Sage Journals: Discover world-class research

Abstract

Online political talk is often deemed toxic to democracy due to the pervasiveness of incivility. This study challenges this perspective by examining the discursive and contextual conditions related to interpersonal incivility in contrast with incivility targeted at political elites on Facebook comments and news websites. Findings suggest that much of the vitriol online is targeted at politicians and characterized by justified opinions. Interpersonal incivility is associated with disagreement, but less likely to be associated with replies—suggesting that users refrain from uncivil direct confrontation. Taken together, these findings indicate that interpersonal-directed and elite-directed incivility are characterized by distinct discursive features and happen in opposite directions. Incivility is more than just shouting, and it is frequently used to criticize political elites and justify opinions than to attack others in a discussion. As such, it should not be inherently associated with toxic behaviors or be considered a problematic feature of online discussions.

Keywords

online incivility social media computer-mediated communication political discussion

Introduction

Informal political conversation is important for citizens to become aware and learn about matters of collective concern, construct personal and collective identities, and become active members of a community through the mutual recognition of shared values (Kim et al., 1999; Moy & Gastil, 2006; Wyatt et al., 2000). With the ubiquitous use of social media in Western democracies, scholars have been increasingly interested in political talk that emerges from informal spaces that are primarily designed for social and entertainment purposes (Rossini & Stromer-Galley, 2019; Wright, 2012). In these spaces, political discussions can emerge from different types of content shared by users, such as personal opinions about current affairs or news stories—which can increase people’s exposure to news and access to more diverse political opinions (Barnidge, 2015; Fletcher & Nielsen, 2018).

Despite substantive evidence of the potential benefits of political talk on- and offline, scholars tend to criticize online discussions for failing to meet standards of civility and deliberation (Anderson et al., 2014; Coe et al., 2014; Rowe, 2015; Santana, 2014). Although prior studies may raise attention to important consequences or implications of online political talk, they mainly focus on the presence of “impolite” and “uncivil” expression, with less attention devoted to understanding whether this type of discourse may have different rhetorical functions, such as targeting particular groups and individuals, or serving to amplify and call attention to certain opinions (Coe et al., 2014; Gervais, 2015; Santana, 2014). As a result, the literature on online incivility tends to portray the behavior as inherently offensive and incompatible with democratically valuable conversations without examining whether participants are expressing opinions or actively attacking one another in uncivil interactions.

To address this gap, this study examines the conditions in which incivility arises in online discussions by examining who is targeted by it. Given the concern that incivility is a sign of toxic online discussions, this study distinguishes between two targets of incivility—other participants in a discussion and political elites—with the goal of disentangling discourse, that is, offensive toward others in a debate from harsh and heated opinions about politics and politicians. Specifically, it examines the contextual and deliberative attributes associated with interpersonal-directed and elite-directed incivility in online political discussions. Considering that citizens have different perceptions of incivility based on whether it focuses on interpersonal attacks or political arguments (Kenski et al., 2020; Muddiman, 2017; Stryker et al., 2016), it is also possible that participants in online discussions use incivility with different goals when talking about politics. These dynamics are analyzed in a large dataset of comments to political news stories in Brazil across two platforms, Facebook and news websites, which have distinct affordances that may shape the ways individuals express political opinions online.

This study makes an important contribution to understanding contextual and conversational dynamics around online incivility and helps unveil some of the distinctive rhetorical functions of this type of discourse. As such, it provides a framework for scholars to consider the role of targets to disentangle harsh criticism toward the political sphere from interpersonal attacks. Contrary to the common sense that incivility online is inherently toxic to participants in a discussion, I find that that much of the vitriol online is targeted at political actors and characterized by justified opinion expression, suggesting that those who discuss politics online might be voicing a critical the political sphere—behaving as “monitorial citizens” (Zaller, 2003), ready to respond to news about issues they care about. However, uncivil comments targeting other participants are likely to surface in the context of disagreement and unlikely to include justified opinions, suggesting that interpersonally targeted incivility may reveal a toxic discussion environment where participants can learn a little about the other side, refraining from substantive debates. However, when participants are actively replying to one another in heterogeneous threads, they are less likely to direct incivility at other participants, suggesting that incivility may serve different rhetoric functions when there is room for substantive and interactive debate.

