God above everyone:
An analysis of the religious rhetoric used by presidential candidates during the
2018 Brazilian campaign on Facebook
Master’s Thesis
Authored by Mariana Sales de Oliveira (Student ID 12846910) Supervised by Linda Bos
Graduate School of Communication Date of Completion: 29th May 2019
Abstract
This study contributes to the scholarship on social media, politics, and religious communication by examining the use of religious rhetoric on Facebook by the four main candidates for the 2018 presidential elections in Brazil during their campaign. Throughout history, Christianity has played an important part in Brazilian society. More so, in the past decade, the country has faced a significant increase in the presence of Evangelical churches and, subsequently, a need for politicians to appeal to the voting bases of the Evangelical Caucus - a group of political actors that dialogue with conservative religious values. By applying key typology to identify the presence of such statements in Facebook posts from Jair Bolsonaro, Fernando Haddad, Ciro Gomes, and Geraldo Alckmin, this paper conducts a content analysis in order to find quantitative factors that measure the extent to which Christian discourse was shared on social media during their campaigns. Findings indicate that politicians use religious rhetoric in distinct frequencies and manners. Still, ideological affiliation is not necessarily what determines whether religious terms are mentioned or not. These conclusions are discussed in line with what has been previously researched in the United States and takes a step further by applying these studies to the Brazilian context.
Keywords: religious rhetoric, social media campaign, content analysis, Brazil, 2018 presidential elections.
Introduction
“Brazil above everything, God above everyone”. This was the slogan that led Jair Bolsonaro (PSL - Social Liberal Party) to his victory in the 2018 Brazilian elections, becoming the country’s 38th president, beating Fernando Haddad (PT - Worker’s Party) in the runoff. As found by the National Institute of Geography and Statistics (IBGE), the number of Evangelicals has shown an increase of 61.5% between 2000 and 2010 (Queiroz, 2019). As a political reflex, The Evangelical Caucus (In Portuguese, Bancada Evangélica) – an organized group of religious congressmen that articulates against themes such as the legalization of drugs, abortion and LGBT rights (Quadros & Madeira, 2018) – has gained power in the past elections, reaching the mark of 84 federal deputies out of the 513 members of the lower house of the National Congress in 2018.
Targeting this specific group of Christian voters, during the presidential campaign in 2018, Jair Bolsonaro settled several alliances with leaders and supporters of the group and embraced a stronger emotional appeal (Knolll, 2019) by often referencing Christian beliefs and values and discussing the impact that these would have in further policymaking. On the other hand, concerned about the difficulty of dialoguing with these voters in recent years, left-wing parties are beginning to devote more time to understand how they can better communicate approach these voters (Prado, 2019).
Concurrently, platforms such as Facebook and Twitter became a central pillar of political campaigning for not only Bolsonaro, but also his main competitors: Haddad, Ciro Gomes (PDT - Democratic Labor Party), and Geraldo Alckmin (PSDB - Brazilian Social Democracy Party). Politicians are increasingly choosing social networks over legacy media because, differently from the latter, online platforms do not have to comply with news them to bypass the journalistic gatekeepers” (Engesser, Ernst, Esser, & Büchel, 2017). An example of that is that Bolsonaro was elected against all the premises for a winning campaign in
Brazil: he had no TV time, no party structure, nor financial resources from a party fund – indicating that his massive participation online was paramount to his victory (Fratini, 2020). Several studies have been conducted on how, why, and with what effects political figures use religious discourse in their communication (Chapp, 2012; Coe & Chapp, 2017; Coe & Chenoweth, 2013; Domke & Coe, 2008; Kaylor, 2010; Shogan, 2006). Through dialoguing and forwarding what has been produced so far, this research proposes to use the typology of Christian discourse developed by American scholars Coe and Chenoweth and apply it to social media campaigning, more specifically to the Facebook campaigns of the four most voted candidates for the presidency in Brazil: Bolsonaro, Haddad, Gomes, and Alckmin. It is important to further highlight that, though starting by discussing the rise and influence of the Evangelical church, this paper looks into the role of Christian rhetoric in general, as it was done by Coe and Chapp, as well as Coe and Chenoweth. Based on the above-presented discussions, this paper is guided by the following research question (RQ):
To what extent did the four main Brazilian candidates for the presidency make use of the religious rhetoric in their Facebook posts during the 2018 campaign in Brazil?
Studies on religious rhetoric and politics so far have looked at formal speeches instead of social media. This study, therefore, proves to be specifically pertinent because it covers the concepts previously explored by taking one step further and looking at online campaigning while performing a quantitative content analysis. Also, considering that the elections in Brazil are a recent event, resources that focus on this specific event are still scarce. Additionally, the previously mentioned scholars have a particular focus on United States politics, whereas here the aim is to apply these findings to the Brazilian case, where cultural, economic, and political settings differ from the prior.
Theoretical Framework
To better comprehend the Brazilian conjuncture that influences the presence or lack of religious rhetoric by politicians in the country, it is first necessary to present a brief explanation of how the political system in Brazil works and the events that led to the 2018 elections. The Brazilian political system works under the model of a federal presidential representative democratic republic, formed by the Union, the 26 states, the Federal District and the municipalities. The head of state is elected by the people, by 4-years mandate. The current Brazilian democratic model was consolidated in 1989 (Codato, 2006), when the first direct elections since the 1960s occurred, after the Constitution was promulgated in 1988. Like other Latin Americas countries, Brazil has endured years of dictatorship. On 1 April 1964, the country suffered a coup d’état that led to 25 years of military ruling. The overthrow was supported by the Catholic Church and much of the media at the time (Sales & Martins Filho, 2018; Toledo 2014).
The 2018 elections
On October 7, 2018, Brazilians voted for the executive and legislative branches at both state and federal levels, choosing not only their president, but also governors, senators, federal deputies, and deputies (AS/COA, 2018). On the occasion, 13 candidates entered the electoral run to become the head of state of the country. As a result, the four politicians that got the highest amount of votes were Jair Bolsonaro (46.03%), Fernando Haddad (29.28%), Ciro Gomes (12.47%) and Geraldo Alckmin (4.76%). According to the Brazilian Constitution, in order to get elected, a candidate must obtain over 50% of the valid votes (Studart, 2013). Since that was not the case in 2018, a second-round was called, taking place on October 27. After three weeks of intense campaigning, Jair Bolsonaro was elected with 55% of the votes, beating Fernando Haddad (45%).
The elections were marked by mass polarization (Borges & Vidigal, 2018), with the parties defending clearly distinct positions, and the voters developing polarized preferences concerning their political values. Social media and polarization on the Internet had already played an important part in the previous elections in 2014, when intense debates took place on social media, especially on Facebook and Twitter, where it was possible to observe the formation of distinct networks (Goya, Fiacadori, & Santos, 2019). These events were followed by protests for the impeachment of former president Dilma Rousseff, which were organized online and were also marked by the polarization of political identities (Penteado & Lerner, 2018).
According to Carreirão (2014), Brazil is a fragmented country when it comes to its party system. As of 2018, the country had 32 political parties. Amidst so many options, for some scholars, the notion of left and right in Brazilian politics are not quite demarcated, which can be explained by the different historic and political periods that the country has undergone (Bolognesi, Babireski, & Maciel, 2019; Tarouco & Madeira, 2013). Still, in Brazil “those who express right-wing ideology are also very frequently conservative” (Tarouco & Madeira, 2013, p. 99). As stated by Brazilian political scientist Claudio Couto, in the political spectrum, Haddad’s PT is ideologically classified as left-wing, Gomes’ PDT as center-left, Alckmin’s PSDB as center-right and Bolsonaro’s PSL as extreme-right (Moraes, 2018).
