RLCS, Revista Latina de Comunicacion Social



Revista Latina




RLCS and Scopus

DOI, Digital Objetc Identifier 10.4185/RLCS-2018-1241en | ISSN 1138 - 5820 | RLCS, 73-2018 | Audio-visual explanation of the author |

Index h of the journal, according to Google Scholar Metrics, g


How to cite this article in bibliograhies / References

L Valera-Ordaz, D Calvo, G López-García (2018): “Political conversations on Facebook. Exploring the role of homophily in argumentation and communicative interaction”. Revista Latina de Comunicación Social, 73, pp. 55 to73.
DOI: 10.4185/RLCS-2018-1245en

Political conversations on Facebook. Exploring the role of homophily in argumentation and communicative interaction

Conversaciones políticas en Facebook. Explorando el papel de la homofilia en la argumentación y la interacción comunicativa

Lidia Valera-Ordaz [CV] [oORCID] [gGS] – Universidad de Valencia, UV – lidia.valera@uv.es

Dafne Calvo [CV] [oORDIC] [gGS] – Universidad de Valladolid, UVa – dafne.calvo@uva.es

Guillermo López-García [CV] [oORCID] [gGS]] – Universidad de Valencia, UV – guillermo.lopez@uv.es

Research on online political talk has been overwhelmingly influenced by deliberative theory, as a result of the big hopes originally placed on the Internet capacity to revolutionize politics. However, two decades of empirical research point out that these informal communicative interactions hardly resemble deliberative ideals. In this study, we explore if Facebook contributors who share ideological stances are more likely to a) justify their points of view and b) interact with each other. We apply content analysis to a sample of 2,800 comments published on the Facebook walls of the four mayor Spanish political parties during General Election campaign 2015. Findings show that homophily is negatively related to both argumentation and interaction. It is the outsiders who are much more likely to reason their opinions, while it is those who hold neutral views who are more inclined towards interaction [1].

Facebook; Online political conversations; Homophily; Argument; Reciprocity; Digital public sphere.

1. Introduction and theoretical framework. 1.1. Research questions. 2. Methodology. 3.1. Sample selection and collection. 2.2. Measures. 4. Results. 3.1. Homophily as a general trend of Facebook conversations. 3.2. Exploring the relationship between homophily and argumentation. 3.3. Exploring the relationship between homophily and reciprocity. 4. Discussion and conclusion. 5. Notes. 6. References.

Escrito en inglés directamente por Lidia Valera Ordaz

 [ Research ] [ Funded ] 
| w | Metadata | File PDF to print | Dynamic presentation - ISSUU | Paper with license Creative Commons | References | XML |
| Series of files for e-books| mobi | htmlz + lit + lrf + pdb + pmlz + rb + snb + tcr + txtz |

1. Introduction and theoretical framework

The emergence of social media has attracted huge academic interest about their implications for digital politics, since the increasing penetration of digital technologies among the population can potentially transform the relationship between politicians and the electorate (Lilleker & Jackson, 2010; Fuchs, 2013). According to Nielsen & Vaccari (2013), researchers have studied the rise and use of the Internet in the political communication realm from different approaches. Some researchers have explored the engagement of citizens with political actors and their mutual interaction during election campaigns; others have discussed whether (ant to what extent) the specific communication platforms designed by political elites have facilitated large-scale direct communication with the citizenry. The present work falls within a third line of research, which attempts to understand how citizens interact with each other in social media, and how these online conversations may be valuable for democracy.

Several authors highlight the significance of Facebook and Twitter as channels of political communication practices and electoral campaign tools, given the extensive number of citizens using them regularly (Kreiss, 2012; Nielsen & Vaccari, 2013; Larsson & Kalsnes, 2014; Yung et al., 2014). In Spain, Facebook is clearly more popular than Twitter, according to different surveys conducted on both the general population (Centro de Investigaciones Sociológicas, 2015) and Internet users (Asociación para la Investigación de Medios de Comunicación, 2016). For instance, according to a recent online survey applied to a big sample of Spanish Internet users, Facebook is the second most visited website, just behind Google, while Twitter only reaches the eight position in the ranking (AIMC, 2016).

Furthermore, the use of social networks was also widespread during the 2015 Spanish election campaign. According to a study conducted by the Centro de Investigaciones Sociológicas (CIS), 69% of the Spanish general population reported that they had used the Internet during the 2015 electoral campaign. Among these, 70% had a Facebook account, while only 25% were on Twitter and 16% on Instagram (CIS, 2016). The relevance of Facebook does not only stem from the number of users who have an account on it, but also from the frequency of usage. Thus, 91% of social media users visit Facebook more frequently than any other social medium, whereas only 4% of them do so with Twitter (CIS, 2016).

This intense use of Facebook entails new opportunities for political participation (Lilleker & Jackson, 2010; Kruikemeier et al., 2014; Warren et al., 2014; Boulianne, 2015). Although Facebook basically attracts people for entertainment and social purposes, this type of use can easily lead to other activities related to public affairs. In other words, users can be incidentally confronted on their Facebook accounts with different political activities, such as initiatives undertaken by their representatives, activities organized by social movements or public protests emerging from the civil society. Indeed, despite most social media users do not pay special attention to political information and the profiles of political parties and their leaders are not extensively followed, their Facebook pages can acquire significant attention in particular moments of civic excitement, such as general elections (Nielsen & Vaccari, 2013; Yung et al., 2014).