Taken together, these findings indicate that interpersonal-directed and elite-directed incivility are characterized by distinct discursive features and happen in opposite directions. While incivility may serve to emphasize arguments and express emotions when citizens are discussing attitudes and behaviors of political elites, it can also signal shouting matches where participants attack one another without exchanging substantive opinions. Importantly, direct replies and mentions reduce the likelihood of interpersonal-directed incivility in heterogeneous debates, suggesting that such “empty” offensive comments are less likely to emerge when participants are directly engaging with one another. These findings highlight the importance of considering how the features of uncivil discourse, such as being directed at different targets, can change the dynamics of online debates and elicit different deliberative attributes, helping scholars to distinguish situations in which online discussions can be relevant or productive in spite of having a heated tone. In sum, these results provide further support to the argument that scholars should focus on understanding online incivility and the conditions in which it arises instead of dismissing it as an inherently problematic dimension of computer-mediated political talk (Chen et al., 2019; Rossini, 2020).

Online and Uncivil

While digital platforms offer many opportunities for those interested in discussing politics online—in forums, newspaper websites, and on social media—scholars have argued that the Internet fails to promote democratically relevant political talk due to the excessive presence of incivility (Coles & West, 2016; Gervais, 2015; Phillips, 2015; Santana, 2014), in spite of evidence that online political talk is a precursor of more sophisticated forms of political engagement (Gil de Zúñiga et al., 2014; Miller et al., 2015; Vaccari et al., 2016). Since incivility can be found in nearly any online platform (Halpern & Gibbs, 2013; Oz et al., 2018; Papacharissi, 2004; Rowe, 2015; Santana, 2014), scholarship in this area needs to move beyond the “detection” of incivility to investigate the context around such expressions, as well as how it may affect participants and bystanders in a discussion.

Scholars have found potentially detrimental effects of online incivility, such as increasing polarization (Anderson et al., 2014), triggering uncivil responses (Gervais, 2015), affecting perceptions of persuasiveness (Chen & Ng, 2016), or perceptions of credibility of news outlets (Meltzer, 2015) and journalists (Searles et al., 2020). There is also some evidence of positive effects: uncivil blog posts may incentivize participation in the comments (Borah, 2014), and some people find uncivil social media messages “entertaining” (Sydnor, 2018).

However, these findings are based in different operationalizations, hindering comparisons—a consequence of the lack of consensus around how incivility should be defined (Jamieson et al., 2015). Political communication scholars tend to approach incivility as violations of politeness, focusing, for instance, on “features of discussion that convey an unnecessary disrespectful tone towards the discussion forum, its participants, or its topics” (Coe et al., 2014, p. 660). In analyses of online talk, common operationalizations include name-calling, ad hominem attacks, profanity, stereotyping, interpersonal disrespect, and using “caps lock” to represent shouting (Chen & Lu, 2017; Coe et al., 2014; Papacharissi, 2004; Sobieraj & Berry, 2011).

However, some scholars approach incivility as a spectrum that ranges from less harmful expressions to hateful and abusive speech, which makes it challenging to disentangle the conditions in which incivility is harmful from when it is not (Rossini, 2020). The use of incivility as a cover term that includes behaviors with varied levels of offensiveness and toxicity has undermined a more nuanced understanding of the role that incivility may play in online discussions (Papacharissi, 2004; Rossini, 2019). While some research suggests that incivility has detrimental effects, there is some empirical support to the argument that, for those involved in online debates, incivility might not be a deterrent for engaging in political talk. For instance, those who talk about politics online more frequently are more likely to demonstrate intention to be uncivil (e.g., using all caps, swearing, or insulting others), as well as to accept these behaviors in discussions (Hmielowski et al., 2014). Incivility is also perceived as engaging and entertaining on social media, and motivates participation—at least from people who are conflict-oriented (Sydnor, 2018, 2019). Moreover, incivility can be associated with relevant deliberative features (Chen, 2017; Rossini, 2020). In this article, incivility is operationalized as the tone, not the substance, of discourse. This includes the use of profanities and vulgarity, personal and ad hominem attacks, name-calling, and aspersions (Coe et al., 2014), which have been typically used as uncivil features of discourse by experimental research (Kenski et al., 2020; Mutz, 2016; Sydnor, 2019). Intolerant expressions, such as discriminatory or offensive stereotyping, violent threats, and hate speech, are excluded from this operationalization (Rossini, 2020).

Most studies on online incivility have primarily focused on identifying the extent to which features signaling uncivil “tone” are present in news comments, forum entries, and social media posts, but a few scholars have also considered the targets of incivility in their analysis. For instance, Rowe (2015) found that interpersonal incivility is more frequent when users are commenting on news websites, when compared to Facebook. Su et al. (2018) conducted a large-scale analysis of Facebook comments on news pages using automated classification and found a higher volume of online incivility “other-directed,” or impersonal, when compared to personal incivility—except in local news outlets. These findings suggest that not all online incivility is inherently toxic to participants, nor directed at them, but they provide limited insight into other potential targets of incivility. As well, while these studies hint at potential mechanisms to explain interpersonal incivility, such as anonymity (Rowe, 2015) and lack of moderation (Su et al., 2018), little is known about the conditions in which interpersonal incivility emerges, as well as about who (or what) is targeted by “other-directed” or impersonal forms of incivility.