Religion in Latin America
More than any other religion, Christianity is the most predominant one in Latin America, which can firstly be historically explained. Believers of this faith have been in the region for about 500 years, when European colonizers first settled in, forming a strong relationship between Christianity and the social order during three centuries of Spanish and Portuguese control (Benavides, 2015) and carrying on its impact to present days. During colonial times, the Catholic church played an important role in political dynamics. Because “the
colonial administration was weak, the church became an important partner in governing” (Ramirez, 2005, p. 2067), meaning that the institution was not only committed to religious tasks but also was often responsible for educating the youth in private religious schools, leading hospitals, keeping statistics on marriages and deaths, lending money, etc. The church was, therefore, a wealthy institution.
Christianity has also had a significant influence when it came to maintaining the status quo. For that, Benavides (2015) heavily critiques how religious leadership would make use of the Bible to justify slavery, for example. In the 17th century, Antônio Vieira, a Portuguese Jesuit praised by Brazilian Catholics up to present days, would often mention the biblical book of Exodus to defend that “‘God does not complain about the Pharaoh, but rather about the overseers,’ meaning by this that whereas the abuses of the slave masters should be punished, the institution of slavery itself was divinely ordained” (p. 536). Nowadays, Catholicism is still the main religion in Latin America, with 425 million adepts in 2014 – constituting almost 40% of the world's Catholic population, as shown by a Pew Research survey (2014). However, not only in Brazil but in the entire region, Catholicism is losing ground to “Evangélicos” - as protestants in the area are often called. The group involves Christians from several Pentecostal and Neo-Pentecostal churches.
According to the National Institute of Geography and Statistics (IBGE), the number of Evangelicals in Brazil has presented na increase of 61% in 10 years. Similarly to Catholics, Evangelicals have a conservative attitude when it comes to moral questions and social issues such as “abortion, homosexuality, same-sex marriage, artificial means of birth control, sex outside of marriage, and drinking alcohol” (Zilla, 2018). On the other hand, as stated by scholar Andrew Chestnut, Catholics are a more heterogeneous group, with segments linked to the left and others to the right. Contrarily, Evangelicals are more politically homogeneous and conservative (Passarinho, 2019).
As a political reflex of this phenomenon, The Evangelical Caucus (In Portuguese, Bancada Evangélica) has gained power in the past elections. As stated by the Brazilian Intersindical Department of Parliamentary Advisory (Diap), they are constituted by representatives who either declare themselves Evangelicals, Protestant or Pentecostal or join the group in voting on topics related to religion and customs, as well as those who also occupy positions in the structures of religious institutions, such as bishops, pastors, missionaries and priests or members of Christian musical groups.
Political marketing strategy and social media
Another key discussion for this paper revolves around the concept of political marketing. The cross-disciplinary perspective comprises studies on politics, marketing, and communications (Scammell, 1999) and is a central element of political campaigning strategies (Merlă, 2018). Different definitions have been proposed throughout the years, but what scholars seem to agree on is that it is the process used by political candidates to promote themselves to their electorate by establishing and renewing long-term political relationships through media (Henneberg, 2002; Jogdankar, 2018; Lock & Harris, 1996). Some go even further and say that marketing strategy is the heart of electoral success (Kotler and Kotler, 1999, p. 4). Regardless, through new instruments and platforms, especially online tools, politicians can more efficiently connect with their target groups and market themselves (Merlă, 2018). Still, it is erroneous to say that all authors agree on these explanations. O’Shaughnessy (2001) strongly criticizes scholars who direct their perception only to areas common to political and consumer marketing, excluding “significant features in the complex environment of political communication” (p. 1056). For him, when discussing political marketing, researchers solely apply the business context to the political process, which is
mistaken due to the fact that the topics have different social values. This present study does that by looking at how religion is used in the political discourse.
According to both Barber (2005), as well as Smith and Hirst (2001), politicians need to focus on creating a long-term advantage in order to carry out a plan to achieve their objectives. As a political marketing strategy, social media have become a key tool for conquering and engaging voters. According to Baldwin-Philippi (2015), what was once used to strictly share press releases and official events, nowadays serves as a platform to “highlight the pleasant interpersonal elements of campaigns” (p. 2). Social media, however, do not work alone. They serve as an extension of existing campaign strategies (Schwartz & Farkas, 2018). Still, as stated by Strandberg (2013), it differs from the more traditional paths because a) it is more affordable; b) the infrastructure is readily available; c) it can be used to recruit volunteers, organize the campaign, mobilize and raise funds; and d) social networks allow candidates to send tailored messages to specific groups. Here is also important to make a more clear distinction between medium and message. McLuhan (1964) defends that message is the “change of scale or pace or pattern” that a new invention “introduces into human affairs” (p. 8), meaning that the message is not specifically the content, but also the change in social dynamics (Federman, 2014). Medium, on the other hand, is any extension of ourselves", commonly being “any new technology” (McLuhan, 1964, p. 7). That relates to this research in the sense that a medium, in this case, Facebook, is used to impact the political dynamics during the Brazilian elections. That is why this paper also looks at the engagement of followers through likes, aiming at quantifying the social impact of religious rhetoric.
More so, in 2018, social media was used by politicians to solidify their communication with citizens. This relation is remarkably important because it provides a direct linkage to people without the interference of philosophical disputes and institutional niceties’ (Kriesi, 2014, p. 34). As nicely mentioned by Engesser et al. (2017), that does not mean that
traditional media is out of the picture. It is still useful for distributing “official statements to larger audiences” (p. 1113). But through social media, politicians can personalize their message in order to speak on a more personal level to their public. This strategy is exactly what this paper expects to exemplify fill by showing how religious rhetoric is used to endorse candidate preferences and leadership through Christian references made online.
Religion and Politics
When it comes to religion and politics, a vast majority of scholars have focused on the United States (Campbell, Green, & Layman (2011); Colombo & Jarratt, 1984; Denton-Borhaug, 2012; Hulsether, 2002; Sherkat et al., 2011; Wald, 1986). Other regions of the world are also specifically discussed, mostly having in common either the predominance of certain religious groups and/or the active participation of these groups in political decision making. Some examples are Poland (Grzymala-Busse, 2018; Szajkowski, 1983; Szlek Miller, 1997), China (Laliberté, 2017; Woo, 2019), countries of Muslim majority (Ayoub, 2003; Kooij, 1980; Lippman, 1982) and Christian democracies in Western Europe (Bale & Krouwel, 2013; Lamberts, 1997), where the religious-secular divide forms a salient political cleavage. Latin American countries, on the other hand, have not been studied as thoroughly, and when they do, scholars speak of the region as a whole and tend to ignore the particularities of the different countries. Additionally, it is also possible to spot a clear religious bias in some of these papers. The work of Escobar (1982), for example, who studies the Evangelical missions in Latin America, though being relevant, holds a clear religious discourse. For instance, the scholar reports to evangelicals as ‘we ’and ‘us’, going on to state that “This is not a time to accommodate the gospel but to ask for a new empowering from the Spirit in order to resist the spirit of this age” (p. 111).
Voters’ religion.