It is also during campaign periods when political actors significantly increase their efforts to reach and persuade voters, using Facebook as a means for their political aims (Yung et al., 2014). Kreiss (2014) suggests that the major challenge that political actors face in electoral campaign periods is not handling participation, but promoting real participation. In other words, campaign volunteers spend considerable time in very demanding unpaid activities, and they need to feel motivated and inspired in order to legitimate their mobilizing and working efforts. From this perspective, the management of social media platforms is aimed at coordinating voluntary work and facilitating collective action. However, these forms of promoting participation through social media do not necessarily imply enhancing deliberation as conceived by deliberative theorists (Kreiss, 2012).

Along with the academic literature on the relationship between social media usage and civic engagement, some researches have focused on how specific factors, such as party size or ideology, influence the use that political actors make of different digital platforms. Cardenal (2013) argues that large parties tend to have more incentives to use a variety of online platforms (including social media) for political mobilization because they have reasonable expectations to win elections and they own more resources to invest in both online and offline activities, in comparison with smaller parties. For their part, Larsson & Kalsnes (2014) focus on the Facebook use by Norwegian and Swedish politicians, and they conclude that representatives with less vote percentage are more likely to use social media. Through their study on the routine uses of social media by Norwegian and Swedish politicians, they observe that it is the individual politicians’ characteristics which most determines social media adoption and use, rather than any other contextual variables (such as party size or ideology).Thus, although using Facebook for electoral purposes requires both financial and human resources, it is not clear that Internet campaigning is dominated by bigger parties (Gibson et al., 2008).

Other researchers have explored the effects of ideology on the Internet use by political actors, but evidence is conflictive. For instance, Lilleker & Jackson (2010) consider right-wing parties to be apparently more open to interaction, but they warn that these predictions need more statistically rigorous tests to be confirmed. For its part, Sudulich argues that left-wing parties are more likely to interact with Internet users online (2013). Cardenal (2013) points out that the degree of ideological cohesion of political parties influences the diffusion of their messages through the Internet. More specifically, political parties with very low or very high ideological cohesion expect their messages to have coherence and they avoid publishing discrepancy, while political organizations with intermediate levels of internal cohesion take the risk of publishing internal discrepancy (Cardenal, 2013).

In general terms, the rise of a more professionalized campaign style has normalized and homogenized the practices of digital politics, regardless of contextual factors (Gibson et al., 2008). Recent studies show that ideology does not influence the use of digital technologies by political parties (Vesnic-Alujevic, 2016). Some even refer to this phenomenon as an ongoing deideologization of political actors’ online activities (Larsson & Kalsnes, 2014). In Spain, the emergence of party primaries or the involvement of citizens during election campaigns are a clear evidence of the increasing americanization of political campaigns since the nineties (Dader, 1999). However, as Lisi argues in analyzing the Portuguese case, despite the increasing professionalization of political campaigns in Western Europe, “the features of the electoral and party systems, as well as the ideological and organizational characteristics of parties are also important elements that help explain campaign professionalization” (2013: 273).

The emergence of digital technologies has also aroused interest in how they could influence citizens’ political participation and political knowledge, and how this could influence civic engagement and political efficacy (Prior, 2005; Vaccari et al., 2015). There is a great deal of literature which has analyzed the characteristics of spontaneous citizen interactions on the Internet (see, for example, Wilhelm, 1998; Jensen, 2003; Graham, 2009; Ellison et al., 2010; Warren et al., 2014). This interest in the so called digital public sphere is explained through the relevance attributed to informal political by political theorists for democracy (Stromer-Galley & Wichowski, 2011; Shah, 2016). Indeed, political talk has been often defined as the soul of democracy.

Thus, researchers have explored online political talk for the last two decades, looking into a number of online platforms, such as newspapers’ comments sections (Graham & Wright 2015; Ruiz et al., 2010), blogs (Kaye et al., 2012; Valera, 2014) or social media, including Twitter (Balcells & Padró Solanet, 2015) and Facebook (Camaj et al., 2009; Sørensen, 2016; Valera-Ordaz, 2017). Special attention has been paid to the online discussions taking place on explicit political fora, i.e. web platforms sponsored by political candidates, political parties and government institutions. Indeed, some researchers have claimed that most studies have concentrated on the Internet practices of formal politics (Wright, 2011; Graham et al., 2015), while those of ordinary citizens on non-explicitly political spaces have rather been overlooked (Coleman & Blumler, 2009). These third spaces, i.e. non-political spaces where political talk also emerges, are, in fact, significantly under-explored.