To address these gaps, this study focuses on understanding the dynamics of uncivil discourse distinguishing between interpersonal-directed incivility—which could signal a toxic discussion environment—and political elites. Similar to Su et al. (2018),¹ Rowe (2015), and Papacharissi (2004), interpersonal-directed incivility identifies uncivil expressions targeted at other participants in a discussion (e.g., to attack or demean them). Unlike prior scholarship contrasting interpersonal incivility with a broader “impersonal” or “other-directed” category, this research examines a specific target political elites.

There are several reasons why political elites are relevant targets of uncivil discourse. First, while political officials are not typically responsive to citizens online, prior work has suggested that the public is often uncivil when engaging with politicians on social media—particularly toward women (Rossini et al., 2021; Southern & Harmer, 2021; Theocharis et al., 2016). However, scholars have focused on uncivil responses to politicians on social media—which could be seen as attempts to harass, silence, and offend elected officials—and attention has been given to the extent to which conversations about politics in response to news stories target politicians with incivility. While uncivil replies to politicians on social media might be seen as harassing, the use of uncivil language to talk about political elites in the context of news stories might be simply a proxy to elicit negative emotions toward the politician or the issue at hand, while still making a substantive contribution to the discussion. Such “strategic” use of incivility could serve different goals, such as raising attention and awareness to one’s arguments (Herbst, 2010), or reinforcing political identities to distinguish “us” and “them” (Berry & Sobieraj, 2016).

Second, political elites are public figures, who receive differential legal treatment, and their position of power subjects them to greater public scrutiny. At least in liberal democracies, politicians must also be open to criticism and disagreement. Despite attempts by politicians to block and silence critics online, there is widespread concern that the practice should be illegal. For instance, in 2019, the US Court of Appeals has ruled blocking citizens on social media as a violation of the First Amendment free speech rights,² and a draft bill currently under discussion by Congress in Brazil aims to rule the practice of politicians blocking critics online illegal.³ For these reasons, politicians are potentially less likely to be personally affected by uncivil comments, while members of the public may suffer direct consequences when targeted by such discourse (Vitak et al., 2017). Although there is evidence of harassment and abuse toward political representatives (Southern & Harmer, 2021), it can be argued that political elites hold a position of power in relation to the public and are less vulnerable to such online attacks—particularly, when these happen when the public is talking about politics, and not addressed to politicians.

Third, politicians often adopt uncivil rhetoric themselves (Mutz, 2016) and there is evidence that such behaviors are seen as appealing and entertaining for conflict-oriented individuals, who also tend to seek and participate in heated debates (Sydnor, 2019). Moreover, there is some evidence that the public may sometimes adopt uncivil rhetoric initiated by politicians—for instance, in 2016, uncivil comments on Hillary Clinton’s Facebook page often included insults that were initiated by her main opponent, Donald Trump (Rossini et al., 2021).

Finally, to the extent that calls for civility often come from those in power to silence dissent, and uncivil strategies may be necessary to amplify calls for change—political or otherwise (Young et al., 2010)—it is possible that elite-directed incivility is the symptom, not the cause, of political dissatisfaction and the decline of trust in formal political institutions.

When analyzing comments on news websites or on social media, it is important to remember that online comments are meant to be seen by others, and are not necessarily meant to invite discussion and back-and-forth conversations (Reagle, 2016). In the age of social media, stories that gain visibility are those that elicit engagement through comments and shares, potentially changing people’s perceptions about the relevance and importance of political topics. Some news websites also use engagement metrics to amplify the visibility of stories, listing the “most read” and “most commented” on the front page. In this sense, online commenters may act as “watchdogs” who contribute to informing their own networks about urgent matters insofar as they influence recommendation systems (Boczkowski et al., 2012). In this context, uncivil discourse around political stories that target political elites might be a consequence of the types of stories that attract higher attention—particularly as social media platforms become relevant channels for citizens to “monitor” the news (Ytre-Arne & Moe, 2018).

Thus, by understanding how expressions of incivility are used to target political elites in contrast with incivility used to attack other participants in a discussion, this study aims at disentangling situations in which incivility might have strategic rhetorical purposes when citizens talk about political issues from when it is used to attack others in a discussion. Not only interpersonal-directed incivility and elite-directed incivility may characterize completely different discussion contexts for both participants and bystanders, but the target of uncivil discourse may also elicit different interpretations. For instance, the public is more likely to classify a message as uncivil when it is focused on personal attacks, and less likely to do so when the focus is a political argument or policy, suggesting that, for bystanders, elite-directed incivility might be more acceptable than interpersonal-directed (Muddiman, 2017; Stryker et al., 2016). Name-calling and vulgarity are perceived as highly uncivil behaviors, while other types of incivility that characterize policy-oriented discussion—such as aspersions—are not (Kenski et al., 2020).

