction effect between gender and private vs. public picture condition was tested. This effect was not significant for attitudes towards the organization, F (1,229)= .39, p= .535, feelings towards the organization, F (1,229)= .52, p= .471, attitudes
Table 1
Summary Table of Estimated Means (Standard Errors) and F-values for attitudes as a function of gender differences
Variable Women (n=142) Men (n=88) F (1,229)
Attitude towards the organization 4.76 (.08) 4.42 (.10) 6.65**
Feeling towards the organization 60.97 (1.93) 53.02 (2.5) 5.80**
Attitude towards the women’s movement 5.12 (.08) 4.94(.10) 1.73
Evaluation of Feminists 5.22 (.12) 4.75 (.15) 5.21**
Feelings towards Feminist activists 54.0 (2.01) 50.44 (2.60) 1.08
*p< .1, ** p <.05,***, p<.01,**** p<.001
Note: Higher scores indicate more positive attitudes, feelings and a less stereotypical evaluation
Covariates appearing in the models are evaluated at the following values: Experienced Discrimination = 2,72, Political Knowledge = 2,07, Political ideology = 4,48, Feminist Network = 3,17.
towards the women’s movement F (1,229)= .07, p= .788, or feelings towards feminist activists, F (1,229)= .13, p= .718. Neither did the interaction between gender and picture condition significantly affect how participants evaluated feminists, F (1,229)= 2.18, p= .141. H1b cannot be supported as exposure to the private vs. the public frame did not interact with gender and thus the difference between men and women did not increase in the private condition.
In H2a it was predicted that women will be more willing to participate in the women’s movement than men (regardless of the condition). Indeed, women reported higher willingness to participate directly than men (direct participation: Mfemales=3.70, SE=.10 Mmales=3.01, SE=. 13), F(1,233)=16.96,Mdifferernce = .69, CI 95% [.36,1.02], p <.001; online participation
(Mfemales=2.75, SE=.11, Mmales=2.21, SE=.14), F(1,233)=7.91,Mdifferernce = .54, CI 95% [. 16,.91], p=005) and H2a is thus supported.
H2b predicted an interaction effect between gender and visual frame was predicted, in a way that women’s willingness to participate will be enhanced after exposure to the private picture and men’s willingness to participate will increase after exposure to the public portrayal of a feminist. In line with the hypothesis, the interaction effect between gender and visual frame on online participation ,F(1,233) = 3.35, p =.069, as well as direct participation, F(1,233) = 3.11, p =.079, was marginally significant. This interaction effect was plotted for the ease of interpretation.
In line with what was predicted, Figure 1 shows that women were more willing to participate online after exposure to the private condition, while for men, the opposite was true. This follow-up analysis showed that males in the private condition (M=2.02, SE=.19) were less willing to participate online than women in the same condition (M=2.88, SE=.17)
and this difference was significant ,, Mdifferernce = -.86, CI 95% [-1.58,-.13], p = .011. However, the difference between women (M=2.62, SE=.21) and men (M=2.40, SE=.19) were cancelled out after exposure to the public picture, Mdifferernce = .22, CI 95% [-.43,.87], p = 1.
!
!
!
A similar pattern was found for direct participation, where the follow-up ANCOVA yielded that after exposure to the private picture, women were more willing to participate directly (M=3.80, SE=.15) than men (M=2.84, SE=.18), Mdifferernce = .96, CI 95% [.33,1.60], p < .001, while there was no significant difference between women (M=3.60, SE=.13) and men (M=3.18, SE=.17) in the public condition, Mdifferernce = .42, CI 95% [-.15,.99], p =.314.
Therefore, consistent with H2b, the gender difference was cancelled out in the public condition as men showed a higher willingness to participate after exposure to the public
Figure 1
Estimated Marginal means for the interaction effect between gender and private vs. public picture on online participation.
picture and women were more willing to participate after exposure to the private picture.
Textual Frames (Conflict vs. Peaceful). In the third hypothesis it was predicted that peaceful
framing will result in more positive attitudes towards the women’s movement, feminist activists and feminist organizations, than exposure to conflict framing.