It is worth noting that most approaches analysing online political talk have relied on highly normative notions of political discussion, grounded in deliberative theory and the Habermasian conception of the public sphere (Friess & Eiders, 2015; Graham, 2015; Valera, 2017), probably as a result of the hopes originally placed on digital technologies to revolutionize politics and transform democracy (Wright, 2011). Thus, researchers have used indicators derived from deliberative democracy theory, and then measured up to what extent spontaneous online citizen conversations resemble the ideal of deliberation, i.e. exhibit deliberative quality. The result of this line of research has been the neglect of the revolutionary scenario predicted by the cyber-optimist school. That is, findings have clearly indicated that online political conversations cannot be conceived as high quality deliberations, which is no surprise given that they are not aimed at decision making (Birchall & Coleman, 2015).

Moreover, research on Facebook political discussions has shown that most conversations taking place in explicitly partisan online fora are dominated by like-minded people, i.e. they take place among contributors who generally share the same ideological stances (Valera, 2012, 2017; Camaj & Santana, 2015; Sweetser & Lariscy, 2008). This structural lack of ideological diversity clearly hinders any possible deliberation conceived as interactive communication which includes the exchange of arguments between individuals who hold different views and political values. According to different studies, partisan Facebook pages rather serve as spaces for sympathizers and activists to gather around their preferred candidates and parties (Robertson et al., 2010), convey support messages to them, and express frustration towards the political opponents (Fernandes et al., 2010). In other words, homophily is a well-established characteristic of partisan online fora, to such an extent that sometimes the dissent found in partisan Facebook pages responds to a radical sector of the political parties who host discussions (Valera, 2012).

Studies also show that conversations on the Facebook profiles of political candidates can be shallow (Camaj et al., 2009), since most contributors convey support messages to the political actors, but do not engage in rational discussion or mutual interaction (Sweetser & Lariscy, 2008). However, the candidates of fringe parties seem to host more interactive conversations, and contributors are more likely to exchange arguments, probably because they lack other traditional resources (Valera, 2017). These discussions are also more homophilic, since participants are hardly exposed to dissent.

In sum, research has clearly established that discussions on the Facebook profiles of political actors do not resemble deliberations as conceived by deliberative theorists (Elster, 2001; Martí, 2006), since they lack some essential deliberation features, such as ideological diversity, although others are generally met, as it is the case of discursive freedom. Discussions rather resemble the “enclave deliberations” described by Sunstein in his work Republic.com, that is, spaces where contributors basically listen to the “louder echoes of their own voices” (Sunstein, 2003: 65).

Some studies have recently moved beyond the confines of deliberative theory in analyzing online political talk, and have explored online discussions according to different democratic logics, using metrics derived from political philosophies such as liberal-individualism, communitarianism and deliberation (Freelon, 2010, 2013). They argue that deliberation is not the only valuable model of democratic communication, and should not be the sole theoretical framework used to explore online political talk (Coleman & Blumler, 2009; Freelon, 2010).

We contend that partisan Facebook pages will probably manifest a communitarian logic, given that most of the contributors who publish comments on them share the ideological stances of the parties hosting the conversations (Camaj et al., 2009; Valera, 2012; 2017; Fernandes et al., 2010), and that social media are basically used by political actors for advertising purposes (Vesnic-Alujevic, 2016; Muñiz et al., 2016). Therefore, we aim to explore if the Facebook pages of political parties are hosting intra-ideological discussions. In other words, if communicative interaction and argumentation are more likely to happen among contributors who share the ideological stances of the parties hosting conversations.

1.1. Research questions

This study explores Facebook political conversations hosted by the four most important Spanish formal political organizations during 2015 electoral campaign. Our goal is to observe if the homophilic nature of online political conversations translates into more in-group interaction among Facebook party followers. Different studies have shown that online deliberation, conceived as rational discussions held among citizens who exhibit strong ideological disagreement, is not to be found in explicitly partisan online platforms. Nor is it on social media, such as Facebook. However, these online spaces might be hosting intra-ideological discussions, that is, conversations between people who share basic political values and beliefs, as it is the case with party followers and sympathizers. In other words, we intend to analyze if those people who share the ideological stances held by the political parties are more likely to justify their points of view, and to interact with each other. We therefore contend that ideological agreement with the political actor who hosts Facebook discussions might influence the willingness towards argumentation and interaction. More specifically, the study tries to answer the following questions:

  • RQ1. Are contributors to the Facebook pages of Spanish political parties who share ideological stances more likely to justify their opinions?

  • RQ2. Are contributors who share ideological stances more likely to interact with each other?


2. Methodology

In order to analyze discussions on the Facebook pages of political parties, we selected the Facebook profiles of the four most important Spanish political parties: the People’s Party (PP), the Socialist Party (PSOE), the Citizens Party (C’s) and Podemos. These four political organizations were selected because they showed the highest voting intention during the 2015 electoral campaign according to different polls, and ended up attracting more of the popular 85% vote on Election Day (20th December 2015).

Our time frame is limited to the two week campaign period when political parties are officially allowed to campaign, because we expect more comments on the Facebook pages of the political parties during this time frame. Specifically, our time frame includes 15 days: from December 4th to December 18th, both days included. Focusing on the campaign period guaranteed us that we would encounter a good deal of activity on the partisan Facebook pages, both in terms of content published by the political organizations and user-generated commentary.