While a few studies have considered in- and out-group targets of incivility, they have mainly focused in the binary between interpersonal and out-group incivility (Maia & Rezende, 2016; Rowe, 2015), or “other-directed” (Su et al., 2018). Such approaches have been concerned with separating interpersonal incivility from any other “target,” thus not enabling a nuanced understanding of the potential rhetorical functions of incivility that is not targeted at other participants in a discussion. As a result, the perspective that incivility is necessarily bad, or harmful, is based upon the premise that it has the intention to offend others or disrupt the conversation.

Nevertheless, to the extent that it is a legitimate concern to identify the conditions in which people are personally attacked with incivility online, as those situations might be harmful to both targets and bystanders, understanding the conditions in which others are targeted with incivility remains a relevant concern. In this study, interpersonal-directed incivility is defined as uncivil behavior targeting other participants in a discussion. These are direct replies, mentions, and messages that clearly address someone else in a discussion (which does not include mentions to public figures) with an uncivil tone (for instance, using name-calling, making disparaging or offensive remarks, etc.). Interpersonal incivility may be detrimental for political talk to the extent that it may offend others and can ultimately impede dialogue. The notion of interpersonal-directed incivility is different than the idea of “personal-level incivility” by Muddiman (2017), which refers to discourse focused on personal characteristics and violations of politeness, respectively, but do not consider the target of incivility (e.g., an attack at a public figure focusing on personal characteristics would be an example of personal-level incivility).

While elite-directed incivility might signal heated rhetoric, disrespect, and deep dissatisfaction with political leaders, it is not inherently aimed at offending others in a discussion and should not necessarily impede substantive argumentation and discussion. Moreover, elite-directed incivility, in the context of news comments, does not necessarily imply that the public is interested in attacking or harassing political figures, In fact, it may be the case that those who routinely discuss politics online tend to adopt an uncivil tone to express their opinions about topics at hand (Hmielowski et al., 2014).

To investigate the extent to which uncivil expressions targeted at different actors are signaling distinctive discursive behaviors, I investigate the relationship between these targets and a set of deliberative attributes, such as the presence of disagreement, the use of justified opinion expression, and the use of direct replies. While I align with the argument that online political talk does not have to adhere to the normative expectations of deliberative discourse (Freelon, 2010)—a decision-making oriented process ideally characterized by justification, respect, inclusion of diverse views, orientation toward the common good, equality among speakers, and absence of political or economic coercion (Habermas, 1996)—the presence of some deliberative attributes may be a sign that uncivil online discussions may still yield some of the positive benefits of informal political talk—such as exposing people to diverse opinions and enable them to learn about others’ viewpoints (Oz et al., 2018; Rossini, 2020), which are intrinsic benefits of political talk.

Justification is a relevant attribute because it suggests that people are articulating their opinions in ways that others can understand which may, in turn, increase awareness and understanding of arguments in a debate (Stromer-Galley, 2007). There is a long-standing debate among deliberation scholars regarding the types of utterances that should count as justification in deliberative discussions, but the field has been consistently moving beyond rationality and recognizing that the deliberative goals can be still be met when other forms of expression are used as justification, such as story-telling, humor, or emotional speech (Black, 2009; Johnson et al., 2017). While this discussion is beyond the scope of this study insofar as informal political talk should not be expected to be fully deliberative (Parkinson & Mansbridge, 2012), the shift from the “form” of expression to the “substance” of what people say provides some support to approach justified opinion expression, regardless of its “quality,” as a signal of potentially productive debate insofar as people might learn something about the reasons behind others’ viewpoints. Another important deliberative attribute is disagreement, as discussions should ideally include diverse voices. Exposure to heterogeneous political arguments can increase one’s own argumentative repertoire as well as awareness of opposing views—which have been associated with increased tolerance and acceptance toward the other side (Eveland & Hively, 2009; Moy & Gastil, 2006; Mutz, 2006).

