Erasmus Mundus Master's in Journalism, Media and Globalization
#EleNão, Women and Instagram: An Analysis of Mídia NINJA and the
Sense of Community in the 2018 Brazilian Elections
Camila Cabral de Mello Viero Student ID: 12906131
MASTER THESIS
Graduate School of Communication Supervisor: dr. Damian C. Trilling
May 29th, 2020
2 Abstract
The 2018 electoral process in Brazil was accompanied by many controversies. Among them the candidacy of Jair Bolsonaro, a conservative extreme-right candidate who uses his political speeches to attack minority groups such as women. In response to these attacks, women from hundreds of Brazilian cities joined forces in social media to try to stop Bolsonaro's election. The protests, which began on the Internet, were soon included in the streets of various localities in the country and the world. In order to join forces with the online women's community that emerged during this period, alternative left-wing media channels, such as Mídia NINJA, began to make posts also contrary to Bolsonaro and often directly aimed at its female users.
This study aims to understand whether the Instagram @midianinja account contributed to mobilize its female audience during the protests that took place on September 29, 2018, during the Brazilian election period. To this end, a quantitative content analysis was carried out from the selection of 102 posts and 1902 comments written exclusively by women. The research was based on theories about online communities and online feminism, taking into account especially the #EleNão movement, which joined forces to help women from all over the country in order to have their voices heard.
Although it found some trends for the contribution of alternative media to the mobilizations, the study did not find statistically relevant results in the analyses of posts directly targeted to women. However, interesting results were found regarding the counter-reactions generated by posts directly contrary to Jair Bolsonaro, demonstrating a strength of the public in favour of the former candidate.
Key Words: Jair Bolsonaro, alternative media, online community, social media,
3 #EleNão, Women and Instagram: An Analysis of Mídia NINJA and the Sense of
Community in the 2018 Brazilian Elections
The Brazilian electoral process in 2018 took place amidst numerous political scandals and polemics related mainly to former candidate and current President Jair Bolsonaro. Loaded with sexist, homophobic and racist rhetoric, Bolsonaro's speeches, while positively reached a large section of the more conservative population, also generated aversion from groups considered minorities. Among these groups were women, who took millions of people to the streets in order to protest against the former candidate. This process happened by means of rapid and extensive spread of online feminism, a form of activism which "has empowered one individual to prompt tens of thousands to take action on a singular issue – within minutes" (Martin & Valenti, 2012).
In recent decades, researchers have been studying online feminism on a large scale, as the movement becomes increasingly accessible to a large portion of society. As explained by Martin and Valenti (2012), it has the aim of sharing stories, discussing problems and organizing collective actions. All these factors are discussed and organized in online platforms, such as social media, the same environment used for online journalism. In the specific case of Brazil, 2018 was a major milestone for this movement, bringing to the surface the strength of online activism in relation to political and social movements. Although many studies deal with this specific subject, there are considerably few studies relating the strength of the online feminist movement to online media participation. Further specifying, there are practically no studies linking feminist activism and those considered alternative media. In Brazil, an alternative media channel – which are usually made up of journalists with leftist political ideals – was widely recognized for its support to the women's movement. The group, called Mídia NINJA, actively participated in the presidential election's process, evoking the society to the
4 protests, performing live coverage and sharing photos and videos taken by their
followers. As most of the mobilizations and protests were led by women, consequently Mídia NINJA's posts were focused on this specific audience. Based on the information presented, this work assumes that, due to its wide online reach among the Brazilian population, the @midianinja Instagram account may have had a significant impact to strengthen a sense of community among its female audience. Therefore, in order to understand the relationship created between the public, the media and the
demonstrations, this article aims to analyse to what extent has Brazilian’s alternative
media contributed to mobilize women during the protests occurred in the 2018 Presidential election campaign?
The characteristics that composed this process may be directly related to what researchers call as the notion of community, when a group of people, for instance, is "fundamentally related to each other as 'whole persons' through social bonds and value consensus that enable stable relationships among the members" (Yuan, 2012, p.667). However, as highlighted by Yuan (2012), it is necessary to take into account that online interactions are more fluid and have their own particularities. Contemporary researchers argue that geography, for example, is no longer a depending factor to create a notion of community, as nowadays the online networks approximate large numbers of individuals – no matter their physical locations – in "support of an activity, interest or identity" (Faraj, Krogh, Monteiro, & Lakhani, 2016, p. 668). Moreover, the online bounds are composed by members who share common interests, generating and establishing intense feelings of solidarity, empathy and support (Yuan, 2012; Zhou, 2011). Following these theories, it is also possible to relate the movement to the concept presented by Fraser (1990) about counterpublic. This notion, according to the author, represents groups outside the dominant public presented in the "public sphere". This idea proposes that
5 members of minority groups, such as feminists and women who participated of the Brazilian protests, "invent and circulate counterdiscourses, which in turn permit them to formulate oppositional interpretations of their identities, interests, and needs" (p.67).
Theoretical Framework Political movements' activism
Social and political movements such as #EleNão have, in their main
characteristic, the establishment of a collective identity that represents values of an individual within a group. According to Melucci (1995), the notion of identity is
referred to three features: "the continuity of a subject over and beyond variations in time and its adaptations to the environment; the delimitation of this subject with respect to others; the ability to recognize and to be recognized" (p.45). When an individual actor faces a collective conflict, for example, he or she and automatically feel a bond with a group. This happens, as explained by the author, not because they share the same interests, but because "they need this bond in order to make sense of what they are doing" (p.48). This action reinforces the notion of identity among the group members, as well as a feeling of internal solidarity, which can be represented by participation in forms of collective mobilization or in social movements.
These movements, when facing political issues in a democratic system, allow citizens to communicate with the government regarding their concerns and preferences. As pointed out by Memoli and Vassallo (2016), citizen's activism in political
movements represent a basic condition of democracy. One of the major characteristics of this factor is the participation in street demonstrations, which is viewed nowadays as an example of a "well-planned effort to intensify the impact of people's dissatisfaction, without waiting for elections" (p.7). However, the new technologies such as Internet also changed the way citizens become politically active, and yet, these changes do not
6 bring a consensus among scholars. Memoli and Vassallo (2016) point out that, in one hand, researchers emphasize that online activism – or "slacktkivism" (p.9) – represents the lack of real effort on the political participation of citizens when compared to previous activists, such as the street demonstrations. For other scholars, however, the online activism brings many benefits. They argue that the use of technologies in political movements has revigorated the participation of young generations, expanding the profile of citizens. These authors emphasize the fact that individuals who share common beliefs and have similar values are attracted to the same online groups, embracing new political debates in places where political apathy was mainly present (p.10).