Regarding the data collection process, we first collected all the Facebook posts published by the four political parties and all the comments published by the citizens in response to them through a specific software called Facepager. After doing so, we randomly selected one post per day per political party for practical reasons, since the total volume of posts and comments clearly exceeded our capability of manual coding. In this way, we covered the whole campaign period and we guaranteed that our sample would not be biased or heavily influenced by special events on specific campaign days, such as electoral debates, which generally attract a great deal of attention.

Table 1: Sample composition

Source: created by the authors.

For practical reasons, we limited the number of comments per post to the first 50 comments. In this way, our final sample consists of 2,800 comments (Table 1). The observed differences in the samples of each political party are due to the fact that sometimes their Facebook posts attracted less than 50 comments, normally because of a very intense publication rate. Still, these differences are not particularly relevant.

In order to tackle our research question, i.e. if ideological agreement affects the willingness of contributors to justify their points of view and to interact with each other, we need to establish metrics accounting for ideological agreement, argumentation and communicative interaction.

First, we use one metric to explore the extent to which online political discussions hosted by political parties on Facebook are characterized by ideological diversity. We do this through a variable called homophily. Homophily (from Latin: love for the same) is defined as the existence of ideological agreement between the content published by the political party hosting the discussion and the comments made in response. Thus, all Facebook comments are coded as: a) contrary, b) favourable or c) neutral to the opinion expressed by the political actors in their Facebook posts. Through homophily we can observe to what extent online political talk on the Facebook pages of Spanish political parties is dominated by like-minded people. The more favourable comments, the more homophilic the conversation dynamic, and, as a consequence, the more ideological correspondence between Facebook followers and political parties hosting discussions.

Table 2: Variables

Source: created by the authors.

We also propose two metrics derived from deliberative democracy theory in order to explore if rational argumentation and interaction are more frequent phenomena among those who share the ideological stances expressed by political parties in their posts. More specifically, we choose two indicators which represent two essential features of deliberations: a) argumentation and b) reciprocal interaction. Argument accounts for the presence of reasoned claims. It is operationalised as the use of causal conjunctions (because, since, etc.) or the explicit inclusion or evidence (such as links to studies, surveys, news stories, etc.) by Facebook users in their comments.

 Reciprocity represents another essential characteristic of deliberation, because participants in deliberation (also in those which take place among people sharing the same ideological values) are supposed to listen to other people’s views, exchange reasons and be opened to be persuaded by others. We have defined reciprocity as the mutual interaction between contributors. We operationalize it through two indicators: a) the use of vocatives (which reflects a desire of interaction), and b) the inclusion of arguments previously exposed by other contributors.

The three outlined variables were applied to our sample through content analysis. The unit of analysis was the individual comment. Still, we need to point out that each comment was analysed in the broader context of the Facebook thread where it was originally published, taking into account the content of the political post and the rest of the comments included in the same thread. All three variables were coded by six coders, who were trained until reliability reached suitable levels. Inter-coder reliability tests were applied to 10% of the sample, i.e. 280 comments. For the three variables inter-coder agreement exceeded 85% and Krippendorff’s alpha values ranged between .65 and .70.

3. Results

Results are presented in the following way. First, we discuss the presence of homophily in the Facebook conversations hosted by Spanish political parties, and we do so by examining any possible differences across political parties (Table 3). After exposing the general significance of ideological homogeneity in Facebook conversations, we move to the exploration of our first research question (RQ1), i.e. the relationship between homophily and argumentation (Table 4). Finally, we analyze the association between homophily and reciprocity, addressing our second research question (Table 5).

All of the results are presented through contingency tables. Besides the descriptive information included in the contingency tables, we performed Pearson chi-square tests to explore if homophily was associated with each of the two outlined metrics (argument and reciprocity). Both tests were significant, indicating that both variables were related with homophily. In order to perform further exploration of both associations, we calculated the standardized adjusted residuals and included them in the contingency tables. The standardized adjusted residuals show which cells contribute the most to the to the Chi-square statistic. That is, which associations are stronger, and which categories of the variables are mores strongly associated.

3.1. Homophily as a general trend of Facebook conversations

Table 3 shows the general ideological composition of Facebook conversations hosted by Spanish political parties, and a detailed description of how homophily varies by political party. In general terms, the majority of contributors publish favorable (45.4) or neutral (43.5) comments, and only a minority of Facebook users express ideological dissent (11.2). In other words, there is a great deal of ideological agreement between the political parties who hold the Facebook profiles and the users who participate in them by publishing a comment during the campaign. Thus, we can clearly confirm our assumption that partisan Facebook pages basically attract like-minded people to express favorable or neutral claims. Indeed, results suggest that comments are probably attributable to party activists and sympathizers, as pointed out by previous research in different political contexts (Camaj et al., 2009; Camaj & Santana, 2015; Fernandes et al., 2010; Valera-Ordaz, 2017).