Prior research suggests that those who engage in political talk online may perceive incivility as normal and are more likely to be uncivil (Hmielowski et al., 2014). Starting from the premise that incivility can be used as a rhetorical asset to express political opinions (Herbst, 2010; Rossini, 2020), and that attacks on political views and arguments are perceived as less uncivil than personal attacks (Muddiman, 2017), I hypothesize that elite-directed incivility and interpersonal-directed incivility will differ in their association with justified opinion expression. Specifically, I expect that elite-directed incivility will be more likely than interpersonal-directed incivility to exhibit an association with justification. This relationship is hypothesized because, unlike direct attacks toward politicians (Southern & Harmer, 2021), elite-directed incivility in the context of discussing news of political should be more likely to contain justified political opinions insofar as incivility does not prevent conversations from being substantive or even “deliberative” (Chen, 2017). However, interpersonal-directed incivility does not necessarily warrant justification—particularly, if we accept the premise that interpersonal incivility aims at attacking and offending others:

H1. Elite-directed incivility is more likely to be associated with justified opinion expression than interpersonal-directed incivility.

Conversely, interpersonal-directed incivility should be likely to emerge in the context of disagreement, with personal attacks taking priority over substantive debate when commenters disagree with one another (Rösner & Krämer, 2016). Elite-directed incivility may not be as strongly associated with disagreement because those expressing uncivil opinions about politicians when commenting the news are not necessarily disagreeing with others. Similarly to politicians not being penalized by their base when using uncivil rhetoric toward opponents (Mutz, 2016), those talking about politics among like-minded others might be inclined to use incivility to emphasize their own political identities—a strategy widely adopted by pundits and political commentators (Berry & Sobieraj, 2016). Moreover, since comments on news stories are not necessarily discussion prompts and may simply reflect one’s own opinions and reactions (Reagle, 2016), the use of uncivil language toward politicians in the comments section may signal an opinion climate where this behavior may be seen as acceptable by those who participate in the comments section, particularly in the context of political dissatisfaction (Rossini, 2020). After all, incivility increases attention and recall of arguments (Mutz, 2016), and uncivil online comments are seen as more entertaining than civil ones (Sydnor, 2018). In short, while interpersonal-directed incivility might be characterized by conflicting opinion expression, the same may not be true when political elites are targeted as commenters do not need to disagree with others in the comments section to express their opinions about the state of affairs with uncivil intensity:

H2. Interpersonal-directed incivility is more likely to be associated with disagreement than elite-directed incivility.

To probe the hypothesized relationship between disagreement and interpersonal-directed incivility and investigate whether elite-directed incivility emerges in conversations in which participants are directly disagreeing with one another, I test an interaction between direct replies—comments in which participants address others by name or using a reply function—and disagreement:

RQ1. Does the hypothesized relationship between disagreement and elite-directed or interpersonal-directed incivility change in the context of direct replies?

The pervasiveness of incivility tends to be associated with inherent features of computer-mediated communication, which lacks non-verbal cues available in face-to-face interactions, as well as with platform affordances, such as the ability to participate anonymously in a discussion, and the ease to enter and leave conversations (Coe et al., 2014; Hmielowski et al., 2014; Maia & Rezende, 2016). Studies comparing anonymous and identified discussions in different online platforms suggest that identification can reduce the presence of incivility (Maia & Rezende, 2016; Rowe, 2015; Santana, 2014; Stromer-Galley & Wichowski, 2011). Conversely, studies focused on online deliberation have found that discussions in which participants have profiles or use real names tend to foster sincerity, rationality, and quality of justification (Davies & Chandler, 2013; Friess & Eilders, 2015; Janssen & Kies, 2005). Consistently with this literature, it can be hypothesized that the use of personal profiles, the maintenance of pre-existing relationship ties, and the visibility of users’ actions on social networking sites, such as Facebook, should act as a social constraint that could prevent participants to use interpersonal-directed incivility, when compared to news websites. The same may not necessarily be true for elite-directed incivility, as those who are target by it are not a part of the conversation, and platform constraints are an open question. Thus,

H3. Users will be less likely to use interpersonal-directed incivility on Facebook when compared to news websites.

RQ2. Is platform significantly associated with the expression of elite-directed incivility?

Internet Use and Political Context in Brazil

Brazil is the fifth most populous country in the world, with over 212 million habitants, and represents the second largest Western online market, with more than 74% of the population using the Internet (Núcleo de Informação e Coordenação do Ponto BR, 2020). Brazil is among the largest markets for social media platforms and accounted for over 102 million Facebook accounts in 2016 (Facebook, 2016). According to the Digital News Report, roughly 54% of the population uses Facebook for news (Newman et al., 2020).

Brazil is a presidential democracy with a multi-party system. When data for this study were collected, in 2015, the political climate was heated in the aftermath of the polarized presidential campaign electing Dilma Rousseff in 2014. The public was heavily divided and demonstrated a deep dissatisfaction with politicians in several public protests, bolstered by corruption scandals and the “Car Wash” Operation—which revealed an illegal scheme to favor business relations with the government and Congress. Since then, polarization has been growing consistently. Considering the heated political context in Brazil, it is plausible to expect that the popular mistrust and dissatisfaction toward the political sphere translated into uncivil expressions in online discussions about current affairs.