Online Communities and the rise of Online Feminism
In the past decades, some authors have defended that a variety of new information and communication technologies, characterized as Web 2.0, have
fundamentally altered the nature of community building, collaboration and organization in economic and social life (Faraj et.al, 2016). Nevertheless, before emphasizing the notion of community inside digital environments, it is necessary to understand the pure sense of community. Regarding the authors, the traditional definition of community recognize a dual nature for the term, where it can be either considered as a space that binds individuals to a geographic locale, or alternatively as an attractor built on a
commonality of bonds, desires, shared features, or common interests (Faraj et.al., 2016). Yuan (2012) complements explaining that the sense of community built by groups is directly related to geographic proximity. In addition, it is also defined by close social relationships, encompassing the idea that human beings are fundamentally related to each other as "whole persons" (p.667), creating social bonds and setting values that enable stable relationships among the members.
7 Although these concepts are often separated by academia, some researchers reinforce the similarities between online and offline communities. To Faraj et.al (2016), for example, it is possible to see how the concept of community is also reflected in two commonly accepted definitions of online community. While Kraut and Resnik (2012, as cited in Faraj et.al, 2016), emphasize the definition of Online Community as a virtual space where individuals reunite with others to converse, exchange information, to learn, play or just to be with each other, other authors such as Sproull and Arriaga (2007, as cited in Faraj et.al, 2016) prefer to highlight the online community as a space of shared experiences, conditions, goals and convictions which create a common interest among its members.
Consequently, online communities can be characterized as groups of people which brings together geographically dispersed individuals in support of activities, interests and/or identities. It also has the power to intensify feelings of solidarity and empathy, supporting interactions in digital environments, which helps to legitimize the conception of digital spaces and communications as a community (Faraj et.al, 2016; Yuan, 2012). However, besides the fact that online communities rely on social media, they are not solely constituted by it. As explained by Faraj et.al (2016), "social media can support communities and their interactions, and facilitate community emergence. However, the main focus of social media is on individual networks and the ties users have built up" (p.671).
Nevertheless, while Internet scholars were studying about the future of digital environments, distinct groups – such as women – started visualizing new opportunities online. For them, this could be a chance to find a political platform of freedom, which would allow them to be stronger and to have their voices heard. As pointed out by Schulte (2011), the Internet had the possibility to offer and escape from gender and
8 gendered bodies, being a space where women wouldn't be constrained by patriarchy and misogyny anymore. However, according the author, this was proven to be an utopian view, as while the Internet could represent a liberatory space for women, it was also an inherently male-dominated technology that would "reinscribe oppression onto the new sphere and would become the virtual continuation of (or worse, a new mean of) real-world gender and sex-based marginalization" (p.735).
Even though the online spaces could not end the misogynist oppression, it gave a new mean to spread women's voices. Some theorists consider that technologies are an advance in providing social and political participation to women (Puente, Cupeiro & Romero, 2016). In Latin America countries, for example, the advocacy of gender equality has expanded along with the development of the Internet, considered a powerful activism tool, having the potential of galvanizing international responses to local violations of women's rights (Friedman, 2005). Puente et.al (2016) explain that, in this context, "the feminist praxis seems to have oriented its efforts towards the creation of new relationship spaces, or online communities that, in some cases, maintain a close relationship with other offline communities" (p.862).
Social media platforms, for instance, are the environments where many of these social and political discussions are taking place, transforming the structure of gender equality advocacy. Its facilitation of rapid interaction and information exchange,
enabling thousand, even millions to take action on a singular issue within minutes, helps to foster feminist ideals while expanding networks that undergird gender equality
advocacy and activism (Martin & Valenti, 2012; Friedman, 2005). According to Matos (2017), this phenomenon can be characterized by Brazilian women, who do not feel well represented by the mainstream media. As a consequence, they have increasingly made use of Internet for self-expression. The author explains that platforms such as
9 feminist blogs, provide online spaces for debates for various groups of Brazilian
women, who are increasingly being able to publish online and argument on "political issues which have overwhelmed the public agenda in the last years" (p.418).
Thus, considering the concepts of online community and online feminism presented, and adding the fact that contemporary communities are based on the sense of belonging on diverse structures such as religion, nationalism, ethnicity, lifestyle and gender (Delanty, 2003) – being gender the most important social and cultural structure in this study –, I first propose the following hypothesis:
H1: Posts targeted directly to women had more engagement (number of likes and/or comments) than posts that were not targeted.
The argument here is that digital tools, such as Instagram, provide the
opportunity for the articulation of ideas and for networking between women in order to mobilize around agendas that can influence the public sphere around gender subjects (Matos, 2017), such as Bolsonaro's speeches against women. In this case, it is possible that the content generated on @midianinja Instagram during the 2018 election campaign have influenced women to engage mostly in posts aimed directly to the female public, as it may have created a space for gender activism.
Instagram and online Journalism
Combined with the Internet, the expansion of Information and Communications Technology (ICT) (Seoane & Hornidge, 2018) and the use of smartphones were also followed by the emerging possibility of a new public space that gives opportunities for players – such as new collective groups, traditional media and the population – to position themselves, participating in discussions on relevant issues and even creating movements that guide the public agenda (Lévy, 2017, as cited in Ramalho & Maia,
10 2018). Here, social media and applications such as Instagram are included as a new possibility of communicating in a public space.
Despite seeing millions of selfies, paradisiacal landscapes, pets and far-fetched dishes, this is not all that the 1 billion monthly active users – being at least 77 million of them Brazilians1 – find when they open the platform every day. This can be justified by the fact that journalism organizations all over the world are learning how to take
advantage of the "attraction of the public to the storytelling capabilities of the site" (Byrd & Denney, 2018, p.47). As an example, The New York Times account on Instagram (@nytimes), in April 2020, counted with 8.4 million followers, while the BBC News (@bbcnews) account had, in the same period, 12.2 million followers. These presented numbers confirm how the possibilities of multiplatform distribution in
journalism are constantly being multiplied and potentiated. These digital
communication environments are increasingly used for information purpose, creating different options and trends for the production of the news (Zago & Belochio, 2017). The organization of strategies used in digital spaces such as Instagram follow the principles of "Transmedia Narrative" (Jenkins, 2008, as cited in Zago & Belochio, 2017), which may be "altering the expectations of audiences in terms of what
newspapers can offer, as they create new forms of construction, exposure and access to journalistic content" (p.790). These narratives open a space for engagement and
collaboration of its public, providing the reader with a new role within the practice of journalism, to which they have now an active participation.
Emotions on social media
In order to connect the media influence with female activism, it is also essential to understand the role of emotions during the construction of a social and political
1Leading countries based on number of Instagram users as of January 2020 (in millions): https://www.statista.com/statistics/578364/countries-with-most-instagram-users/
11 movement. Instagram and other social media platforms allow its users to share thoughts, opinions and feelings with other individuals in a very easy way. The simplicity of it provides various opportunities for ordinary people to engage in a wide range of social activities (Lee & Chau, 2017), including politics activism. According to Hou and Bonanno (2018), in the past years the world has witnessed increasing groups that are affected by socio-political movements aiming social justice and democracy. In
accordance with several academics, in this context we must consider the importance of emotions, as they are indispensable under the spotlight of social and political movement and are considered prominent drivers of action (Hou & Bonanno, 2018; Au & Chew, 2018), having the capacity of directly influencing the behaviour of individuals.