If we examine the differences by political party through the standardized adjusted residuals, we see that neutral comments are especially more frequent in the case of the Socialist Party, while they are significantly more unlikely to be published in discussions hosted by Podemos and the People’s Party. The presence of dissent is constant across parties, and no significant differences were found. More interestingly, dialogues hosted by political parties which hold more extreme views (PP for the right-wing and Podemos for the left-wing) attract significantly more favorable comments, so that homophily is more extended in conversations hosted by more extreme political parties.

Table 3: Homophily by political party (in percentages)

SR: Standardized residual
* Values > +1.96 or < -1.96 indicate bivariate association at 95% confidence level.
** Values > +2.58 or < -2.58 indicate bivariate association at 99% confidence level.
Source: created by the authors.


3.2. Exploring the relationship between homophily and argumentation

Table 4 shows the cross tabulation of argument and homophily, and the univariate distribution of argument. In general, only 18% of the Facebook comments published on the walls of political parties contain reasoned claims, while the vast majority of them (82%) consist of non-justified political preferences. But after cross tabulating both variables, some differences emerge depending on the ideological agreement of each comment with the views expressed by the political parties.

Indeed, the chi-square test performed to explore the association between argument and homophily is clearly significant, i.e. both variables are associated, as showed by the standardized residuals. More specifically, 37% of the comments expressing dissent contain some form of argumentation, a rate which more than doubles the general presence of reasoned claims (18%). In the case of comments holding neutral or favourable views, however, argumentation is significantly less likely to happen. It only appears in 14.5% and 16.7% of the comments.

What do these results mean? They mean that comments expressing dissent on the Facebook pages of political parties are much more likely to include reasoned claims, in comparison with those containing neutral or favourable views. Interestingly enough, it is not the contributors supporting the political parties, i.e. sharing their ideological stances, who more frequently justify their opinions. Even though they enjoy a favourable discursive environment, where the majority of users share their preferred values and beliefs, they are significantly less inclined to justify their points of views. It is the outsiders who are significantly more willing to engage in rational argumentation. Probably, this happens because outsiders face a hostile environment, which is clearly dominated by party activists and sympathizers, and using argument to support their opinions helps them to feel more confident to get their voices heard.

Table 4: Argument by homophily (in percentages)

SR: Standardized residual
* Values > +1.96 or < -1.96 indicate bivariate association with 95% confidence level.
** Values > +2.58 or < -2.58 indicate bivariate association with 99% confidence level.
Source: created by the authors.

 In sum, our findings point out that argumentation is not more likely to happen among like-minded people, but rather among contributors who comment on partisan Facebook posts but do not share political parties’ ideological stances. Findings clearly show that expressing ideological disagreement is associated with a greater tendency to engage in rational argumentation.

4.3. Exploring the relationship between homophily and reciprocity

Table 5 includes the cross tabulation of reciprocity and homophily, and the univariate distribution of reciprocity. It shows that only a 16.4% of the total comments reflect a desire to interact with other users, be it through the use of vocatives or through the inclusion of arguments previously used by other contributors. The majority of comments, however, do not present any inclination towards communicative interaction (83.6%).

Table 5: Reciprocity by homophily (in percentages)

SR: Standardized residual
* Values > +1.96 or < -1.96 indicate bivariate association with 95% confidence level.
** Values > +2.58 or < -2.58 indicate bivariate association with 99% confidence level.
Source: created by the authors.

The result of the Pearson chi-square test is also significant regarding the association of homophily and reciprocity, as shown by the standardized residuals. In other words, the desire to interact with other users is not equally distributed across ideological orientations. Facebook commentators who express ideological support to the political parties are significantly less inclined towards communicative interaction, while those users who express neutral views are much more likely to exhibit a desire of interaction.

More specifically, only 9% of those who share partisan ideological orientations exhibit some form of reciprocity, while 23.9% of contributors publishing neutral comments are willing to interact with each other. In the case of individuals expressing dissent, there are no significant differences. In simple terms, reciprocal interaction is less frequent among those who support political parties ideological orientations, and more likely to happen among contributors holding neutral views, who are more frequently pulled into interaction with other users.

4. Discussion and conclusion

This paper has explored Facebook political discussions hosted by the four most important Spanish political parties (People’s Party, Socialist Party, Citizens and Podemos) during the 2015 electoral campaign. Our goal was to observe if the well-established homophilic nature of Facebook conversations hosted by political organizations translated into more inclination towards argumentation and communicative interaction. Our intuition was that these online spaces could be hosting intra-ideological discussions, that is, conversations of political nature between people who share basic values and beliefs, as it is the case with party followers and sympathizers.

Given that the presence of ideological agreement makes it easier for individuals to express their opinions, elaborate them, and interact with each other, since they face a very low risk of social disapproval or isolation (Kuran, 1995; Noelle-Neumann, 1995), we expected those users sharing partisan ideas and beliefs to be more inclined toward argumentation and communicative interaction. But our findings clearly neglect nearly all of our assumptions.