Before argumenting on religious rhetoric in politics, the relationship between political leaning and religious values must also be highlighted. In the section above, several religions were mentioned to exemplify the relation between religion and politics. However, the focus here is on Christianity due to its dominance in Brazil and the relevance to the research question of this study. According to Layman and Hussey (2007), religion starts in everyday practices. Participating in religious activities, such as attending services, “exposes individuals to cues from clergy on moral and political issues” (p.183). McAdams and Lance (2013) take it one step further by defending that the more religiously committed a group is, the more politically conservative they are, which is explained firstly by the already mentioned values towards sexuality and reproductivity, as well as that, according to the mentioned scholars, people that maintain the same religious tradition often also have similar political attitudes because “they are exposed to homogenous political messages that are consistent with the religious tradition’s theological doctrine; the more exposure to those messages, the more consistent individuals ’political attitudes would be with the doctrinal core of their religious tradition” (p. 488).
When it comes to technology and, more specifically, social media, scholars (Faimau & Behrens, 2016; Helland, 2000) believe that the relationship can be seen through two different perspectives: religion online and online religion. For religion online, the internet is used to simply communicate ideas as a “tool of the top-down communication structure” (Faimau & Behrens, 2016, p. 70). Online religion, on the other hand, represents a space for open interaction, without a hierarchical structure. This is directly related to the concept of technologization of discourse that, according to Fairclough (2013, p. 137), relates to “the process of intervention in the sphere of discourse practice with the objective of constructing a new hegemony (…)”. Seeing that from a social media perspective, it is when certain
linguistic strategies are implemented to influence people, or more specifically, followers. This distinction then proves to be relevant to this paper and the investigation of the extent to which religious rhetoric is used to persuade voters through social media.
Religious rhetoric in politics.
It is important to also discuss what has been written on religious rhetoric within politics so far. As said before, when it comes to religion and politics, certain regions and countries of the world have clearly been more on the spotlight over others. Not surprisingly, when the debate turns to religious rhetoric, that is also the case. More than that, all the key literature centers solely on the North American scenario. Domke and Coe (2010) conceptualize religious rhetoric within the political sphere by calling it a technique used by politicians to “invoke their sincerely held religious beliefs in a fashion that allows them to define key issues, differentiate themselves from their opponents and, ultimately, advance their political aims” (Coe & Chapp, 2017, p. 110).
To operationalize this concept, a framework originally proposed by Coe and Chenoweth (2013) will be used to identify words and phrases within the Facebook posts that are “either distinctly Christian in all contexts (e.g., Christ, the New Testament, Catholic) or that are likely to be imbued with Christian meaning when used within a speech that also explicitly mentions Christianity (e.g., God, faith, pray)” (p. 375). To detail these parameters, the authors divide them into six main coding groups for terms that might appear throughout the data: “God (e.g., ‘the Creator’), the Bible (e.g., ‘according to the Gospel’), physical manifestations of religion (e.g., ‘the parish’), meta-physical manifestations of religion (e.g., ‘heaven’), nonleader religious figures (e.g., ‘Christians’), and religious figures in leadership positions (e.g., ‘pastor’)” (Coe & Chapp, 2017, p. 116).
Here, carrying on with the logic behind the previous section, Brazilian citizens that have more affinity with religion would be more engaged in politics and politicians that speak in rhetoric that dialogues with their conservative beliefs. This can be used to explain the hypothesis that defends that:
Hypothesis 1: Facebook posts containing religious rhetoric receive more reactions from the public through likes and comments than the ones without any references.
By analyzing the American elections of 2008, Coe and Chapp (2017) found that Christians are a “reliable part of the Republican coalition” (p. 118). That does not mean that religious rhetoric is inexistent in discourses by liberal candidates. Barack Obama, from the Democrat Party, was found to approach public Christianity less often than its competitor. Still, he would do so by focusing on “its more abstract forms” (Coe and Chenoweth, 2013, p. 389). Making a comparison to the Brazilian scenario, the more left-wing candidate, Haddad, is expected to have mentioned Christianity less often. In the same manner, taking into consideration the discussion about religion and how right-wing politicians in Brazil tend to have more conservative values, it is expected that Bolsonaro will have referred to Christianity more often. Based on that, it is possible to assume that:
Hypothesis 2: The use of religious rhetoric in Jair Bolsonaro’s posts will be higher than the other three candidates.
Hypothesis 3: The use of religious rhetoric in Fernando Haddad’s posts will be lower than the other three candidates.
Methods
The research draws on a quantitative content analysis of the Facebook posts of Jair Bolsonaro, Fernando Haddad, Ciro Gomes, and Geraldo Alckmin during their 2018 presidential campaign. The choice of using quantitative research methods is due to the fact that a) it allows the researcher to work with a large amount of data; b) it provides numerical measurements for the observation; and c) makes it possible to formulate hypotheses among variables (David & Sutton, 2011). Furthermore, content analysis was chosen as the most suitable approach because it allows the study to systematically and objectively analyze the data collected to identify specific characteristics within the messages (Haggarty, 1996; Holsti, 1968).
As previously mentioned, the candidates maintained a strong presence on social media, especially on Twitter and Facebook. For this analysis, however, the latter was chosen based on a) its popularity, considering that is the platform with the highest amount of active subscribers in Brazil (Imme, 2020); and b) it is the main channel of communication of president Bolsonaro, not forgetting that it was on Facebook that he shared his first speech as an elected president - live-streaming the event even before heading to the official press conference (Hidalgo, 2018).
Time frame
This content analysis utilizes Facebook posts from the four main Brazilian candidates for presidency from the first official date that political parties were allowed to appoint their representatives for the elections (according to the official electoral calendar) - July 20, 2018 - until the day Bolsonaro was elected president - on October 28th, 2018, totaling 101 consecutive days. This particular time frame of two months leading up to the presidential elections was selected because it is the period in which politicians are the most exposed and active on media due to the election. It is also able to provide a sufficient amount of data for
this analysis. In 2018, the presidential elections happened in two rounds, the first one taking place on October 7th, and the second one three weeks later, on October 28th. For this paper it was decided to use both periods, considering that we aim at quantifying the relative number of religious rhetoric in comparison to a total of posts, regardless of the lack participation of certain candidates in the second election round.
In 2018, 13 candidates entered the electoral run to become the president of Brazil. The number was the highest in comparison to the elections that succeeded the redemocratization of the country in 1989. For this paper, however, only the four most successful politicians were selected. By that, it means that these are the ones with the largest amount of votes in the first round: Bolsonaro (46.03%), Haddad (29.28%), Gomes (12.47%) and Alckmin (4.76%). The remaining candidates were inexpressive, having below 2,5% of valid votes.
Data collection and sample
Furthermore, Facebook posts from the presidential candidates were chosen as the coding unit. Therefore, in order to first collect the relevant data, a preliminary step is to extract all the necessary posts from the four different Facebook pages. The texts were collected using Facepager, an open-source data collection and extraction application (Pancer & Poole, 2016) that facilitates the data exporting process by retrieving information from different sections of the social networking platform (e.g., content from posts and the amount of reactions, shares, and comments). With such a tool, it was possible to extract the texts from the posts with their respective dates and the number of likes for each candidate within the previously chosen time frame. All this information was gathered and organized into four distinct CSV files and later compiled into one Excel sheet. The latter was organized into the following categories: candidate code, post, date and time, and likes. At first, 1621 posts were
found in total for the four candidates within the given time frame. Some of them, however, were disregarded for this study. It was established that posts containing the following characteristics would be removed from the list of posts: those that a) do not contain any text (only photos or videos), b) have been shared repeatedly (none of them having religious references), c) contain only hashtags (none of them having religious references), or d) have just emojis. That is how the final sample size was obtained (N = 1260). Table 1 shows the final distribution of valid posts for this study.
Table 1. Candidates and the amount of analyzed Facebook posts.