As explained by Lee and Chau (2017), the socio-political context represented in social media can be charged with intense emotions, as the "controversial and polarizing" (p.23) nature of politics is expected to trigger sentiments and, consequently, draw public attention. This dynamic has the power to impact mobilizations, demobilizations, and solidarity among people during social movements, and it is the sanctioning emotion that encourages them to show dissatisfaction with individuals or governments that maintain contrary postures (Hou & Bonanno, 2018). Thus, when people understand the sources of the political threat, they start to powerfully mobilize the movements and protests, being the last one considered a form of "collective action and rely heavily on
participants’ sense of belonging to the movements and whether the cause of action resonates with the participants’ emotions" (Hou &Bonanno, 2018, p.121). In this
manner, emotions such as indignation, dissatisfaction, pride, among others, contribute to create a sense of belonging from individuals to certain groups that defend the same ideals.
12 Nevertheless, it is necessary to emphasize that emotions generated first in digital environments need to be analysed differently than in offline environments. As explained by Au and Chew (2018), emotions are usually expressed in a combination of
non-linguistic mechanisms – such as facial expressions and vocal intonations. However, the authors emphasize that textual communication, such as the majority of social networks, it is possible to present modes of expression no less rich than in the face-to-face
communication, as they can be amplified by expressive textual methods such as emoticons, hashtags and slangs. It is thus argued that users may rely on verbal communication strategies, such as content and linguistics, in order to compensate for the lack of nonverbal cues in the online environment (Waterloo, Baumgartner, Peter, & Valkenburg, 2018, p.1815).
The fact is, even in an environment where face-to-face communication is not present – or at least is not the main mean –, the content generated by groups and individuals can be rich in stance, opinions and evaluations. Huffaker (2010, as cited in Lee & Chau, 2017), also noted that in an online space – giving an example of discussion forums –, messages charged with emotions tended to receive more feedback than the ones with no emotion at all. Being so, it is believed that the content shared at
@midianinja during the 2018 electoral period are likely to trigger emotions while disseminating facts and opinions. As explained by Jasper (1998), "affective and reactive emotions enter into protest activities at every stage" (p.404). The author interprets that emotions are part of collective actions, which reinforce a sense of solidarity among the group, called by him as a "we-ness" (p.418). This way, adding emotions to the
previously presented concepts of online community and online feminism, the second and third hypotheses proposed are the following:
13 H2: Post targeted directly to women evoked more positive emotions in the comments than posts that were not targeted.
H3: Posts targeted directly to women had the sense of community/"we-ness" more present in the comments than posts that were not targeted.
Finally, it is important to analyse if emotions related to groups against Jair Bolsonaro might have triggered possible counter-reactions. For this reason, the following sub-research questions will also be taken into consideration:
RQ1: Did posts directly contrary to Jair Bolsonaro (i.e. using the #NotHim) lead to a greater number of counter-reactions?
RQ2: Did the use of positive emotions in the posts' descriptions lead to a greater number of counter-reactions?
Methods Case under study
Mídia NINJA activism
In order to contextualize the choice of case under analysis in this research, it is necessary to take into consideration some facts regarding the media landscape in Brazil. Although Internet access is considerably high in the country – 74.9% of its households have access to Internet2, being 97% within this percentage accessed by smartphones –, the use of television as the main means of communication is a reality. As explained by Barros (2018), in the Brazilian current outline the media system is characterized by being a "Trusteeship Model" (p.3), when the State has the option to privilege media
2IBGE (2018) - Brazilian Institute of Geography and Statistics: https://agenciadenoticias.ibge.gov.br/agencia-sala- de-imprensa/2013-agencia-de-noticias/releases/23445-pnad-continua-tic-2017-internet-chega-a-tres-em-cada-quatro-domicilios-do-pais
14 activity as private and commercial. The system is also considered "No Law" (p.3), because of the absence of legislation regulating it, and without concerns with the so-called crossed ownership – when the same group controls different media, such as TV, radios and newspapers. In this context, political and family oligarchies are consolidated, controlling not only the media, but sometimes the politics of certain regions. Taking into consideration the traditional media – such as radio and television broadcasting –, only six national conglomerates and five regional media groups, through partnerships with local groups, reach almost a hundred percent of the Brazilian territory (Cabral, 2016). However, during the four consecutive terms of office of the Workers' Party (PT) in Brazil, initially led by Luiz Inácio Lula da Silva and later by Dilma Rousseff – from 2003 to 2016 –, two main factors have put pressure on the predominance of private mass media. According to Seoane and Hornidge (2018), first, the economic growth in the country has facilitated the expansion of ICTs and smartphones among the
population. Secondly, Lula introduced policies in the cultural sector that indirectly influenced the media sector. Consequently, these policies "stimulated new practices, thus challenging pre-existing commercially ones, influencing the discourse of many cultural and media organizations" (p.8).
Among these organizations, is the "Independent Narratives, Journalism and Action" collective, known as Mídia NINJA. This online media is classified as a left-wing alternative media, which according to Seoane and Hornidge (2018), relates to any media organization striving for social change in its particular context of operation. Created in 2013, the group bets on "the exchange of direct information between people, without the presence of the old intermediaries" (Barros, 2018, p.9), and it aims to finance its work with its readers, demystifying the tradition of journalism and placing a network of collaboration in content production. This new type of media belongs to a
15 group or collective, as presented by Ramalho and Maia (2018), that address their
discourses towards "combating social inequalities and prejudices, fighting for the realization of rights, and effectively carrying out a social activism based mainly on ICTs, constituting so-called mediativism" (p.188).
Midia NINJA started to gain ample attention already in 2013, achieving high level of followers on social media. The accelerated growth occurred due to its original coverage of several riots in the country, which initiated as a demand to reduce the public transport fares, and soon incorporated other factors such as the fight against corruption and against the 2014 World Cup (Saad-Filho, 2013 as cited in Seoane & Hornidge, 2018). Their coverage became famous after transmitting scoops of police violence that, at first, the mainstream media ignored. According to Saad-Filho (2013, as cited in Seoane & Hornidge, 2018), by streaming the news using smartphones through social media, the group had an unprecedent chance to make the citizen's voice public. All of the content was created with little or none editorial control, contrasting with
broadcasting media.
For this specific research, Mídia NINJA was chosen as the object of study as result of its large number of followers on social media, as well as for its direct and substantial involvement in covering the #EleNão movement. It will be analysed only the Instagram account (@midianinja), which had, until April 2020, 2 million followers.