Methods

The data for this study comprised public comments on news stories published by the Facebook page of Portal UOL, Brazil’s largest online news portal, and comments in news stories posted to the websites where the news stories originated. UOL was selected for being the largest and most accessed online news portal in the country, and for hosting a diverse array of news outlets (e.g., national and regional newspapers, entertainment websites, and opinion blogs) in addition to having its own newsroom, which contributes to a more diverse sample in terms of topics, stories, and perspectives when compared to research using a single news outlet (e.g., Coe et al., 2014; Rowe, 2015).

Stories posted by Portal UOL’s Facebook page over the course of 6 months were collected using DiscoverText. Then, a set of stories was selected using the constructed week sampling technique to ensure representation of the variability of news on weekdays in the sample (Riffe et al., 2005). Following recommendations for online news, 2 weeks were constructed to represent 6 months of the news cycle (Connolly-Ahern et al., 2009). The initial sample contained 1,669 news stories posted by Portal UOL, before filtering topics. Using DiscoverText, each story was classified as being political or not, based on their headlines, adopting a broad scope of political topics including formal political topics—references to politicians, government-related issues and policies, corruption, any of the three branches of power—and a broader scope of social topics that are related to political issues, such as violence and education, as well as organized protests, stories about racism or discrimination, and stories about social struggles and inequalities. The assessment was made by the author, and a post hoc intercoder agreement test was conducted on two samples⁴ of the data to ensure the reliability of the sampling criteria (Krippendoff’s α = .79). After removing duplicated items (e.g., the same link posted twice on Facebook), and those with one or no comments, the final sample had 157 political news stories, which were categorized in the following subtopics by two independent coders: formal politics, public policy (education, security, violence), celebrities (e.g., engaged in political activities, or victims of discrimination), civil society, minorities, and international affairs (Krippendorff’s α = .72).

This study compares comments on Facebook to those posted in the original source. The comments in the original sources were scrapped using a Python⁵ script, and Facebook comments were collected with Facepager, an open-source academic software (Junger & Keyling, 2012/2018) that provides an interface to interact with Facebook’s Graph Application Programming Interface (API).⁶ While it is not possible to make inferences about the demographics of users in either of these sites,⁷ this approach aimed at keeping the topics of discussion constant. The sample of 157 stories led to a universe of 55,053 comments on Facebook and on news sites, with Facebook comments accounting for around 70% of this total (n = 38,594). For content analysis, a random stratified sample of comments⁸ was devised, accounting for the proportion between Facebook (70%) and news sources (30%), and the number of comments on each thread (e.g., to proportionally represent threads with 1000, 100, or 10 comments). To enable the observation of thread, consecutive messages were sampled instead of random comments, using a randomly generated number as a starting point. The final sample for manual content analysis had a total of 12,337 comments.

Content Analysis

This study employs systematic content analysis (Neuendorf, 2002), using a coding scheme that incorporates a mix of categories inspired by prior research (Coe et al., 2014; Stromer-Galley, 2007) and original ones. Messages were coded to detect several discourse features such as disagreement, opinion expression, incivility, and target (see Supplemental Appendix).

Opinion expression was coded in the following two subcategories: justified and non-justified opinions, with the former referring to comments in which there is an attempt to justify one’s position, and the latter referring to remarks that revealed a commenter’s take on a topic, without any explanation (Krippendorff’s α = .91 news comments and .74 Facebook).

Disagreement was coded based on the content and the context of the discussion, identified when a message diverged from the previous messages in a thread;⁹ or when users explicitly diverged from another person in form of mentions or replies (Krippendorff’s α = .89 news comments and .82 Facebook).

Reply: Messages were coded as a direct reply when they mentioned another commenter by name or used reply affordances in each platform, that is, comments in reply to comments and not to the original story or post (Krippendorff’s α = .96 news comments and .77 Facebook).

Incivility was classified based on discursive features signaling a heated, disrespectful tone in the discussion. The category includes different types of incivility (e.g., mockery, disdain, pejorative language, profanity, personal attacks focused on demeaning characteristics, personality, ideas, or arguments). Following other studies, messages were considered civil when they did not contain any type of incivility (Coe et al., 2014; Su et al., 2018), which does not mean they were respectful or polite (Krippendorff’s α = .87 news comments and .79 Facebook).