First, our results confirm that homophily is the rule in the online discussions which take place in the Facebook pages of political parties (Camaj et al., 2009; Camaj & Santana, 2015; Valera-Ordaz, 2012; Valera-Ordaz, 2014). Most commentators actually share the same ideas and values as those expressed by political actors in their Facebook posts. Others simply express neutral views, but dissent always represents a minority for the four analyzed political parties. Second, our findings also show that homophily is associated with both argumentation and reciprocity. But these associations adopt an unexpected shape.

Regarding our first research question (RQ1), our findings point out that ideological agreement with the political actor hosting Facebook discussions does clearly not increase the willingness towards argumentation. Quite the opposite, ideological agreement is associated with a lower likelihood of justifying opinions. In other words, argumentation in Facebook conversations is more likely to happen among contributors who express dissent towards the political parties which host the discussions.

Those who make the effort to express dissent and expose themselves to a majority of users who will probably disagree with them, i.e. who take the risk of facing a hostile discursive environment, do so by more frequently reasoning or justifying their opinions. Interestingly enough, ideological disagreement seems to trigger more elaboration and justification of political opinion. Still, it should be noted that we have only performed a bivariate exploration of this relationship, so that future studies will have to apply multivariate analysis (including other possible influencing factors) in order to confirm these empirical findings.

Regarding our second research question (RQ2), which asked if ideological agreement is associated with more communicative interaction, our results indicate that it is those contributors who hold neutral views who more frequently interact with each other, while Facebook users who explicitly share partisan ideological views engage far less in reciprocal interaction. This suggest that users who intervene in the Facebook walls of political parties pursue different goals: while some basically want to express their ideological agreement and convey support messages, others look for communicative interaction.

In sum, homophily does not necessarily translate into intra-ideological discussions, where party activists and sympathizers interact with each other reasoning their opinions. Indeed, results suggest different patterns of behavior by those who comment on partisan Facebook walls. While a first group of contributors publish favorable comments to convey support to the parties but tend to avoid both communicative interaction and rational argumentation, other group of participants openly expresses opposition to the views and ideas held by the parties. These contributors are much more likely to justify their opinions with arguments and therefore engage in rational argumentation. For their part, a third group of users does not explicitly declare their ideological agreement or disagreement with the views and values expressed by the parties (since they publish neutral comments), but they are significantly more willing to engage in mutual interaction with other users.

In conclusion, the implications of our findings suggest that it is the users who do not explicitly share the ideological stances of political parties who introduce the most “deliberative potential” to the conversations taking place on Facebook. Indeed, it is them who more intensely justify their opinions through evidence, and who more often seek to interact with each other. However, future studies will have to confirm if the outlined bivariate associations between homophily and argumentation and communicative interaction, respectively, are strong enough controlling for other factors. In other words, future research will have to apply multivariate analysis (such as logistic regressions) to corroborate our empirical findings.

  • This study has been funded by the national research project “Estrategias, agendas y discursos en las cibercampañas electorales: medios de comunicación y ciudadanos”, funded by the Spanish Ministry of Economy (2017-2020) and directed by Professor Guillermo López García (reference CSO2016-77331-C2-1-R). The study has also benefited from postdoctoral research grant APOSTD2016 from regional Valencian government.


5. Note

[1] A previous version of this study was presented at the IPSA conference “Political Communication in Uncertain Times: Digital Technologies, Citizen Participation and Open Governance”, in September 2017 at the University of Pamplona.


6. References

AIMC. Asociación para la investigación de medios de comunicación. (2016): 18º Navegantes en la Red. Encuesta AIMC a usuarios de Internet. Available at: http://download.aimc.es/aimc/ROY76b/macro2015.pdf [Accessed 5 Feb. 2017].

J.  Balcells & A. Padró-Solanet, (2015): Tweeting on Catalonia’s independence: The dynamics of political discussion and group polarization. Paper presented at the IPSA Regional Conference on Communication, Democracy and Digital Technology, Rovinj (Croatia), 2-3 October 2015 (forthcoming in Media Studies).

C. Birchall & S. Coleman (2015): Creating spaces for online deliberation. In: S. Coleman & D. Freelon (eds) Handbook of Digital Politics. Cheltenham, UK: Edward Elgar Publishing, 264-280.

S. Boulianne (2015): Social media use and participation: A meta-analysis of current research. Information Communication & Society, 18 5, pp. 524–538. DOI: 10.1080/1369118X.2015.1008542

L. Camaj, S. C. Hong, G. Lanosga & Y. Luo (2009): Political discourse on Facebook: A new public sphere? Paper presented at the annual meeting of the Association for Education in Journalism and Mass Communication, August 2009, Boston (USA).

L. Camaj & A. D. Santana (2015): Political deliberation on Facebook: The role of moderator’s technical role and political ideology for online deliberation. Journal of Information, Technology & Politics (accepted).

A. S.  Cardenal (2013): Why mobilize support online? The paradox of party behaviour online. Party Politics, 19 1, pp. 83-103.

Centro de Investigaciones Sociológicas (2015): Estudio panel CIS nº 7715: Preelectoral y Postelectoral Elecciones Generales 2015. Available at: http://www.cis.es/cis/opencms/-Archivos/Marginales/Globales/7715/Ft7715.pdf [Accessed 5 Feb. 2017].