The #EleNão movement
In the second half of 2018, a major movement has emerged in Brazil illustrating how the strength of online communities can reach the political arena of a country. It started after several controversial speeches by the former candidate Jair Bolsonaro, which were mainly targeted to political opponents, public policies and social minorities, such as LGBTQI+ (Lesbian, Gay, Bisexual, Transgender, Queer, Intersex +), black
16 people, women and immigrants. On October 29, 2018, CartaCapital – a Brazilian
magazine –, illustrate some of the problematic statements. Perhaps the most emblematic one happened in 2003, when Bolsonaro said to Mrs. Maria do Rosário – a Brazilian Congresswoman – in front of several journalists, "I would never gonna rape you because you don't deserve it". Years later, in 2014, he said that his colleague "does not deserve [to be raped] because she is very ugly, she is not my type, I would never rape her" (CartaCapital, 2018). Besides having caused legal consequences for Bolsonaro, the declarations also led to a great rejection of the Brazilian female electorate, which during the electoral polls of 2018 reached 52% (Jiménez, Mendonça, Oliveira, Rossi & Coletta, 2018). This significant number was soon reflected on social media, as with the creation of groups such as Mulheres Unidas Contra Bolsonaro (Women United Against
Bolsonaro) on Facebook. After two months of its creation, the group had already reached more than 4 million women, and soon these online protests also took on a big dimension on the streets.
On September 29, 2018, millions of women filled several cities in Brazil and around the world in the act called #EleNão (#NotHim). In a statement to BBC Brasil at that time, Céli Regina Jardim Pinto, author of the book A History of Feminism in Brazil, said that the protest was the largest manifestation of women in the history of the country (Rossi, Carneiro & Gragnani, 2018). In addition, she explained that the
movement has embraced agendas that go far beyond feminism. "It started with women, because Bolsonaro said low-level phrases in relation to women, and it encompassed many things, such as the defence of democracy and human rights", she added (Rossi et.al, 2018). Therefore, the identity of the protests was widely recognized by the use of a symbol commonly used on social media: the hashtag. As explained by Recuero, Zago, Bastos and Araújo (2015), the use of hashtags allows social media users to participate in
17 creating and sharing particular resources through tags, which can be multiple because its meaning is socially constructed. According to Pinheiro-Machado and Burigo (2018), the #EleNão cannot, however, be considered just a simple hashtag, but an extraordinary base movement that at the same time organizes a political act and serves as a focal point for other women's movements, online and face-to-face. Daflon (2019) complements, explaining that the mobilizing success of the slogan #EleNão is an expression of the scale and plurality of contemporary feminism.
Research Design and Sampling
Taking into account the growth of alternative media and its important contribution to the #EleNão movement in Brazil, the present study aims to analyse Mídia NINJA's impact in mobilizing women during the 2018 protests against the former candidate Jair Bolsonaro. For this purpose, the research is conducted through a
quantitative content analysis, which assigns numeric values to symbols of
communication – such as words and emojis – in order to look at the data by means of a systematic and replicable examination, using statistical methods to describe the
communication (Riffe, Lacy, Watson, & Fico, 2019, p.23).
Considering the relevance of the Mídia NINJA's Instagram account on the left-wing political scene in Brazil and its significant background in covering street
demonstrations, the @midianinja user was chosen as the case study of this research. The specific choice to analyse the media collective in Instagram was due its largest number of followers when comparing to Twitter, the second biggest social media channel of the group (2.2 million on Instagram against 641,000 on Twitter3). The timeframe selected for the analysis was from 27 September 2018 to 1 October 2018. This period includes the date of the main protest during the presidential elections on 29 September and has a
18 two-day window before and after the protest. This choice was made to increase the coverage time, as before the 29th the group was already carrying out postings with the theme of the protests. And after the date, they continued with the posting of photos and videos of the events that took place in Brazilian cities and around the world. In total, from the selected dates, @midianinja made 508 posts among photos, videos, designed arts and print screens. In order to select the data for this study, a stratified sampling combined with a systematic random sampling was carried out, where 1 in each 10 posts were selected, resulting in a total of 102 posts.
Regarding the comments, only the participation of women was taken into
account, since the study and the hypotheses presented take into consideration the female public only. Thus, in each post up to 25 comments written by women were selected, disregarding male and/or unidentifiable users. Also, comments in which the user only tagged another user were not taken into account. It is necessary to emphasize that due to an issue on Instagram – when the website is opened on devices such as computers or laptops – posts with a very high comment count do not load the comments in their totality. Therefore, the site does not allow the complete view of all comments written in the publication. Consequently, the first comments available in each post were selected, totalling in 1902 comments for the analysis.
Operationalization
Due to the lack of empirical research in this specific topic of analysis, a specific codebook – available in Appendix A – was developed as the instrument of quantitative data collection. The codebook was based on the previous literature presented, taking mainly into account the sense of community, or "we-ness", as presented by Jasper (1998), and the emotions classification created by Page and Mapstone (2010). Due to
19 particularities in Bolsonaro's campaign and after a random look at the selected posts and comments as a primer step, additional emotions were included in these categories.
The first group of items was developed at the post level. This group measured the amount of posts – including photos, images, videos and words – that were directly targeted at the female audience, the amount of posts that were contrary to Jair
Bolsonaro, and how many posts included positive emotions in words and/or emojis. It is worth pointing out that no posts posted on the selected dates were in favour of
Bolsonaro. All were against him or neutral/factual. The second group of questions was developed at the comment level. This group, for its part, measured the presence or absence of positive and negative emotions in words and/or emojis, the presence of words representing a female group, words and/or emojis referring to the feeling of pride, and words and/or emojis directly against or in favour of Bolsonaro.
Each variable was subjected to an intercoder reliability test conducted to a subsample of 20 posts and 122 comments and encoded by two coders. Only the Post Identification variable was excluded, as both coders assessed the same posts and comments. In order to measure the level of agreement, the Krippendorff's Alfha values were calculated in one round, in SPSS, using the Kalpha macro (Hayes & Krippendorff, 2007). All the variables measured resulted from good to excellent. In the post level, three variables reached α = 1.0; and three between 0.77 < α < 0.9. In the comment level, eight variables reached between 0.85 < α < 0.97; and one reached α = 1.0. For a
complete overview of the intercoder reliability of the variables, see Appendix B. Descriptive Summary Statistics
In order to better visualize the data collected for the analysis, tables 1 and 2 present the statistical summary descriptions of each variable considered in this study. The first table shows data of the variables in the posts level – total of 102 posts for
20 analysis – while the second table presents data of the variables in the comment level – 1902 comments in total.
Table 1. Descriptive Summary Statistics in the Post Level.
Table 2. Descriptive Summary Statistics in the Comment Level
21 The data presented at table 1 represent the descriptions among the 102 posts selected for this research. All the data collected was measured in the nominal level, taking into account 0 as No, 1 as Yes, and 99 as non-identifiable. Only the first four variables could be answered as non-identifiable. In order to achieve these outcomes, all the non-identifiable answers were recorded as missing values so as to avoid interfering with the binary results. The data that resulted in Table 2 included the comments present in each post, which means that the results presented in table 2 show the average of "yes" responses among the evaluated posts for each variable. The data were also obtained in the nominal level. However, it did not contain the non-identifiable option among its answers.