Candidate Amount of posts Total
Jair Bolsonaro 189 N = 1260 Fernando Haddad 437 Ciro Gomes 341 Geraldo Alckmin 293 Operationalization
With the required data in hands, the step that followed was to apply the previously mentioned typology of Coe and Chenoweth (2013) to the retrieved texts by having one single coder manually coding the data over the course of one and a half weeks. The choice of manually coding the data instead of using automated applications is due to the fact that the texts are considerably short, making it easier to do so. Also, several words that were used by the candidates in their posts have more than one meaning or interpretation. An example of this is the word são, Portuguese for ‘saint’ or ‘wholy’, that is also present in the name of capital cities such as São Paulo and São Luis. The same happens to its female equivalent, santa, that is also present in city names like Santa Catarina. Words like these make it difficult to identify if the Facebook posts contain a specific type of religious rhetoric or it is solely
mentioning places. Once again manual coding proves to be more efficient, considering that context can also be taken into account.
The dictionary was based on the original one used by Coe and Chenoweth that was provided by professor Kevin Coe himself after e-mail contact. There, six different types of religious rhetoric were coded: (1) God, (2) the Bible, (3) physical manifestations of religion, (4) metaphysical manifestations of religion, (5) religious figures in leadership positions, and (6) nonleader religious figures. The original dictionary, however, contained several words that are specific to the US scenario and do not apply to the Brazilian context. That is why after a careful read of the entire document, some words were excluded (e.g.: Amish, Quaker, Mennonite), and some common expressions related to religion in Brazil were added (e.g.: Assembleia de Deus [God’s Assembly], Neo-Pentecostal, Universal Church [Igreja Universal]). For the final dictionary. They were all organized in a codebook (see Appendix A). It is important to say that, considering that the analyzed texts are in Brazilian Portuguese, all terms present in the codebook are in English and Portuguese.
Reliability
To measure the level of agreement, an inter-coder reliability test was conducted by a second coder who was given a subsample of 64 posts, with 16 texts for each candidate, which represents 5% of the total sample size (N = 1260). It is relevant to highlight that the second coder was chosen considering their proficiency in both Brazilian Portuguese and English. All the texts that were previously considered to have religious rhetoric were added to this selection. The additional posts were randomly selected. This specific method was chosen due to the small amount of religious rhetoric found in the posts, which will be further explained in the next chapter. To calculate Krippendorff’s Alpha for the religious rhetoric, the six different variables were first merged in one round, using the Kalpha macro (De Swert, 2012; Hayes & Krippendorff, 2007) in SPSS. The test shows that, overall, the dimensions
are reliable, obtaining α = 0.83. Furthermore, following the guidelines of De Swert (2012), the bootstrapping procedure indicates that there is only 2.5 percent chance that the Kalpha would be below 0.80 if the whole population would be tested. As a second step, each variable was coded individually, obtaining different values: God (α = 1.0), the Bible (α = 1.0), Physical Manifestations (α = 0.65), Meta-physical Manifestations (α = 0.79), Leaders (α = 0.9) and Nonleaders (α = 0.33) (See Appendix B). The discrepancy between the reliability of the overall results and the specific ones shows that, although terms were coded for different categories on certain occasions, the religious rhetoric was still detected. The small amount of rhetoric found also plays a part here. Still, the results from the Nonleaders category are treated with prudence.
Results
This analysis begins by presenting the overall results from the coding process. Then, it moves onto the discussion of the relationship between Facebook likes and religious rhetoric, and finally, it further interprets the data retrieved from far-right candidate Jair Bolsonaro and Worker's Party candidate Fernando Haddad. After manually coding the 1260 posts, the results are as shown in Table 2. The numbers represent the amount of times that terms within the six different parameters suggested by Coe and Chenoweth (2013) were used by each of the four different politicians.
Table 2. Candidates for presidency and the number of words mentioned within each religious rhetoric parameter of the typology proposed by Coe and Chenoweth’s (2013)
Candidate God Bible The
Physical Manifesta-tions Meta-physical Manifesta-tions Leaders Non-lead ers Total Jair Bolsonaro 5 1 2 1 3 0 12 Fernando Haddad 0 0 1 0 1 5 7 Ciro Gomes 0 0 0 0 0 0 0 Geraldo Alckmin 0 0 0 1 1 0 2
The first observation from the outcome of the coding process is that, in fact, the use of religious rhetoric in the posts was extremely limited. Of the 1260 total of posts, 16 posts presented some sort of Christian discourse, some of them having more than one reference. Still, several differences were identified amongst the candidates. As can be noticed in table 2, Jair Bolsonaro was the only candidate who mentioned terms within the first two-categories: ‘God’ (N = 5) and ‘the Bible’ (N = 1). On the other hand, Fernando Haddad was the only politician that used term within the ‘Nonleaders’ category (N = 5). Moreover, dividing the results presented in Table 2 and the amount of posts per candidate shown in Table 1, the percentages in Table 3 were obtained for the amount of religious rhetoric for each politician.
Table 3. The percentage of religious rhetoric present in texts posted by the candidates on their Facebook pages
Candidate Percentage of religious rhetoric
Jair Bolsonaro 6.34%
Fernando Haddad 1.60%
Ciro Gomes 0
Geraldo Alckmin 0.68%
Once again, from the percentages found above it is possible to affirm that the general use of religious rhetoric was considerably small for all the candidates. That can be proven by the number presented by Geraldo Alckmin, who only had religious rhetoric in 0.68% of his posts, and Ciro Gomes, who has not made any references. These results will be further discussed in the following sections, in accordance with the previously presented hypotheses.
Facebook likes and religious rhetoric
The first hypothesis of this study stated that Facebook posts containing religious rhetoric receive more likes from the public than posts that do not contain any references. In order to statistically prove this assumption, a series of analyses of variance (ANOVAs) were run in SPSS.
In the first round, a one-way ANOVA was conducted to compare the individual level posts to the total amount of likes of (1) the texts that contain religious rhetoric and (2) the texts that do not contain any religious rhetoric. Taken together, the results show that there is a significant difference between the two groups at the p < 0.05 level [F (1, 1258) = 7.23, p = 0.007]. According to this analysis, the posts that contain religious rhetoric (M = 36300.7, SD = 29292.2) received more likes than the posts that do not contain any religious rhetoric (M = 14032.1, SD = 32963.6), therefore supporting the hypothesis.
A second ANOVA round was performed, this time splitting the data into the four different candidates. The results for each of the candidates are presented In Table 4.
Table 4. Mean scores for the comparison of the number of likes for posts with and without religious rhetoric per candidate
Candidate Presence of religious rhetoric Variable: Likes M(SD) P F Jair Bolsonaro Yes 48913.9 (26695.2) 0.64 0.22 No (69109.2) 58749.5 Fernando Haddad Yes 12980.7 (848) 0.74 0.11 No 10749 (11600)
Ciro Gomes Yes - - -
No - Geraldo Alckmin Yes 1908.5 (970.8) 0.86 0.03 No 2396.7 (3817.6)
Here the results are different from the ones obtained in the first ANOVA rounds. When split apart, the difference between conditions was found insignificant (p < 0.05) for all four candidates: Bolsonaro (p = 0.64), Haddad (p = 0.74), Gomes (not presented) and Alckmin (p = 0.86). The notable distinction of results between the two rounds of ANOVAs can be explained by the fact that, in this case, the candidates are a confounding factor. Each of them has a different number of followers, average likes on their posts, and algorithm. That means that aggregating all their values, as done in the first round, might result in an unbalanced analysis of the given data. Furthermore, there is also a significant difference between the number of posts that contain religious rhetoric (N = 16) in comparison to the posts that do not contain any religious rhetoric (N = 1244), which also influences the final
results. It has to be noted, however, that by splitting up the sample more, I compare an even smaller group of posts containing religious rhetoric to a large group not containing it.