This hypothesis was partly supported. Table 2 shows that conflict vs. peaceful framing had a direct effect on attitudes and feeling towards the feminist organization as well as feelings towards feminist activists, but did not affect participants’ attitudes towards the women’s movement or their evaluation of feminists. In line with what was predicted exposure to the peaceful frame resulted in more positive feelings and a more positive attitude towards the feminist organization. To the contrary, participants reported more positive feelings toward feminist activists when reading the conflict focused message.
H3b predicted an interaction between gender and textual framing in a way that the effects should be more evident among female participants. However, the interaction between gender and conflict vs. peaceful framing did not reach significance in regard to attitudes towards the organization, F(1,233)=.06, p= .881, feelings towards the organization, F(1,233)=.11, p=.735, attitudes towards the women’s movement, F(1,233)=1.51, p=.220,
Table 2
Summary Table of Estimated Means and F-values for attitudes as a function of conflict vs. peaceful manipulation
Variable Conflict (n=118) Peaceful (n=112) F (1,222)
Attitude towards the organization 4.37 (.08) 4.80 (.08) 13.30****
Feeling towards the organization 60.03 (2.05) 53.80 (2.15) 5.80*
Attitude towards the women’s
movement 5.12 (.08) 4.94(.10) 1.73
Evaluation of Feminists 5.22 (.12) 4.75 (.15) 4.41
Feelings towards Feminist activists 55.26 (2.23) 49.48 (2.14) 3.49*
*p< .1, ** p <.05,***, p<.01,**** p<.001
evaluation of feminists, F(1,233)=.82, p=.365, or feelings towards feminist activists, F(1,233)=.76, p=.385.Thus, H3b is rejected.
H4a predicted that exposure to conflict framing will result in a higher willingness to participate directly and online when compared to peaceful framing, and H4b predicted that the framing effect will interact with gender, in a way that the more positive effect of the conflict frame is more evident among female participants. Participants who read the conflict frame did report a higher willingness to participate directly (M=3.45, SE=.11) than those who were exposed to the peaceful condition (M=3.26, SE=.11), yet this difference was not significant, F(1,233) = 1.57, p =.211. Neither did the effect of conflict vs. peaceful framing on online participation, reach significance, F(1,233) = 1.03, p =.312 and the predicted interaction effect of gender and textual frame was not found to have a significant effect on online, F(1,233) = . 460, p =.498, or direct, F(1,233) = .003, p =.960, participation. Therefore, Hypotheses 4a and 4b are rejected.
Overall Effects. Finally, it was tested whether the two experimental factors interacted. The
analysis showed that the four conditions significantly affected people’s attitudes towards the organization, F(3,218) = 4.44, p =.005. In line with what was predicted, participants who were exposed to the private picture and the peaceful text had the most positive attitude
towards the feminist organization (M= 4.81, SE=.13), while participants who were exposed to the public picture and the conflict-oriented text scored lowest (M= 4.31, SE=.12). The
difference between those two conditions was significant, Mdifferernce = .50, CI 95% [.04,.96], p = .028. Additionally, participants in the public-peaceful condition had a more positive attitude towards the feminist organization (M= 4.78, SE=.11) than participants who saw the same picture but read the conflict text (M= 4.31, SE=.12) and this difference was also significant,
Mdifferernce = .47, CI 95% [.04,.90], p= .024. Yet, all other pairwise comparisons were insignificant.
Participants’ feelings towards the organization, F(3,218) = 1.56 p =.200, their evaluation of average feminists, F(3,218) = .25, p =.865, their feelings towards feminists, F(3,218) = 1.01, p =.387, and their attitudes towards the women's movement, F(3,218) = .07, p =.977, were not significantly affected by the four conditions. Neither was there a significant interaction effect between gender and the conditions on attitudes towards the feminist
organization, F(3,218) = .07, p =.974, feelings towards the organization, F(3,218) = .53 p =. 660, evaluation of average feminists, F(3,218) = .94, p =.421, feelings towards feminist activists, F(3,218) = .34, p =.793, and their attitudes towards the women's movement,
F(3,218) = .53, p =.661. Since the conditions only affected participants’ attitudes towards the organization, but no other attitude measures were affected and there was no interaction between condition and gender, we can only participially accept H5.