Findings
The first hypothesis, which predicts that posts targeted directly to women had more engagement (number of likes and/or comments) than posts that did not describe the female audience, was measured by running an independent-samples t-test in SPSS. The test was conducted in order to compare the means regarding the number of likes and comments in both variables (targeted to women and not targeted to women). From the 102 analysed posts, 60 were directly targeted to women, 19 were not directly targeted to women, and 23 were considered as non-identifiable – when it was not possible to identify the gender or the feminist symbols in photos and/or videos. The 23 non-identifiable posts were excluded from this specific analysis. The independent t-test showed that even though it seems that posts that were not targeted to women had more likes (n = 19, M = 17551.26, SD = 14199.46) than those which were targeted (n = 60, M = 14380.67, SD = 12961.65), this difference was not significant; t(77) = 0.908; p = 0.36, 95%CI = [-3780.86, 10122.05]. In addition, when analysing the average of comments under the same variables (targeted or not targeted to women), posts not targeted to
22 women also had more comments (n = 19, M = 488.79, SD = 646.69) than posts that were directly targeted to women (n = 60, M = 200.68, SD = 274.94), demonstrating a slightly but non-significant difference in the scores; t(20) = 1.89; p = 0.07, 95%CI = [-30.00, 606.21]. The null hypothesis is supported, demonstrating that posts directly targeted to women did not result in more engagement when compared to posts not targeted to women (see Figure 1).
23 With the aim of testing H2, which states that posts targeted directly to women evoked more positive emotions in the comments than posts that did not describe the female audience, an independent t-test was also performed. In order to measure the variables, the non-identifiable posts (n = 23) were again excluded from the analysis. The results showed that there was no significant difference between the positive emotions in the non-targeted posts (n = 19, M = 11.63, SD = 7.22) when compared to the use of positive emotions in posts targeted to women (n = 60, M = 10.65, SD = 5.85); t(77) = 0.60; p = 0.55, 95%CI = [-2.27, 4.23]. For the purpose of comparison, negative emotions were also measured among the same conditions, which resulted also in no significant differences between targeted (n = 60, M = 4.27, SD = 8.34) and non-targeted (n = 19, M = 4.32, SD = 8.22) posts; t(77) = 0.022; p = 0.98, 95%CI = [-4.31, 4.40]. Once again, the hypothesis was rejected, indicating that the use of positive emotions in the comments was not higher in posts that were directly targeted to women, when compared to posts that were not targeted to women (see Figure 2).
The same independent t-test was also conducted in order test the H3, which indicates that posts targeted directly to women had the sense of community – or "we-ness", as presented by Jasper (1998) – more present in the comments than posts that did not describe the female audience. This hypothesis was measured in two combined ways, taking into account comments that referred to groups of women (female words of the Portuguese language) as well as whether the comments contained the positive feeling of pride measured in words and/or emojis. Posts directly targeted to women (n = 60, M = 8.65, SD = 5.39) had more presence of the sense of community/we-ness than posts not targeted to women (n = 19, M = 7.89, SD = 5.74). Although the averages indicate a trend towards a greater presence of a sense of community/we-ness in posts aimed
24 directly at women, the results showed no significance (t(77) = -0.52; p = 0.60, 95%CI = [-3.62, 2.11]), so the hypothesis cannot be statistically confirmed (see Figure 3).
25 Figure 3. Confidence Interval on the presence of sense of community/we-ness.
Finally, the two sub-research questions were also analysed from an independent t-test. The RQ1, which aimed to find if posts directly contrary to Jair Bolsonaro lead to a greater number of counter-reactions, resulted in significant differences. Posts directly against Bolsonaro (n = 93, M = 1.29, SD = 1.96) demonstrated a greater number of counter-reactions when compared to neutral posts (n = 9, M = 0.00, SD = 0.00); t(92) = -6.33; p = 0.00, 95%CI = [-1.69, -0.88]. Here, it is important to emphasize that there were no posts in favour of Jair Bolsonaro. By means of comparison, it was also analysed the number of comments against Bolsonaro, taken the same two variables – against Bolsonaro or neutral – into consideration. Posts directly against Bolsonaro (n = 93, M = 4.92, SD = 3.14) had more comments against Bolsonaro than neural posts (n = 9, M = 0.78, SD = 1.20); t(21) = -8.02; p = 0.00, 95%CI = [-5.22, -3.07]. From these results, we can state that the posts directly against the former candidate Jair Bolsonaro led to a greater number of counter-reactions when compared to neutral posts (see Figure 4).
26 Figure 4. Confidence Interval on posts against Jair Bolsonaro and the
counter-reactions.
Regarding the RQ2, which aimed to find if the use of positive emotions in the posts' descriptions lead to a greater number of counter-reactions, a trend was found, but without significance. Posts which contained positive emotions (n = 46, M = 1.39, SD = 2.26) had more counter-reactions in its comments than posts considered neutral (n = 56, M = 1.00, SD = 1.56); t(100) = -1.02; p = 0.30, 95%CI = [-1.14, 0.36]). In this case, although the results are not significant, a trend was found in which there are more
27 counter-reactions in posts that used positive emotions in their description, when
compared to posts considered neutral (see Figure 5).
Figure 5. Confidence Interval on positive emotions and the counter-reactions.
Discussion
The extent of women's engagement in the battle against President Jair
Bolsonaro's candidacy during the 2018 elections is in accordance with the body of this study, which underlines the online platforms as a place where social and political actions are happening. These places are being able to transform the structure of gender equality advocacy and activism (Martin & Valenti, 2012; Friedman, 2005), opening an important space for online communities such as online feminism to grow. These changes also play an important role in the media channels, including the so-called alterative media, which are beginning to gain more space for coverage of events where traditional broadcasters often do not have enough space – or are even barred by editorial choices.
According to the events that took place during the 2018 election process in Brazil, this study aimed to identify possible connections between the role of alternative
28 left-wing media – represented in this research by the Mídia NINJA group – and the protests led by women. The main goal was to understand if there was any impact through the Instagram account @midianinja to the large demonstrations that took place in hundreds of cities in Brazil and around the world. To this end, data from the spectrum of posts and comments were analysed, in which the target audience, existing positive and negative emotions, words aimed at specific groups of women, the feeling of pride, and opinions directly against or in favour of Jair Bolsonaro were taken into account.