Uncivil messages were also coded by targets, initially classified as other users, political elites (parties, politicians, public officials, cabinet members, etc.), people or groups featured on the news, the media (including journalists), political minorities, and so on. These messages could also be unfocused when coders were unable to identify a clear target (Krippendorff’s α = .88 news comments and .84 Facebook). Interpersonal-directed and elite-directed incivility are based on two of these initial subcategories—other users in a discussion and political actors, binary-coded post hoc.

Content analysis was conducted by two independent coders, who performed an intercoder agreement test using approximately 5% (n = 636, proportionally distributed between 413 Facebook comments and 223 news comments)¹⁰ of the sample.

Results

Before answering research questions and testing hypotheses, I present descriptive results of the content analysis to provide an overview of the extent to which different targets were used, as well as the prevalence of uncivil discourse. Uncivil discourse was observed in 37.8% of the comments. As demonstrated in Figure 1, the distribution of targets of incivility—with percentages based on the subset of uncivil messages—shows that politicians are the main target of uncivil discourse in online comments in both platforms, and the differences between targets per platform are significant, χ²(5) = 264.91, p < .0001.

Figure 1.

Distribution of targets of incivility per platform (%).

To further understand how these behaviors are manifested online, I use two logistic regression models using the targets of interest as the dependent variables: interpersonal-directed incivility and elite-directed incivility (Table 1), which were recoded based on two of the original targets (other users and political elites). Because targets were only coded for messages classified as uncivil, these models are based on a subset of the data (N = 4,669). The models tested a set of discursive predictors, such as justified opinion expression, disagreement, and replies,¹¹ as well as the platform where the comment was published (Facebook = 1). The interaction between disagreement and replies was added to the model to investigate whether interpersonal dynamics change the effects of disagreement, addressing the first research question. The topics of the news stories were included as control variables.

Table 1.

Logistic Regression Models Predicting Targets of Incivility.

	Interpersonal incivility			Elite incivility
	B	SE	2.5%, 97.5%	B	SE	2.5%, 97.5%
(Intercept)	−2.97***	(0.31)	[−3.57, −2.37]	−1.96***	(0.24)	[−2.42, −1.49]
Reply	2.40***	(0.19)	[2.03, 2.77]	−0.65***	(0.14)	[−0.93, −0.38]
Justified opinions	−0.45***	(0.13)	[−0.71, −0.19]	0.21*	(0.08)	[0.05, 0.38]
Facebook	−0.06	(0.14)	[−0.33, 0.21]	0.12	(0.09)	[−0.06, 0.29]
Disagreement	3.54***	(0.24)	[3.06, 4.02]	−2.29***	(0.29)	[−2.85, −1.72]
Disagreement × reply	−1.14***	(0.30)	[−1.73, −0.56]	0.55	(0.33)	[−0.10, 1.20]
Topic: Formal politics	−0.75**	(0.29)	[−1.32, −0.18]	3.36***	(0.23)	[2.91, 3.82]
Topic: Minorities	−0.58	(0.30)	[−1.18, 0.01]	0.60*	(0.24)	[0.14, 1.07]
Topic: Policy	−1.32***	(0.39)	[−2.08, −0.55]	1.27***	(0.26)	[0.77, 1.77]
Topic: Civil society	0.04	(0.48)	[−0.90, 0.98]	1.82***	(0.34)	[1.16, 2.49]
AIC	1,958.70			4,618.95
Pseudo R²	.57			.44

AIC: akaike information criterion; SE: standard error.

p < .05. **p < .01. ***p < .001.

Comparing the two models, it is clear that interpersonal and elite incivility are associated with distinct deliberative attributes: while elite-directed incivility is positively associated with justified opinion expression and has a negative relationship with disagreement and with the use of direct replies, interpersonal-directed incivility is negatively associated with justified opinions and strongly and positively associated with disagreement, confirming the first and second hypotheses. When the interaction between replies and disagreements is taken into account, the relationship with interpersonal-directed incivility changes direction and becomes negative, while not significant for elite-directed incivility (Figure 2). These findings address the first research question, demonstrating that uncivil comments that represent an interaction between participants are different than comments that do not engage with others—in both cases, the association changed directions. Even though the interaction term was not significant in the model for elite-directed incivility, it is noteworthy that the coefficient is positive and approached significance (p = .09), suggesting that conversations between participants change the nature of uncivil discourse.

Figure 2.

Predicted probabilities of interaction terms.