Right-wing candidate and religious rhetoric
The second hypothesis stated that candidate 1, right-wing politician Jair Bolsonaro, was the one who made use of religious rhetoric the most among the four candidates. Firstly, this hypothesis can be answered by looking at the data retrieved from the content analysis. Table 2 and Table 3 support the assumption that Bolsonaro was the candidate who used more religious rhetoric. The then-member of the Social Liberal Party (PSL) used Christian discourse 12 times in total. That constitutes 6.34% of his posts, a percentage higher than the ones presented by Haddad, (1.6%), Gomes (0), and Alckmin (0,68%).
To statistically test this hypothesis, another one-way ANOVA was carried out. This time, the means for the presence of religious rhetoric for each candidate was compared, proving that Bolsonaro had indeed made use of this specific type of discourse more frequently, as presented in Table 5.
Table 5. Mean scores for each candidate per inference of religious rhetoric
Candidate Variable: presence of religious rhetoric M(SD)
Jair Bolsonaro 0.07 (0.33)
Fernando Haddad 0.02 (0.26)
Ciro Gomes 0 (0)
Geraldo Alckmin 0.01 (0.08)
The calculations show that there is a significant difference between the four candidates at the p < 0.05 level [F (3, 1256) = 5.86, p = 0.001]. The mean score for Bolsonaro (M = 0.07, SD = 0.33), is greater than the numbers found for all the other candidates, as
plotted in Figure 1 below. Moreover, the post hoc comparisons using the Tukey HSD test indicated that the mean score for Bolsonaro was significantly different when compared to the other three candidates, Haddad (p = 0.011), Gomes (p = 0) and Alckmin (p = 0.003).
Figure 1. Candidates and their mean values for the presence of religious rhetoric in their Facebook posts
As previously displayed in Table 2, Bolsonaro used terms within five out of the six rhetoric categories. Moreover, he was the only one who used terms within the first two categories ‘God’ and ‘the Bible’. The politician referred five times in five different posts to the ‘Supreme Being’ (Coe & Chenoweth, 2013), directly mentioning and intoning the word ‘God’ in phrases such as “Fiquem com Deus” [Stay with God], “Que Deus continue abençoando…” [I hope God keeps blessing]. An example of that is a post shared on July 27th, where he states: “Quem esquece seu passado está condenado a não ter futuro. - A família é a base de uma sociedade sadia. - Deus nos ilumine em todas as decisões que
0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8
tomarmos, em especial, naquela do dia 07 de outubro” [Whoever forgets their past is doomed to have no future. - The family is the basis of a healthy society. - God enlighten us in all the decisions we make, especially that of October 7th]. Furthermore, by directly quoting the Bible, Bolsonaro shared a biblical verse, in John 8:32, on August 2nd: “…E conhecereis a verdade, e a verdade vos libertará.” [Then you will know the truth, and the truth will set you free].
Left-wing candidate and religious rhetoric
The third hypothesis states that candidate 2, left-wing politician Fernando Haddad, was the one who made use of the religious rhetoric the least among the four candidates. Just like the results for Jair Bolsonaro, this hypothesis can also be answered by going back to the data arisen from the results of the content analysis. That being said, Table 2 and Table 3 dismiss the assumption that Haddad was the candidate who used less religious rhetoric. Only behind Bolsonaro, the Worker’s Party (PT) candidate used Christian discourse 7 times in total. That constitutes 1.6% of his posts, a percentage greater than the ones presented by Ciro Gomes (0) and Geraldo Alckmin (0,68%). That is again confirmed by the ANOVA performed for the previous hypothesis and the results displayed in Table 5, showing that the mean score for Haddad (M = 0.02, SD = 0.26) is lower than the mean score of Bolsonaro (M = 0.07, SD = 0.33), but is still higher than the numbers found for the other two candidates: Gomes (M = 0, SD = 0) and Alckmin (M = 0.01, SD = 0.08). In addition, the post hoc comparisons using the Tukey HSD test indicated that the mean score for Haddad was significantly different when compared to Bolsonaro (p = 0.01). However, Haddad did not significantly differ from Gomes (p = 0.604) and Alckmin (p = 0.88). These findings, therefore, reject the hypothesis that Haddad would be the one with the least amount of religious rhetoric amongst the four candidates.
As displayed in Table 2, Haddad used terms within three out of the six rhetoric categories. Moreover, he was the only one who used terms within the last category, ‘Non-leader religious figures’. Differently from Bolsonaro, however, who presented religious rhetoric throughout different posts, Haddad marked a significant score here because in one single post he mentioned several times the names of several different protestant churches, terms under the non-leaders category, from which the representations he met with on October 10th: “Encontro de Fernando Haddad com lideranças evangélicas da Metodista, Presbiteriana, Batista, Anglicana, Assembleia de Deus, Comissão Justiça e Paz, Betesda e a Frente Evangélica pelo Estado de Direito. #HaddadPresidente #Haddad13” [Fernando Haddad's meeting with evangelical leaders from Methodist, Presbyterian, Baptist, Anglican, God’s Assembly, Justice and Peace Commission, Bethesda and the Evangelical Front for the Rule of Law #HaddadPresidente # Haddad13].
Discussion and Conclusion
The present study aimed at identifying and understanding the use of religious rhetoric during the 2018 presidential elections in Brazil by Jair Bolsonaro, Fernando Haddad, Ciro Gomes, and Geraldo Alckmin. The paper indeed found differences in the use of this specific type of rhetoric throughout the Facebook posts of the four candidates during their campaign period, which corroborates the findings of previous scholars (Chapp, 2012; Coe & Chapp, 2017; Coe & Chenoweth, 2013).
That can be first seen through different manners in which this discourse was used by the candidates. Bolsonaro was the only candidate who used terms within the first two categories, ‘God’ and ‘the Bible’, the most. Contrarily, Haddad was the only one who used terms within the last category, ‘Non-leader religious figures’. That shows to be consistent with the conclusions obtained by Coe and Chapp (2017) that suggest that the six categories
are usually used in a different fashion by candidates throughout their campaigns depending on how they want to address the particular exigencies they face (Campbell & Jamieson, 2008; Domke, & Coe, 2008). That was the case of Haddad, who met up with people from different religious groups, so that he would be able to dialogue with their voters, and published a Facebook post on it, showing that he still felt the need to share it. On the other hand, the study disagrees with the authors when it comes to which of the six categories are used the most. Coe and Chapp defend that politicians tend to make more references to “God, faith, prayer, heaven, and the like, with fewer references to the more tangible and specific aspects of Christian faith (such as the Bible or particular religious denominations).” (p. 388). Apart from Bolsonaro, who specifically mentioned God several times and used a Bible verse, Brazilian candidates did not mention any abstract expressions. Instead, the most common terms were related to leaders and nonleaders who have a direct tie to Christianity.