Although there are a large number of posts directly aimed at women, sharing photos and/or videos exalting the participation and leadership of the female public in the demonstrations against Bolsonaro, it was not possible to identify specific influences between the @midianinja posts and the protests. First, it was expected that posts targeted directly to women would have greater engagement than non-targeted posts. This hypothesis was denied, leading to the belief that engagement could be driven by other factors that were not measured in this study, such as themes directed to other communities also considered as minorities – for example LGBTQI+ and black people –, or even how is the image presented – if it is a photography, a print screen, a video, a poster. Considering that Instagram is a social media of strong visual impact, the image may have an important role in its users' engagement. As observed by Roberts (2005, as cited in Schwalbe, Silcock, & Candello, 2015), the Internet, including the social media platforms, is changing the traditional gatekeep model turning "solid 'gates' into little more than screen doors" (p.465), where the published image starts to be worth as even as or more than words. The same considerations may also have influenced the results obtained from the analysis of the outcomes of the second hypothesis, which was not confirmed, and measured if posts directly targeted to women had more positive emotions in the comments when compared to the not-targeted posts. In this case,
29 although the averages among the variables are considerably close, they cannot also be used as an argument to indicate the impact of the posts on the manifestations.
In relation to the third hypothesis – which predicted that posts directly addressed to women would have the greatest presence of a sense of community/we-ness in the comments –, although there is no statistical significance, the results showed a trend that could possibly confirm the hypothesis in future studies. Overall, the average on
comments that included words and/or emojis which indicate a sense of community/we-ness among women was slightly higher in posts directly targeted to women, showing that with a larger sample, there is a possibility of confirming the hypothesis. The data also indicate that in accordance with the previous theory presented, Latin America women have increasingly used social media as an activism tool in order to have their voices heard. As pointed out by Puente et.al (2016), they seem to have created new relationship spaces, called online communities, where they discuss actions, show solidarity and empathy with each other, and take their ideals to the offline spaces as well.
Lastly, when taking into account all the posts directly against the former candidate Jair Bolsonaro, the significance found between this variable and the counter-reactions generated in the comments was valuable. Relating these results to the themes often reported during the electoral process, it must be taken into consideration that, in addition to these posts possibly attracting the attention of people in favour of Bolsonaro, it may also have generated a greater amount of comments posted by bots. As explained by Ituassu, Lifschitz, Capone and Mannheimer (2019, as cited in Pimentel & Tesseroli, 2019), computational propaganda in the form of acting robots and manipulation of algorithms played a central role in the Brazilian political system. According the authors, the digital political communication of Bolsonaro's successful campaign presents clear
30 elements of the use of computational propaganda. From these elements, it can be
stipulated that part of the counter-reactions may have come from the automation of responses in favour of Bolsonaro. These characteristics may also have impacted the results obtained in the second sub-research question, which predicted that posts that used positive emotions would generate more counter-reactions in favour of Bolsonaro. Although these latter results are not statistically significant, they show a tendency to expect a greater number of counter-reactions in posts that have used positive emotions in its descriptions.
The results point to some limitations in this study. For future research, it would be relevant to increase the period of analysis in order to compare women's participation before, during and after the street demonstrations. Not only to measure whether or not women's participation on the @midianinja page increased during demonstrations, but also to understand if there was any change in the type of content commented by the users. Thus, it would be possible to test whether there was a considerable increase in the use of expressions indicating a sense of community/we-ness on the exact day of the protests when compared to prior and subsequent days, possibly relating women's activism in the Brazilian elections with the presented theoretical background on online feminism and online communities. Furthermore, in this analysis only comments written by women were taken into account. Although these comments strongly contribute to interpretations in relation to online feminist communities, they limit possible
comparisons with other genders and do not include transgender women in their theoretical definitions, who are widely ignored in theories about feminism. Therefore, the research could be enriched with comparative data between different genders, in order to have sufficient results to understand if groups formed only by women had a greater impact on @midianinja posts during the street demonstrations.
31 Regardless these limitations and taking into account that not many statistically significant results were found, it is possible to identify features present on
@midianinja's posts that reinforce the elements present in online communities. Women's participation in the political scenario has gradually grown over the past decades. And a great ally of this growth, specifically in Latin America, is the use of social media as a platform for political and social activism, being them mediatic or not.
32 References
Au, A., & Chew, M. (2017). How Do You Feel? Managing Emotional Reaction, Conveyance, and Detachment on Facebook and Instagram. Bulletin of Science, Technology & Society, 37(3), 127–137.
https://doi.org/10.1177/0270467618794375
Cabral, E.D.T. (2016) Mídia Concentrada no Brasil: Até Quando? (Concentrated Media in Brazil: Until When?) Revista Latinoamericana de Ciencias de la
Comunicación, v.13, n.24 (13).
<http://www.alaic.org/revista/index.php/alaic/article/view/725/409> Barros, B.M.C. (2018). A Democratização dos Meios de Comunicação e a
Descentralização da Informação no Brasil a Partir das Novas Mídias
(Democratization of the Media and Decentralization of Information in Brazil from the New Media). Direito e Cidadania, v. 2, 2018.
Byrd, R.D., Denney, P. (2018) Using Their Own Voice: Learning to Tell Stories with Instagram. Teaching Journalism & Mass Communication, Vol.8, nº2, pp.47-55 <http://www.aejmc.us/spig/journal>
CartaCapital (2018) Bolsonaro em 25 frases polêmicas (Bolsonaro in 25 controversial sentences). Oct 29th. <https://www.cartacapital.com.br/politica/bolsonaro-em-25-frases-polemicas/>
Daflon, V. (2019) Dilemas e Perspectivas dos Feminismos no Brasil Contemporâneo (Dilemmas and Perspectives of Feminisms in Contemporary Brazil). Sociologia e Antropologia, v.09.01: 315-320, Jan-Abr 2019.
Dantas, J., & Falcão, P. (2018). Mídia Ninja, Online Citizen Journalism and the Professionalization of Journalism in Brazil. Âncora, v.5 n.1, p.97-110. DOI: 10.22478
33 Delanty, G. (2003). Conclusion — Theorizing community today. Last chapter in
Community, 186-195. London: Routledge.
Di Carlo, J., Kamradt, J. (2018). Bolsonaro e a Cultura do Politicamente Incorreto na Política Brasileira (Bolsonaro and the Politically Incorrect Culture in Brazilian Politics). Programa de Pós-Graduação em Ciências Sociais - UFJF v. 13 n. 2 ISSN 2318-101x (online) ISSN 1809-5968 (print).
Faraj, S., Krogh, G., Monteiro, E., & Lakhani, K. (2016). Online community as space for knowledge flows. DOI: https://doi.org/10.1287/isre.2016.0682
Fraser, N. (1990). Rethinking the Public Sphere: A Contribution to the Critique of Actually Existing Democracy. Social Text, No. 25/26, pp.56-80.