Additionally, H6 was tested that predicted that women in the private-conflict condition will be more willing to participate in the women’s movement than women in the peaceful condition, while men will be more willing to participate when exposed to the public-conflict condition. Yet, we did not find exposure to the different conditions to have a
significant effect on willingness to participate online, F(3,218) = .97, p =.410, or directly, F(3,218) = 1.55, p =.203, nor did the interaction between gender and condition affect the willingness for online, F(3,218) = 1.36, p =.256, or direct participation, F(3,218) = 1.77, p =. 153. Thus, Hypothesis 6 has to be rejected.
Discussion
Advocating gender equality is a topic that has been around for almost a century and has recently become revitalized in its’ relevancy (e.g. Redfern & Aune, 2013) Yet, in order for the movement to gain wider social recognition and public support, it is important for feminist and women’s right organizations to know how certain ways of communication affect how people perceive the movement and how it affects their willingness to take action in it. Although scholarship on which frames have been circulating in the mass media, and to a much lesser degree in the movement’s own communication efforts, is growing, little is known about the actual effects of these frames. Additionally, recent efforts to specifically target men and include them in the feminist movement, (e.g. HeforShe Campiagn, 2014) indicate that men and women might respond differently to certain messages about feminism – another issue that is drastically understudied.
This thesis aimed to make theoretical and methodological contributions. Theoretically, it combined different approaches rooting in social psychology to explain framing effects. In particular, theories of source similarity and self-referencing, as suggested by message persuasion scholars, served to explain variances in attitudes as well as gender differences. Further, this paper built on the psychological elements of collective identity, efficacy and injustice as a useful framework to investigate what effects frames may have on intended participation.
Importantly, this study could demonstrate that it matters how feminist organizations communicate. While the general attitude towards the women’s movement was not as much affected, people’s attitudes and their feelings towards the feminist organization were found to be more positive after reading a peaceful, rather than a conflict-oriented text, especially when this text was presented together with a private portrayal of a feminist. To the contrary, a
conflict-oriented message together with a feminist portrayed as a public person had
counterproductive effects on how the organization was viewed. In other words, participants perceived the feminist organization that was presented in the treatment as most trustworthy, important and capable of actively securing and advancing equal rights when exposed to a peaceful message and a picture of a feminist in a private setting. Interestingly and
contradictory to what we have predicted and to what has been found in previous research (e.g. Arpan et al., 2006), the conflict oriented frame did result in more positive attitudes towards feminist activists.
Although we were not able to show that participation was positively affected by a conflict-oriented rather than a peaceful message, results did point in this direction, which underlines the importance of differentiating between attitudinal and behavioral outcomes when investigating the effects of conflict or injustice frames.
Furthermore, our findings evince that the same frame can have opposite effects on different target groups. In regard to social movement theory, this finding provides evidence to the proposition that those prone to adopt collective identity may be persuaded to participate in collective action in different ways than those who are more distant to the collectiveness of the disadvantaged group (e.g. Van Zomeren, Postmes, &, Spears; Veenstra & Haslam, 2000). Specifically, we found that women and men responded differently to a private vs. a public portrayal of a feminist. Despite very small differences, results suggest that women were more likely to participate in collective actions such as demonstrations, petitions or joining a
women’s organization, when exposed to a feminist that was portrayed in a private setting which potentially evoked the feeling that oneself is capable of changing the current status and thus appears to be more empowering for women. On the other hand, men seem to be more likely to be engaged through instrumental factors, such as how efficacious and thus successful
the organized movement appears to be in the political sphere. This, as well as the finding that women were overall more positive towards the movement and more willing to participate, suggests that in order to target men and thus make gender „everybody’s issue“, different communication approaches may be needed. Future research could examine to what degree the inclusion of male activists in campaigns or a clearer focus on how gender inequality affects men (Cornwall, 1997), and motivates them to join the movement.