Centro de Investigaciones Sociológicas (2016): Estudio CIS nº 3126: Postelectoral Elecciones Generales 2015. Available at: http://www.cis.es/cis/export/sites/default/-Archivos/Marginales/3120_3139/3126/Ft3126.pdf [Accessed 5 Feb. 2017].

S. Coleman & J.G. Blumler (2009): The Internet and Democratic Citizenship. Cambridge: Cambridge University Press.

J. L.  Dader (1999): La retórica mediática frente a la cultura política autóctona: la encrucijada de la comunicación política electoral española entre la americanización y el pluralismo democrático tradicional. Cuadernos de Información y Comunicación, 4, pp. 63-87.

N. B. Ellison, C. Steinfield & C. Lampe (2010): Connection Strategies: Social Capital Implications of Facebook-enabled Communication Practices. New Media & Society, 13 6, pp. 873–892. DOI: 10.1177/1461444810385389

I. Elster (ed.) (2001): La democracia deliberative. Barcelona: Gedisa.

J. Fernandes, M. Giurcanu, K. W. Bowers & J. C. Neely (2010): “The writing on the wall: A content analysis of college students' Facebook groups for the 2008 presidential election”. Mass Communication and Society, 13 5, pp. 653-675.

D. G. Freelon (2010): “Analyzing online political discussion using three models of democratic communication”. New Media & Society, 12 7, pp. 1172-1190.

D. G. Freelon (2013): “Discourse, architecture, ideology, and democratic norms in online political discussion”. New Media & Society, 20 10, pp. 1-20.

D. Friess & C. Eilders (2015): “A Systematic Review of Online Deliberation Research”. Policy & Internet, 7 3, pp. 319-339.

C. Fuchs (2013): “Twitter and Democracy: A New Public Sphere?” In Social Media: A critical Introduction, pp. 179–209, Los Angeles: SAGE.

R. K. Gibson, W. Lusoli & S. Ward (2008): “Nationalizing and normalizing the local? A comparative analysis of online candidate campaigning in Australia and Britain”. Journal of Information Technology & Politics, 4 4, pp. 15-30.

T. Graham (2009): What's Wife Swap got to do with it?: Talking Politics in the net-based public sphere. Amsterdam: Amsterdam School of Communications Research.

T. Graham (2015): “Everyday political talk in the internet-based sphere”. In: S. Coleman  & D. Freelon, (eds) Handbook of Digital Politics. Cheltenham: Edward Elgar, pp. 247-263.

T. Graham & S. Wright (2015): “A tale of two stories from ‘below the line’: comment fields at The Guardian”. International Journal of Press/Politics 20 3, pp. 1-22.

T. Graham, D. Jackson & S. Wright (2015): “From everyday conversation to political action: Talking austerity in online ‘third spaces’”. European Journal of Communication 30 6, pp. 648–665.

J. L. Jensen (2003): “Public Spheres on the Internet: Anarchic or Government-sponsored- a comparison”. Scandinavian Political Studies, 26 4, pp. 349-374.

B. K. Kaye, T. J. Jonson & P. Muhlberger (2012): “Blogs as a source of democratic deliberation”. In: T. Dumova & R. Fiordo R (eds) Blogging in the global society: Cultural, political and geographical aspects. Hershey: Information Science Reference, pp. 1-18.

D. Kreiss (2012): Taking Our Country Back: The Crafting of Networked Politics from Howard Dean to Barack Obama. Oxford: Oxford University Press.

D. Kreiss (2014): The virtues of participation without power: Campaigns, Party Networks, and the Ends of Politics. The sociological quarterly. DOI: 10.1111/tsq.12068.

S. Kruikemeier, G. Van Noort, R. Vliegenthart & C. H. De. Vreese (2014): “Unraveling the effects of active and passive forms of political Internet use: Does it affect citizens’ political involvement?” Media & Society, 0 0, pp. 1–18. DOI: 10.1177/1461444813495163.

T. Kuran (1995): Private Truths, Public Lies: The Social Consequences of Preference Falsification. Cambridge, MA: Harvard University Press.

A. O. Larsson & B. Kalsnes (2014): “Of course we are on Facebook”: Use and non-use of social media among Swedish and Norwegian politicians. European Journal of Communication Published Online, pp. 1–16. DOI: 10.1177/0267323114531383

D. Lilleker & N. Jackson (2010): Towards a more participatory style of election campaigning? The impact of Web 2.0 on the UK 2010 General Election. Policy & Internet, 2(3), 69–98.

M. Lisi (2013): “The professionalization of campaigns in recent democracies: The Portuguese case”. European Journal of Communication, 28 3, pp. 259-276. DOI: 10.1177/0267323113475463

J. L. Martí (2006): La república deliberativa: una teoría de la democracia. Madrid: Marcial Pons.

C. Muñiz, J. L. Dader, N. Marlen Téllez & A. Salazar (2016): “¿Están los políticos políticamente comprometidos? Análisis del compromiso político 2.0 desarrollado por los candidatos a través de Facebook”. Cuadernos.info, 39, pp. 135-150. DOI: 10.7764/cdi.39.970

R. K. Nielsen & C. Vaccari (2013): Do People “Like” Politicians on Facebook? Not Really. Large-Scale Direct Candidate-to-Voter Online Communication as an Outlier Phenomenon. International Journal of Communication, 7, pp. 2333–2356.