In total, Bolsonaro was the candidate that used religious rhetoric the most, 12 times, supporting the second hypothesis. This is in line with what has been discussed by a group of scholars who defend that, in Brazil, right-wing politicians tend to have a stronger connection to religious and traditional values (Alencar, 2018; Cepêda, 2018; Chaia & Brugnano, 2015). The main slogan of the president, “Brazil above everything, God above everyone” is a prime example of religious rhetoric. Haddad, on the other hand, differently from what the third hypothesis proposed, was not the one who used religious rhetoric the least. Instead, he was second among the four politicians, making 7 Christian references in total. From a political marketing perspective these results make sense because, since religion is so important in Brazilian politics, a candidate that makes it to the second round should mention it at some point. Again, the results found for this hypothesis dialogues with the findings of Coe and Chapp (2017) who concluded that there was no such thing as a “God gap” among parties in the 2012 American elections, meaning that the political leaning did not actually matter when
it came to the use of Christian discourse, instead their personal impulses would “shine through as the moment moves them” (p. 122). That could be applicable here as well. Although it was not the case for right-wing politician Bolsonaro, differences in the use of religious rhetoric were shown to be insignificant among the other three candidates. Whether this finding is due to differences in political leaning or personal references is up for further research.
When it comes to the relationship between the number of likes and the presence/lack of religious rhetoric in the Facebook posts, this study shows that, in reality, there is not a significant difference between the two groups when looking into the numbers for each of the candidates separately. The incapability to support the hypotheses brings light to some discussions related to the gap in more specific literature. Scholarship shows that indeed Facebook is a digital device with the capacity to mediate religious behaviors, interactions, and discourses (Faimau & Behrens 2016). Still, the effect on the engagement of followers is uncertain. At first impression, the number of likes and the presence of rhetoric seemed to be correlated. Nonetheless, when looking at the data more closely and taking into account the particularities of each candidate, this relation disappeared. A stronger causal test, perhaps with an experiment, could shed more light on this discussion.
Although several differences were observed amongst the frequency and manner in which Christian references were made throughout the analyzed Facebook posts, another main finding is that, in fact, little religious rhetoric was found. Center-left candidate Ciro Gomes, for example, has not made any Christian references. That relates to the results presented by Coe and Chenoweth (2013) when they state that some voter groups receive more attention than others, depending upon the nature of the elections. In the 2012 US presidential elections, evangelicals also received less candidate attention than it was expected. “For Romney, they were a safe enough bloc that he felt comfortable giving them
just his usual dose of mainstream religious rhetoric; for Obama, they were likely out of reach” (p. 123). That is a possible explanation for what happened here with Bolsonaro and Haddad.
Moving into further limitations, a factor that might have impacted the results found is that, for this paper, it was chosen to code only Facebook posts, although candidates were also active on other social media platforms such as Twitter (Goya, Fiacadori, & Santos, 2019) and Instagram. Also, visual aspects such as photos and videos (Parmelee & Roman, 2019), as well as speeches and television appearances (Eduardo, 2014; Ekström, 2001) could have also been taken into account but were opted out due to the necessary concision. Such topics, however, can be later explored in depth. Furthermore, the timeframe was another limitation, considering that Bolsonaro and Haddad campaigned for three more weeks than Gomes and Alckmin due to the runoff voting. Lastly, although one of the theoretical focus of this paper is on the Evangelicals in Brazil, the dictionary used comprises words related to Christianity in general. That is because if only Evangelical terms were to be coded, even fewer results would have been found. Perhaps future analyses can look at the extent to which political leaders allude to different religious constituencies.
Regardless, the study provides relevant results that can serve as a stepping stone for further investigation in the field such as the use of religious rhetoric in the period that succeeded the 2018 elections. After having been sworn in as president, Bolsonaro has promoted Evangelical services at his presidential headquarter (Vargas & Behnke, 2019), and appointed several religious personalities to assume key government positions - the most famous being Damares Alves, a pro-life and anti-homosexual rights evangelical pastor (Gomes, 2020) who took over the Ministry of Women, Family and Human Rights, former Ministry of Human Rights. Here, it would be interesting to investigate if such decisions played a part in the way the elected president uses religious rhetoric on his social media
pages during the mandate.
Further research can also focus on the effects of this specific type of discourse on voters or if applied to social media, followers. Weber and Thornton (2012), for example, argue that religious tendencies instigate a sense of traditionalism that influences how political candidates are considered by citizens. Albertson (2015) takes one step further and analyzes how religious appeals have distinct meanings to different audiences depending on their affinity with the theme. Such ideas could be explored in order to explain the extent to which religion constructs a sense of political identity amongst the Brazilian electorate (Mccarthy, Olson, & Garant, 2019). These studies, however, as most of the key papers used for this paper, are based on American politics, proving the importance of this and possible sub-sequential research.
To sum up, this study proves the generalizability of the typology developed by Coe and Chap. That is because it succeeds to apply the original research to a format that goes beyond verbal speeches and also focuses on a country other than the United States. On that note, the research adds up to what has been written on Christianity and politics and validates its relevance to the field by studying the changes in the Brazilian political scenario. All in all, it was found that, although the use of religious rhetoric by Brazilian presidential candidates was limited, they used it in different manners and frequencies while trying to appeal to voters on social media. Such a conclusion leaves room for further exploration of scholarship on political communication and religion.
References
Albertson, B. L. (2015). Dog-whistle politics: Multivocal communication and religious appeals. Political Behavior, 37(1), 3–26. doi:10.1007/s11109-013-9265-x
Alencar, G. (2018). Evangélicos e a Nova Direita no Brasil: os Discursos Conservadores do “Neocalvinismo” e as Interlocuções com a Política [Evangelicals and the New Right in Brazil: Conservative Discourses of “Neocalvinism” and Interlocutions with Politics]. Programa de Pós-Graduação em Ciências Sociais - UFJF, 13(2), 101–117.
AS/COA (2018). A Guide to 2018 Latin American Elections. Retrieved from https://www.as-coa.org/content/guide-2018-latin-american-elections
Ayoub, M. (2003). The crisis of Muslim history: religion and politics in early Islam. Oxford: Oneworld.
Baldwin-Philippi, J. (2015). Using technology, building democracy: digital campaigning and the construction of citizenship. Oxford: Oxford University Press.
Barros, L. & Silva, M.S. (2020). Right-wing populism in the tropics: The rise of Jair Bolsonaro. VoxEU. Retrieved from https://voxeu.org/article/right-wing-populism-tropics-rise-jair-bolsonaro
Benavides, G. (2015). Christianity in Latin America. In J. Wright (Ed.), International Encyclopedia of the Social & Behavioral Sciences (pp. 534–539). Amsterdam: Elsevier.
Bolognesi, B., Babireski, F. B., & Maciel, A. P. (2019). Conhecendo o vazio: congruência ideológica e partidos políticos no Brasil [Knowing the void: ideological congruence and political parties in Brazil]. Política & Sociedade, 18(42), 86–116.
Campbell, D., Green, J., & Layman, G. (2011). The Party Faithful: Partisan Images, Candidate Religion, and the Electoral Impact of Party Identification. American Journal of Political Science, 55(1), 42–58.
Campbell, K. K., & Jamieson, K. H. (2008). Presidents creating the presidency: Deeds done in words. Chicago, IL: University of Chicago Press.
Carreirão, Y. S. (2014). O sistema partidário brasileiro: um debate com a literatura recente [The Brazilian party system: a debate with recent literature]. Revista Brasileira de Ciência Política, 14(1), 255–295.
Cepêda, V. A. (2018). A nova direita no Brasil: contexto e matrizes conceituais [The new right-wing in Brazil: context and conceptual matrices]. Dossiê - Interpretações do Brasil Contemporâneo, 23(2), 40–73.
Chaia, V. & Brugnano, F. (2015). A nova polarização política nas eleições de 2014: radicalização ideológica da direita no mundo contemporâneo do Facebook [The new political polarization in the 2014 elections: ideological radicalization of the right in the contemporary world of Facebook]. Revista de arte, mídia e política, 7(21), 99– 129.