Methodologically, this study also has some central advantages. To the author’s knowledge, it is the first experimental study that tested the effects of frames applied in
feminism. By using a factorial design, this study was able to isolate the casual impact of both, textual and visual frames as well the interaction between the two. In addition, by relying on equivalence rather than emphasis framing internal validity is further strengthened as the measured effects can be linked directly to the manipulations (Scheufele & Iyengar, 2012). Given that this experiment assigned participants randomly to the conditions, the potential for spurious associations and reverse causation are also redressed, which furthermore adds to the strength of the causal inferences that were made. Furthermore, the stimulus was created as a screenshot of website that used elements from real feminist organizations in order to make the treatment as realistic as possible.
Yet, the findings of this study should be interpreted in regard to some central limitations. First, problems of generalizability of the findings have to be addressed. While generalizability is not a strong suit of experimental research in general, the results of this study drew on a convenience sample in which highly educated, young people and women were overrepresented, which further decreases the generalizability. Another weak point that is connected to the sample may be the number of participants. With between 54 and 71
participants per condition, statistical significance is hard to reach when investigating rather small, communication effects. Thus, chances of making type two errors increase.
Furthermore, attitudes and willingness to participate were measured shortly after the exposure to the mission statement of the women’s organization. Therefore, we cannot draw any conclusions of whether these effects last long enough to actually change attitudes or spark participation. For further evidence future research could test whether the effects that were found are expanded or constrained when measuring long-term effects. Despite our efforts to create realistic stimuli that are more likely to be encountered in a real context, it is unlikely that a majority of the participants would pay close attentions to such contents in their day-to-day lives, which reduces the ecological validity of this experiment.
Conclusion
Although communication and gender studies scholars have investigated the frames circulating about feminism and have discussed their potential influence on public opinion (e.g. Barnett, 2005; Beck, 1998, Bronstein, 2005), our study was the first to systematically trace these effects and shows that communication can have an effect on how individuals perceive feminist organizations and their causes. Practically, the findings of this study imply that feminist organizations may have to craft different messages to enhance reputation or find sympathizers than when trying to mobilize people for protest or collective action. Further, the results suggest that whether a campaign is aimed at women, men or both might be an
References
Arpan, L. M., Baker, K., Lee, Y., Jung, T., Lorusso, L., & Smith, J. (2006). News coverage of social protests and the effects of photographs and prior attitudes. Mass
Communication & Society, 9(1), 1-20.
Barnett, B. (2005). Feminists shaping news: A framing analysis of news releases from the National Organization for Women. Journal of Public Relations Research, 17(4), 341-362.
Beck, D. B. (1998). The" F" word: How the media frame feminism. NWSA Journal, 10(1) 139-153.
Benford, R. D., & Snow, D. A. (2000). Framing processes and social movements: An overview and assessment. Annual review of sociology,26, 611-639.
Berryman-Fink, C., & Verderber, K. S. (1985). Attributions of the term “feminist”: A factor analytic development of a measuring instrument. Psychology ofWomen Quarterly, 9, 51–64.
Bochner, S. (1994). The effectiveness of same-sex versus opposite-sex role models in advertisements to reduce alcohol consumption in teenagers. Addictive behaviors, 19(1), 69-82.
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. Bradley, P. (2004). Mass media and the shaping of American feminism, 1963-1975 Univ.
Press of Mississippi.
Bronstein, C. (2005). Representing the third wave: Mainstream print media framing of a new feminist movement. Journalism & Mass Communication Quarterly, 82(4),783-803.
Budgeon, S. (2001). Emergent Feminist (?) Identities Young Women and the Practice of Micropolitics. European Journal of Women's Studies, 8(1), 7-28.
Burnkrant, R. E., & Unnava, H. R. (1995). Effects of self-referencing on persuasion. Journal of Consumer Research, 22(1),17-26.
Chong, D., & Druckman, J. N. (2007). Framing theory. Annual Review of Political Science, 10, 103-126.
Cohen, C. J., & Kahne, J. (2012). Participatory politics: New media and youth political action. YPP Research Network.