E. Noelle-Neumann (1995): La espiral del silencio: opinión pública, nuestra piel social. Barcelona: Paidós.

M. Prior, (2005): “News vs. Entertainment: How Increasing Media Choice Widens Gaps in Political Knowledge and Turnout.” American Journal of Political Science, 49 3, pp. 577-592.

S. P. Robertson, R. K. Vatrapub & R. Medina (2010): “Off the wall political discourse: Facebook use in the 2008 U.S. presidential elections”. Information Polity, 15, pp. 11-31. DOI: 10.3233/IP-2010-0196

C. Ruiz, P. Masip, J. L. Micó, J. Díaz-Noci & D. Domingo (2010): “Conversation 2.0. and democracy. An analysis of reader’s comments in Catalan online newspapers”. Communication & Society, 23 2, pp. 7-39.

D. V. Shah (2016): “Conversation is the soul of democracy: expression effects, communication mediation, and digital media”. Communication and the Public, 1(1), 12-18. DOI: 10.1177/2057047316628310

M. P. Sørensen (2016): “Political conversations on Facebook: the participation of politicians and citizens”. Media, Culture & Society, 38 5, pp. 664-685.

J. Stromer-Galley & A. Wichowski (2011): “Political discussion online”. In: Consalvo M & Ess C (eds) The Handbook of Internet Studies. United Kingdom: Blackwell, pp. 168-187.

M. Sudulich (2013): Do ethos, ideology, country and electoral strength make a difference in cyberspace? In: P. Nixon, R.  Rawal, & D. Tercea (eds), Politics and the Internet in Comparative Context: Views from the Cloud. Routledge, pp. 75-94.

C. R. Sunstein (2003): República.com: Internet, democracia y libertad. Barcelona: Paidós.

K. D. Sweetser & R. W. Lariscy (2008): “Candidates make good friends: An analysis of candidates’ use of Facebook”. International Journal of Strategic Communication, 2 3, pp. 175-198.

C. Vaccari, A. Chadwick & B. O’ Loughlin (2015): “Dual screening the political: media events, social media, and citizen engagement”. Journal of Communication, 65 6, pp. 1041-1061. DOI: 10.1111/jcom.12187

L. Valera-Ordaz (2012): “Deliberation or Radicalization of Partisan Discourse? An analysis of political discussions conducted on the Facebook pages of Spanish political candidates.” Text & Visual Media, 5, pp. 139-168.

L. Valera-Ordaz (2014): “El debate público en la blogosfera política española durante la campa-a electoral de 2011: ¿Hacia un espacio público enriquecido o fragmentado?” Revista Trípodos, 34, pp. 153-170.

L. Valera Ordaz, G. López García (2014): “Agenda y marcos en las webs de PP y PSOE en la cibercampaña de 2011”. Revista Latina de Comunicación Social, 69, pp. 41 a 66. http://www.revistalatinacs.org/069/paper/1000_Valencia/03_Valera.html 
DOI: 10.4185/RLCS-2014-1000

L. Valera-Ordaz (2017): “Comparing the democratic value of Facebook discussions across the profiles of Spanish political candidates during 2011 General Election.” Revista Internacional de Sociología, pp. 75 1. DOI: 10.3989/ris.2017.

L. Vesnic-Alujevic (2016): European Elections and Facebook: Political Advertising and Deliberation? In: Dezelan, T. & Vobic, I. (ed), (R)evolutionizing Political Communication through Social Media. IGI Global, pp. 191-209.

A. M. Warren, A. Sulaiman & N. I. Jaafar (2014): “Understanding civic engagement behaviour on Facebook from a social capital theory perspective”. Behaviour & Information Technology, 34 2, pp. 163–175. DOI: 10.1080/0144929X.2014.934290

A.G. Wilhelm (1998): “Virtual sounding boards: How deliberative is online political discussion?” Information, Communication & Society 1 3, pp. 313-338.

S. Wright (2011): “Politics as usual? Revolution, normalization and a new agenda for online deliberation”. New Media & Society, 142, pp. 244-261.

B. Yung, L. Y. Leung, N. Territories & H. Kong (2014): “Facebook as change? Political engagement in semi-democratic Hong Kong in its transition to universal suffrage”. Journal of Asian Public Policy, pp. 37–41. DOI: 10.1080/17516234.2014.922147.


How to cite this article in bibliographies / References

L Valera-Ordaz, D Calvo, G López-García (2018): “Political conversations on Facebook. Exploring the role of homophily in argumentation and communicative interaction”. Revista Latina de Comunicación Social, 73, pp. 55 to73.
DOI: 10.4185/RLCS-2018-1245en


Article received on 10 December 2017. Accepted on 14 January.
Published on 17 January 2018.