Friedman, E. (2005). The Reality of Virtual Reality: The Internet and Gender Equality Advocacy in Latin America. Latin American Politics and Society, 47(3), 1–34. https://doi.org/10.1111/j.1548-2456.2005.tb00317.x
Gallego, E.S. (org) (2018). O ódio como política: a reinvenção das direitas no Brasil (Hatred as politics: the reinvention of the rights in Brazil).
https://bit.ly/2FNbyD2
Hayes, A. F., & Krippendorff, K. (2007). Answering the call for a standard reliability measure for coding data. Communication methods and measures, 1(1), 77-89. Hou, W., & Bonanno, G. (2018). Emotions in Everyday Life During Social Movements:
Prospective Predictions of Mental Health. Journal of Counseling Psychology, 65(1), 120–131. https://doi.org/10.1037/cou0000236
Jasper, J. (1998). The Emotions of Protest: Affective and Reactive Emotions In and Around Social Movements. Sociological Forum, 13(3), 397–424. DOI: https://doi.org/10.1023/A:1022175308081
34 Jiménez, C., Mendonça, H., Oliveira, R., Rossi, M., Coletta, R.D. (2018). Mulheres
quebram o jejum das ruas no Brasil com manifestações contra Bolsonaro (Women break the fast of the streets in Brazil with demonstrations against Bolsonaro). El País Brasil, Sep. 30, 2018.
https://brasil.elpais.com/brasil/2018/09/30/politica/1538270819_523141.html Lee, C., & Chau, D. (2018). Language as pride, love, and hate: Archiving emotions
through multilingual Instagram hashtags. Discourse, Context & Media, 22, 21– 29. https://doi.org/10.1016/j.dcm.2017.06.002
Martin, C., & Valenti, V. (2012). #FemFuture: Online Revolution. Barnard Center for Research on Women,8, 1-34.
Matos, C. (2017). New Brazilian feminisms and online networks: Cyberfeminism, protest and the female ‘Arab Spring’. International Sociology, 32(3), 417–434. doi: 10.1177/0268580917694971
Melucci, A. (1995). The Process of Collective Identity In Johnston, H., & Klandermans, B. (1995) Social Movements and Culture. University of Minnesota Press.
http://www.jstor.org/stable/10.5749/j.ctttt0p8.6
Memoli, V., & Vassallo, F. (2016). Political Activism Research: Studying the Evoution of Political Behavior. Partecipazione e Conflitto, 9(1), 1–18.
https://doi.org/10.1285/i20356609v9i1p01
Page, K., Dk, & Mapstone, M. (2010). How does the web make youth feel? Exploring the positive digital native rhetoric. Journal of Marketing Management: Academy of Marketing Conference 2010, “Transformational Marketing,” 26(13-14), 1345–1366. https://doi.org/10.1080/0267257X.2010.523709
Pimentel, P.C., Tesseroli, R. (Orgs.). (2019). O Brasil vai às urnas: as camoanhas para Presidente na TV e Internet. Systagma Editores.
35 Pinheiro-Machado, R., Burigo, J. (2018). #EleNão deixou de ser uma simples hashtag: é
um movimento feminista e político que pode mudar o Brasil (#NotHim is no longer a simple hashtag: it is a feminist and political movement that can change Brazil). The Intercept Brasil, Sep 28th
https://theintercept.com/2018/09/28/elenao-movimento-feminista-politico/ Puente, S., Cupeiro, S., & Romero, D. (2016). Ciberfeminismo contra la violencia de
género: análisis del activismo online-offline y de la representación discursiva de la víctima 1/Cyberfeminism against Gender-Based Violence: Analysis of online-offline Activism and the Discursive Representation of the Victim. Estudios Sobre el Mensaje Periodistico, 22(2), 861–877.
https://doi.org/10.5209/ESMP.54240
Ramalho, R., & Maia, K. (2018). Estudos metodológicos: a aplicação da Análise de Discurso para interpretar a produção de sentido de coletivos midiativistas (Methodological studies: the application of Discourse Analysis to interpret the production of meaning of mediativist collectives). Verso e Reverso, 32(81), p.187-200, Sep.-Dec. 2018. DOI: 10.4013/ver.2018.32.81.02
Recuero, R., Zago, G., Bastos, M.T., & Araújo, R. (2015). Hashtags Functions in the Protests Across Brazil. SAGE Open, April-June 2015: 1-14. DOI:
10.1177/2158244015586000
Riffe, D., Lacy, S., Watson, B., & Fico, F. (2019). Analyzing Media Messages: Using Quantitative Content Analysis in Research: Vol. Fourth edition. Routledge. Rossi, A., Carneiro, J.D., & Gragnani, J. (2018). #EleNão: a manifestação histórica
liderada por mulheres no Brasil vista por quatro ângulos (#NotHim: the historical manifestation led by women in Brazil seen from four angles). BBC News Brasil. https://www.bbc.com/portuguese/brasil-45700013
36 Schulte, S. (2011). [Review of Surfing Feminism’s Online Wave: The Internet and the
Future of Feminism (Book Review)]. Feminist Studies, 37(3), 727–744. Feminist Studies, Inc.
Schwalbe, C., Silcock, B., & Candello, E. (2015). Gatecheckers at the Visual News Stream: A new model for classic gatekeeping theory. Journalism Practice: Photojournalism and Citizen Journalism: Co-Operation, Collaboration and Connectivity, 9(4), 465–483. https://doi.org/10.1080/17512786.2015.1030133 Seoane, M.V., & Hornidge, A.K. (2018). The social shaping of media technologies’
multiple uses: the case of Mídia NINJA in Brazil. Information, Communication & Society, DOI: 10.1080/1369118X.2018.1500623
<https://www.tandfonline.com/doi/full/10.1080/1369118X.2018.1500623> Waterloo, S., Baumgartner, S., Peter, J., & Valkenburg, P. (2018). Norms of online
expressions of emotion: Comparing Facebook, Twitter, Instagram, and WhatsApp. New Media & Society, 20(5), 1813–1831.
https://doi.org/10.1177/1461444817707349
Yuan, E.J. (2012). A culturalist critique of ‘online community’ in new media studies.
new media & society, 15(5), 665–679. DOI: 10.1177/1461444812462847
Zago, G., & Belochio, V. (2017). Mapping the paths of Transmedia Journalism: the death of Teori Zavascki in Zero Hora and in readers' interactions.
Contemporânea comunicação e cultura, v.15, nº.03, set-dez 2017, pp.789-815. ISSN: 18099386
Zhou, T. (2011). Understanding online community user participation: A social influence perspective. Internet Research, 21(1), 67-81. DOI:
37 Appendix A
Codebook
I. POST LEVEL:
Q1: Does the photo and/or video in this post contain only female people (women, girls)? Please read the note1 before answering.
0 = No 1 = Yes
99 = It doesn't contain people
Note1: Please only take gender into consideration when it is clearly identifiable. If it is a crowd, a poster or someone whose gender cannot be identified, answer "no".
Q2: Does the photo and/or video contain one or more poster representing the female audience and/or the female cause (women's fight against Bolsonaro and/or sexism)? Please read the note2 before answering.
0 = No 1 = Yes
99 = Non identifiable
Note2: Posters that represent the female audience/cause may contain: female symbol (cross down with a circle, may contain a hand drawn within the circle), feminine nouns (she, her, woman, girl, girls, ladies, etc.), Portuguese language words written in the female gender (i.e., "fraquejada", "unidas", "representadas" etc.).
38 Q3: Does the description in this post refer to female protest participants? Consider "Yes" if the description is written with female nouns and/or Portuguese language words written in the female gender.