Chapp, C. B. (2012). Religious rhetoric and American politics: The endurance of civil religion in electoral campaigns. Ithaca, NY: Cornell University Press.
Codato, A. (2006). A political history of the Brazilian transition from military dictatorship to democracy. Revista de Sociologia e Política, 2(1).
Coe, K., & Chapp, C. (2017). Religious rhetoric meets the target audience: Narrowcasting faith in presidential elections. Communication Monographs, 84(1), 110–127.
Coe, K., & Chenoweth, S. (2013). Presidents as priests: Toward a typology of Christian discourse in the American presidency. Communication Theory, 23(1), 375–394. doi:10.1111/comt.12020
Colombo, F., & Jarratt, K. (1984). God in America: religion and politics in the United States. New York, NY: Columbia University Press.
David, M., & C. D. Sutton (2011). Social Research: Social Research: An Introduction. London: SAGE.
Denton-Borhaug, K. (2012). Beyond Iraq and Afghanistan: Religion and Politics in United States War-Culture. Dialog, 51(2), 125–134. doi:10.1111/j.1540-6385.2012.00669.x
De Swert, K. (2012). Calculating inter-coder reliability media content analysis using Krippendorff’s Alpha. Retrieved from
https://www.polcomm.org/wp-content/uploads/ICR01022012.pdf
de Vreese, C. H., Esser, F., Aalberg, T., Reinemann, C., & Stanyer, J. (2018). Populism as an Expression of Political Communication Content and Style: A New Perspective. The International Journal of Press/Politics, 23(4), 423–438.
doi:10.1177/1940161218790035
Domke, C. & Coe, K. (2008). The God Strategy: How Religion Became a Political Weapon in America. New York, NY: Oxford University Press.
Eduardo, L. (2014). As estratégias do discurso político: uma análise de imagens e procedimentos linguísticos [Strategies of political discourse: an analysis of images and linguistic procedures]. Palimpsesto - Revista do Programa de Pós-Graduação em Letras da UERJ, 13(19), 459-475.
Ekström, M. (2001). Politicians Interviewed on Television News. Discourse & Society, 12(5), 563–584. doi:10.1177/0957926501012005001
Engesser, S., Ernst, N., Esser, F., & Büchel, F. (2017). Populism and social media: how politicians spread a fragmented ideology. Information, Communication &
Society, 20(8), 1109–1126. doi:10.1080/1369118X.2016.120769
Escobar, S. (1982). Beyond Liberation Theology: Evangelical Missiology in Latin America. International Bulletin of Missionary Research, 6(3), 108–114.
Faimau, G. & Behrens, C. (2016). Facebooking religion and the technologization of the religious discourse: A case study of a Botswana-based Prophetic Church. Online - Heidelberg Journal of Religions on the Internet. 11(1), 66–92.
Fairclough, N. (1992). Discourse and social change. Cambridge: Polity Press.
Federman, M. (2004). What is the Meaning of the Medium is the Message?. Retrieved from
http://individual.utoronto.ca/markfederman/MeaningTheMediumistheMessage.pdf.
Fratini, J. (2020). Campanhas políticas nas redes sociais [Political campaigns on social media]. São Paulo: Matrix.
Gomes, K. (2020). "É o momento de a igreja ocupar a nação", diz Damares Alves [It is the moment for the church to occupy the nation”, says Damares Alves]. Deutsche Welle. Retrieved from
https://www.dw.com/pt-br/%C3%A9-o-momento-de-a-igreja-ocupar-a-na%C3%A7%C3%A3o-diz-damares-alves/a-52559550
Goya, D.H, Fiacadori, G., Santos, P. D. (2019). A polarização ideológica no Twitter: um estudo sobre as redes de retweets durante as eleições presidenciais de 2018
[Ideological polarization on Twitter: a study of retweets networks during the 2018 presidential elections]. Compolítica, 8(1).
Grzymala-Busse, A. (2018). Religion, Politics, and Values in Poland: Continuity and Change since 1989. In Sabrina P. R. & Irena B. (Eds.), Journal of Religion in Europe, 11(3), 278–279.
Haggarty, L. (1996). What is content analysis? Medical Teacher, 18(2), 99–101. doi:10.3109/01421599609034141
Hayes, A. F., & Krippendorff, K. (2007). Answering the call for a standard reliability measure for coding data. Communication Methods and Measures, 1, 77–89.
Helland, C. (2000). Online-Religion/Religion-online and Virtual Communities. In: J.K. Hadden, J.K. and Cowan, D.E. (Eds.), Religion on the Internet: Research Prospects and Promises, London: JAI Press/Elsevier Science.
Henneberg, S.C. (2002). Understanding Political Marketing. In: N. J. O'Shaughnessy and S. C. Henneberg (Eds.), The Idea of Political Marketing, (Vol 1, pp. 93–
170). Westport, CT: Praeger.
Hidalgo, Emilio S. (2018). O primeiro discurso de vitória por Facebook Live [The first victory speech by Facebook Live]. El País. Retrieved from
https://brasil.elpais.com/brasil/2018/10/28/politica/1540752778_816687.html
Holsti, O. R. (1968). Content analysis, In: Lindzey, G. & Aronson, E. (Eds.), The Handbook of Social Psychology (Vol. 2). Reading, MA: Addison-Wesley.
Hulsether, M. (2002). Religion, culture and politics in twentieth-century United States. Edinburgh: Edinburgh University Press.
Imme, A. (2020). Ranking das redes sociais: as mais usadas no Brasil e no mundo, insights e materiais gratuitos [Ranking of social networks: the most used in Brazil and in the world, free insights and materials]. Blog de Marketing Digital de Resultados. Retrieved from https://resultadosdigitais.com.br/blog/redes-sociais-mais-usadas-no-brasil/
Jogdankar, H. (2018). Marketing Management in Politics. Raleigh, NC: Lulu Publication.
Kaylor, B. T. (2010). Presidential campaign rhetoric in an age of confessional politics. Lanham, MD: Lexington.
Knoll, T. (2019). Seeking Liberation in Brazil: Though growing right-wing religious groups helped pave Jair Bolsonaro’s path to power, Brazil’s faith community has at other times driven progressive social and political change. Can Liberation Theology stage a comeback? NACLA Report on the Americas, 51(3), 227–231.
doi:10.1080/10714839.2019.1650483
Kriesi, H. (2014). The populist challenge. West European Politics, 37(2), 361–378.
Kooij, C. (1980). Islam in Qadhafi’s Libya : religion and politics in a developing country. Amsterdam: Universiteit van Amsterdam, Antropologisch-sociologisch Centrum.
Lamberts, E. (1997). Christian Democracy in the European Union, 1945/1995: Proceedings of the Leuven Colloquium, 15-18 November 1995. Leuven: Leuven University Press.
Laliberté, A. (2017). Religion, Resistance, and Contentious Politics in China. Review of Religion and Chinese Society, 4(2), 151–166. doi:10.1163/22143955-00402001
Layman, G. C., & Hussey, L. S. (2007). George W. Bush and the Evangelicals: Religious Commitment and Partisan Change among Evangelical Protestants, 1960–2004. In Campbell D. (Ed.), Matter of Faith: Religion in the 2004 Presidential Election. Washington, DC: Brookings Institution Press.
Lock, A. & Harris, P. (1996). Political marketing - vive la différence!. European Journal of Marketing, 30(10). 14–24.
Mccarthy, A., Olson, L., & Garand, J. (2019). Religious Right, Religious Left, Both, or Neither? Understanding Religio-Political Identification. Journal for the Scientific Study of Religion, 58(3), 547–569. doi:10.1111/jssr.12618
McLuhan, M. (1964). Understanding Media: The Extensions of Man. New York: McGraw Hill.