Coleman, R. (2010). Framing the pictures in our heads: Exploring the framing and agenda-setting effects of visual images. In D’Angelo & Kyupers (Eds.),Doing news framing analysis: Empirical and theoretical perspectives . New York: Routledge, 233-261. Cornwall, A. (1997). Men, masculinity and 'gender in development'. Gender & Development,
5(2), 8-13.
Davis, N. J., & Robinson, R. V. (1991). Men's and women's consciousness of gender inequality: Austria, West Germany, Great Britain, and the United States. American Sociological Review, 56(1),72-84.
Devereux, C. M., & Devereux, J. (2005). Feminism: What Are We Supposed to Do Now?. ESC: English Studies in Canada, 1(2), 9-11.
Dean, J. (2010). Feminism in the papers: Contested feminisms in the British Quality Press. Feminist Media Studies, 10(4), 391-407.
Drury, J., & Reicher, S. D. (2005). Explaining enduring empowerment: A comparative study of collective action and psychological outcomes. European Journal of Social
Drüeke, R., & Winker, G. (2005): Neue Öffentlichkeiten durch frauenpolitische Internet- Auftritte in Schachtner, C. & Winker, G. (Ed.), Virtuelle Räume – neue Öffentlichkeiten. Frauennetze im Internet. Frankfurt/Main, Germany: Campus Verlag, p.31-49.
Ellemers, N. (1993). The influence of socio-structural variables on identity management strategies. European review of social psychology, 4(1), 27-57.
Entman, R. M. (1993). Framing: Toward clarification of a fractured paradigm. Journal of Communication, 43(4), 51–58.
European commission (2012): The current situation of gender equality in Austria – Country Profile. Retrieved from: http://ec.europa.eu/justice/gender-equality/
filesepo_campaign/130911_epo_country_profile_austria.pdf
European commission (2013): Women and men in leadership positions in the European Union, 2013. Retrieved from: http://ec.europa.eu/justice/gender-equality/files/ gender_balance_decision_making/131011_women_men_leadership_en.pdf
European commission (2014): Report on Progress on equality between women and men in 2013. Retrieved from: http://ec.europa.eu/justice/gender-equality/files/documents/ 141404_annual__report_en.pdf
Fassinger, R. E. (1994). Development and testing of the Attitudes Toward Feminism and the Women's Movement (FWM) Scale. Psychology of Women Quarterly, 18(3), 389-402. Feminism [Def.1, US English] In Oxford Dictionaries Online Retrieved November 15, 2014,
from http://www.oxforddictionaries.com/definition/american_english/feminism. Frijda, N. H. (1986). The emotions. Cambridge, England: Cambridge University Press.
Gamson, W. A. (1995). Constructing social protest. Social movements and culture, in ,Klandermans, B. & Johnston, H. (Eds.), Social Movements and Culture 4, University of Minnesota Press: 85-106.
Gamson, W. A., & Wolfsfeld, G. (1993). Movements and media as interacting systems. Citizens, Protest and Democracy (528), 114-125.
Geiger (2002): Mediale Vermittlung Feministischer Öffentlichkeiten in Neissl, J. (Ed.), Geschlechterperspektiven in den Medien (= Beiträge zur Medien- und
Kommunikationsgesellschaft 9). Vienna, Austria: Studien Verlag, 91-111.
Kahneman, D. (2002). Maps of bounded rationality: A perspective on intuitive judgment and choice. Nobel prize lecture, 8, 351-401.
Kim, Y. M. (2009). Issue publics in the new information environment selectivity, domain specificity, and extremity. Communication Research, 36(2), 254-284.
Klandermans, B. (1984). Mobilization and participation: Social-psychological expansions of resource mobilization theory. American sociological review, 49(5), 583-600.
Lugones, M. C., Spelman, E. V. (1983). Have we got a theory for you! Feminist theory, cultural imperialism and the demand for ‘the woman's voice’. In Women's Studies International Forum,6(6), 573-581.
Lind, R. A., & Salo, C. (2002). The framing of feminists and feminism in news and public affairs programs in US electronic media. Journal of Communication, 52(1), 211-228.
Liss, M., Crawford, M., & Popp, D. (2004). Predictors and correlates of collective action. Sex Roles, 50(11-12), 771-779.