Please read note3 to get examples.
0 = No 1 = Yes
Note3: Feminine nouns may be: she, her, woman, girl, girls, ladies, etc. Portuguese language words written in the female gender may be, for example, "fraquejada", "unidas", "representadas", etc.
Q4: Does the post contain photo and/or video with text, images, posters, speeches that are directly against Jair Bolsonaro?
0 = No 1 = Yes
99 = Non identifiable
Q5: Does the post description contain words and/or hashtags that are directly against Jair Bolsonaro (i.e., #EleNão, #EleNunca)?
0 = No 1 = Yes
Q6: Does the post description contain words and/or emojis that represent positive emotions?
Please read note4 in order to have examples of words and emojis representing positive emotions.
39 0 = No
1 = Yes
Note4: consider positive emojis such as:
Hearts, hands clapping, heart eyes, closed hand as in the feminist symbol, arm strength emoji, flowers, kiss, tearfully, "high-five", rainbow, etc.
Consider positive emotions based on the classification below. These words are examples that may guide your response. You may identify words written with different spelling, or that are different from the example but have similar meanings. In these cases, answer "Yes".
Happiness Love Empowering Interested Pride Optimistic Happy Joyful Glad Excitement Good Amazed Ecstatic Content Enthusiastic Love Loved Connected Affection Kindness Beautiful Empowered Privileged Independent Free Alive Amazed Lucky Powerful Courageous Support Curious Entertained Informed Clever Intelligent Proud Successful United Hopeful Inspired Thankful Confident Valued Respected
Page, K., Dk, & Mapstone, M. (2010). How does the web make youth feel? Exploring the positive digital native rhetoric. Journal of Marketing Management: Academy of
40 https://doi.org/10.1080/0267257X.2010.523709
(added group 5 and 6 after this literature)
II. COMMENT LEVEL
Please, consider ONLY posts written by women/girls. If the gender is not identifiable, do not select this comment. For each post, select up to 25 comments, depending on its availability.
Q7: To which post this comment belong? Answer with the Post ID, from P1 to P102, according the code provided in the Excel sheet.
Q8: Does the specific comment under analysis present words that express positive emotions?
Please read note4 in order to have examples of positive emotions.
0 = No 1 = Yes
Note4: The words below are examples that may guide your response. You may identify words written with different spelling, or that are different from the example but have similar meanings. In these cases, answer "Yes".
Happiness Love Empowering Interested Pride Optimistic Happy Joyful Glad Excitement Good Love Loved Connected Affection Kindness Empowered Privileged Independent Free Alive Curious Entertained Informed Clever Intelligent Proud Successful United Hopeful Inspired Thankful Confident Valued
41 Amazed Ecstatic Content Enthusiastic Beautiful Amazed Lucky Powerful Courageous Respected
Positive emotions classification based on
Page, K., Dk, & Mapstone, M. (2010). How does the web make youth feel? Exploring the positive digital native rhetoric. Journal of Marketing Management: Academy of
Marketing Conference 2010, “Transformational Marketing,” 26(13-14), 1345–1366.
https://doi.org/10.1080/0267257X.2010.523709 (added group 5 and 6 after this literature)
Q9: Does the specific comment under analysis present words that express negative emotions?
Please read note5 in order to have examples of negative emotions.
0 = No 1 = Yes
Note5: The words below are examples that may guide your response. You may identify words written with different spelling, or that are different from the example but have similar meanings. In these cases, answer "Yes".
Anger Fear Shame Sadness Confusion
Frustrated Annoyed Apprehensive Stressed Scared Terrified Unsafe Defeated Naked Exposed Insignificant Small Sad Vulnerable Emotional Empty Confused Uninterested Distracted Tired
42 Denied Cheated Lost Inadequate Uncomfortable Antisocial Stupid Groggy Bored
Negative emotions classification based on
Page, K., Dk, & Mapstone, M. (2010). How does the web make youth feel? Exploring the positive digital native rhetoric. Journal of Marketing Management: Academy of
Marketing Conference 2010, “Transformational Marketing,” 26(13-14), 1345–1366.
https://doi.org/10.1080/0267257X.2010.523709
Q10: Does the specific comment under analysis contain emojis that represent positive emotions?
Please read note6 in order to have examples of emojis that represent positive emotions.
0 = No 1 = Yes
Note6: consider positive emojis such as:
Hearts, hands clapping, heart eyes, closed hand as in the feminist symbol, arm strength emoji, flowers, kiss, tearfully, "high-five", rainbow, etc.
Q11: Does the specific comment under analysis contain emojis that represent negative emotions?
Please read note7 in order to have examples of emojis that represent negative emotions.
0 = No 1 = Yes
43 Note7: consider negative emojis such as:
Broken heart, angry face, sad crying face, sad face, disgust face, nausea face, lifted middle finger, etc.
Q12: Does the specific comment under analysis contain words representing a female group (i.e., "juntas", "somos", "unidas", "elas", "nós", etc.) and/or a sense of belonging such as "eu fui", "eu tava", "eu estava", etc.?
As we are considering only women's comments, words such as "somos" and "nós" are included in this category.
0 = No 1 = Yes
Q13: Does the specific comment under analysis contain words referring to the feeling of pride (i.e., "orgulho", "lindo", "lindas", "arrepiante", "arrepiei", "choro de emoção", "emocionada", etc.)?
0 = No 1 = Yes
Q14: Does the specific comment under analysis contain emojis that represent pride (i.e., heart, hands clapping, heart eyes, closed hand as in the feminist symbol, arm strength emoji)?
0 = No 1 = Yes
Q15: Does this comment contain words and/or hashtags that are directly against Jair Bolsonaro (i.e., #EleNão, #EleNunca)?
44 0 = No
1 = Yes
Q16: Does this comment contain any counter-reactions in favor of Bolsonaro (i.e., #EleSim, 17, #Bolsonaro17, emoji of a hand imitating a gun, etc.)?
0 = No 1 = Yes
45 Appendix B
Krippendorff's Alpha values for codebook's variables
Nr. Variable Level of
Measurement
Kα Situation
Q1 FemOnly Nominal 0.7754 Not changed
Q2 FemPoster Nominal 0.9008 Not changed
Q3 FemPart Nominal 0.8030 Not changed
Q4 ImgAgainst Nominal 1.0000 Not changed
Q5 WordAgainst Nominal 1.0000 Not changed
Q6 PosEmot Nominal 1.0000 Not changed
Q8 PosEmot Nominal 0.8996 Not changed
Q9 NegEmot Nominal 0.8610 Not changed
Q10 EmojiPos Nominal 0.9753 Not changed
Q11 EmojiNeg Nominal 0.8667 Not changed
Q12 GroupAim Nominal 0.8805 Not changed
Q13 PrideFeel Nominal 0.8565 Not changed
Q14 EmojiPride Nominal 0.8893 Not changed
Q15 AgainstBolso Nominal 0.9246 Not changed