Moraes, I (2018). O que significa esquerda, direita e centro na política? [What does left, right and center mean in politics?]. O Estado de S. Paulo. Retrieved from
https://politica.estadao.com.br/noticias/eleicoes,o-que-significam-direita-esquerda-e-centro-na-politica,70002314116
Pancer, E., & Poole, M. (2016). The popularity and virality of political social media: hashtags, mentions, and links predict likes and retweets of 2016 U.S. presidential nominees ’tweets. Social Influence, 11(4), 259–270.
doi:10.1080/15534510.2016.1265582
Parmelee, J. H., & Roman, N. (2019). Insta-Politicos: Motivations for Following Political Leaders on Instagram. Social Media + Society, 5(2).
Passarinho, N. (2019). Por que igrejas evangélicas ganharam tanto peso na política da América Latina? Especilista aponta 5 fatores [Why have evangelical churches gained so much weight in Latin American politics? Specialist points 5 factors}. BBC.
Retrieved from https://www.bbc.com/portuguese/brasil-50462031
Pew Research Center (2014). Religion in Latin America - Widespread Change in Historically Catholic Region. Retrieved from
https://www.pewforum.org/2014/11/13/religion-in-latin-america/
Penteado, C. L. C. & Lerner, C. (2018). A direita na rede: mobilização online no impeachment de Dilma Rousseff. Em Debate, 10(1), 46–53.
Prado, T. (2019). Partidos de esquerda tentam reatar diálogo nas igrejas evangélicas [Left-wing parties try to resume dialogue in evangelical churches]. O Globo. Retrieved from https://oglobo.globo.com/brasil/partidos-de-esquerda-tentam-reatar-dialogo-nas-igrejas-evangelicas-23949492
Quadros, M. P. R., & Madeira, R. M. (2018). Fim da direita envergonhada? Atuação da bancada evangélica e da bancada da bala e os caminhos da representação do conservadorismo no Brasil [End of the embarrassed right? Performance of the
evangelical caucus and the bullet bench and the ways of representing conservatism in Brazil]. Opinião Pública, 24(3), 486–522. doi:10.1590/1807-01912018243486
Queiroz, C. (2019). O crescimento da fé evangélica [The rise of the evangelical faith]. Nexo. Retrieved from https://www.nexojornal.com.br/externo/2019/12/09/O-crescimento-da-f%C3%A9-evang%C3%A9lica
Ramirez, S. E. (2005). Religion: Latin America. In M. C. Horowitz (Ed.), New Dictionary of the History of Ideas (Vol. 5, pp. 2066–2068). Detroit, MI: Charles Scribner's Sons.
Sales, C. M. R., & Martins Filho, J. R. (2018). The Economist and Human Rights
Violations in Brazil During the Military Dictatorship. Contexto Internacional, 40(2), 203–227. doi:10.1590/s0102-8529.2018400200009
Scammell, M. (1999). Political Marketing: Lessons for Political Science. Political Studies, 47(4), 718–739. doi:10.1111/1467-9248.00228
Schwartz, S., & Farkas, J. (2018). Please Like, Comment and Share our Campaign! How Social Media Managers for Danish Political Parties Perceive User-Generated Content. Nordicom Review, 39(2), 19–33. doi:10.2478/ nor-2018-0008
Sherkat, D., Powell-Williams, M., Maddox, G., & de Vries, K. (2011). Religion, politics, and support for same-sex marriage in the United States, 1988–2008. Social Science Research, 40(1), 167–180. doi:10.1016/j.ssresearch.2010.08.009
Shogan, C. J. (2006). The moral rhetoric of American presidents. College Station, TX: Texas A&M University Press.
Strandberg, K. (2013). A social media revolution or just a case of history repeating itself? The use of social media in the 2011 Finnish parliamentary e l e c t i o n s . N e w M e d i a & S o c i e t y , 1 5 ( 8 ) , 1 3 2 9 – 1 3 4 7 . doi:10.1177/1461444812470612
Studart, P. H. M. (2013). Quando, afinal, há segundo turno em uma eleição?. Tribunal Superior Eleitoral. Retrieved from http://www.tse.jus.br/o-tse/escola-judiciaria- eleitoral/publicacoes/revistas-da-eje/artigos/revista-eletronica-eje-n.-6-ano-3/quando-afinal-ha-segundo-turno-em-uma-eleicao
Szajkowski, B. (1983). Next to God..., Poland : politics and religion in contemporary Poland. London: Pinter.
Szlek Miller, S. (1997). Religion and Politics in Poland: The Abortion Issue. Canadian Slavonic Papers, 39(2), 63–86. doi:10.1080/00085006.1997.11092143
Tarouco, G. S., & Madeira, R. M. (2013). Esquerda e direita no sistema partidário brasileiro: análise de conteúdo de documentos programáticos [Left and right in the Brazilian party system: content analysis of programmatic documents]. Revista Debates, 7(2), 93–114.
Toledo, C. N. (2014). A democracia populista golpeada. 1964: visões críticas do golpe: Democracia e reformas no populismo. Campinas: Unicamp.
Vargas, M. & Behnke, E. (2019). Bolsonaro reúne lideranças evangélicas em culto no Planalto [Bolsonaro gathers evangelical leaders in worship at the Planalto]. Estadão. Retrieved from
https://politica.estadao.com.br/noticias/geral,bolsonaro-reune-liderancas-evangelicas-em-culto-no-planalto,70003129353
Wald, K. D. (1986). Religion and Politics in the United States. New York, NY: St. Martin’s Press.
Weber, C., & Thornton, M. (2012). Courting Christians: How political candidates prime religious considerations in campaign ads. Journal of Politics, 74(1), 400–413. doi:10.1017/S0022381611001617
Woo, T. T. (2019). Religion and Politics in the People’s Republic of China: An Appraisal of Continuing Mistrust and Misunderstanding. Religions, 10(5). doi:10.3390/rel10050333 Zilla, C. (2018). Evangelicals and Politics and Latin America. Stiftung Wissenschaft und
Politik. Retrieved from
Appendix A: Codebook
CODEBOOK: CONTENT ANALYSIS
The coding process should be manually performed through an Coding Form (annex) that contains a total of 1260 posts retrieved from the Facebook pages of the four frontrunners of the Brazilian 2018 presidential elections: Jair Bolsonaro (JB), Fernando Haddad (FH), Ciro Gomes (CG) and Geraldo Alckmin (GA).
Part 1 - Content Selection
All 1260 posts from the data list must be coded according to the the six categories that will be further explained. The entirety of the content posted within the proposed timeframe have been previously selected and assembled, except those that a) do not contain any text (only photos or videos), b) have been shared repeatedly, c) contain only hashtags, d) have only emojis.
Part 2 - Form Organisation
The Coding Form is organised as follows:
- Candidate code: the initials of the candidate’s names;
-
Post: the content of each post;-
Date and time: when the content was posted;-
Number of likes (previously retrieved);-
Number of comments (previously retrieved);-
Six different categories to be coded for (see Part 3).Part 3 - Guidelines
Step 1. The coder must go through the conceptualisation of the six different parameters (see Part 3) and get familiar with the words that constitute the categories.
Step 2. Go to the Coding Form and read each of the posts.
Step 3. Identify if words for each of the categories have been mentioned. Fill the empty spaces for each category with:
0 = Absent 1 = One mention 2 = Two mentions 3 = Three mentions 4 = Four mentions 5 = Five mentions