The third hypothesis was focused on platform affordances and predicted that the public nature of Facebook, where participants use their real identities, would act as a constraint to interpersonal-directed expressions of incivility. It was hypothesized that comments on Facebook would be less likely to target other participants to the discussion when compared to news websites. The hypothesis was not confirmed, as platform was not a significant predictor of interpersonal-directed incivility even though the coefficient was in the expected direction. Finally, in response to the second research question, elite-directed incivility was slightly more likely to occur in Facebook comments—however, as the confidence interval spans zero, this result is not significant. Figure 3 presents a plot of the coefficients for both models to demonstrates that the contextual and discursive features of interpersonal-directed and elite-directed incivility are substantively different, suggesting that uncivil comments may emerge in distinct conversational contexts.

Figure 3.

Plotted coefficients (B) for targets of incivility.

Discussion

The goal of this study was to examine the conversational dynamics associated with online incivility directed at different targets, focusing on the distinction between elite-directed incivility—targeted at politicians, public officials, party leaders—and interpersonal-directed incivility, which targets other participants in a discussion. Identifying the contextual and discursive features associated with these behaviors is crucial to better understand the conditions in which online discussions might be toxic or offensive to individuals or are characterized by heated opinion expression, and the findings of this study provide evidence that the latter is more common. Considering that engagement with news stories is an important metric for algorithmic filtration, it is possible that people who are uncivil toward politicians in the comments are contributing to “sound the alarm” and expose others in their network to relevant political issues, behaving as monitorial citizens (Zaller, 2003).

When looking at specific targets, this study finds that incivility is primarily targeted at political elites, which is consistent with prior research suggesting that personal targets are less prevalent than other targets of incivility (Su et al., 2018). While interpersonal-directed incivility is not the most common behavior, it is still important to unveil the dynamics at play, as these might signal discussions that are toxic. As hypothesized, interpersonal-directed incivility is positively associated with disagreement (Stromer-Galley et al., 2015)—but negatively associated with direct replies in heterogeneous discussions. These findings suggest that interpersonal-directed incivility might be used with the intention of offending others, and such interpersonal attacks seem to undermine, instead of invite, further discussion. Interpersonal-directed incivility, outside of the context of direct mentions and replies, appears to signal a lack of interest in engaging in substantive debates with the other side, prioritizing personal attacks over justified arguments. However, when participants directly engage with others in the context of disagreement, the comment is less likely to have an interpersonal target. Taken together, these results suggest that while some conversations might become toxic, that is not always the case as those who meaningfully engage with others are significantly less likely to target them with uncivil rhetoric.

Prior work on perceptions of incivility can help interpret these results, as people generally perceive it as acceptable when participants in a discussion are uncivil on the grounds of their positions and arguments (Muddiman, 2017; Stryker et al., 2016). The finding that interpersonal-directed incivility is less common when commenters are directly responding to each other provides further evidence that substantive heterogeneous debates may occur in the comments section in spite of incivility. However, the dynamics around interpersonal-directed incivility also suggest that those who are targeted do not engage with such comments, indicating that interpersonal-targeted incivility may undermine the possibility of dialogic interaction.

As well as those who are more frequently engaged in political discussions online tend to be more uncivil (Hmielowski et al., 2014), continued participation in these debates might influence participants’ perception of interpersonal-directed incivility as an acceptable behavior online. The perception that the discussion environment is open to incivility may lead participants to use vulgar and violent language, conforming to perceived “norms” (Rösner & Krämer, 2016). There is evidence that people who are conflict-oriented find incivility enjoyable and entertaining, which might suggest that for those who participate in the discussions, incivility comes with the territory (Sydnor, 2019).

Still in the realm of interpersonal-directed incivility, the results challenged the hypothesis that the public nature of Facebook would act as a constraint to antinormative behaviors (Donath & Boyd, 2004; Suler, 2004). Contrary to the hypothesis that social connections and personal identities would prevent participants from being uncivil toward others in a platform where these behaviors are visible to others in the network, as compared to the more anonymous environment of news websites, the variable for platform was not a significant predictor of interpersonal targets. Even though prior research had pointed to anonymity as the culprit for online incivility (Maia & Rezende, 2016; Papacharissi, 2004; Rowe, 2015; Santana, 2014), this study provides further evidence that identification alone is not enough to prevent uncivil discourse, and that perceived social group norms may have a stronger impact than anonymity in facilitating uncivil expressions (Chen, 2017; Rösner & Krämer, 2016). Another possible explanation is that Facebook users may be interacting with others outside of their own networks when commenting on news stories, and therefore, might not feel as constrained by their social ties.

Finally, this study points to the political sphere as the main target of uncivil comments suggesting that those who discuss politics online often adopt a harsh critical tone toward elected officials. The result is not surprising in the context of polarized opinions and political distrust in Brazil. Moreover, the public might find incivility acceptable when targeted at politicians (Theocharis et al., 2016), or political arguments and opinions (Muddiman, 2017), and uncivil rhetoric may be seen as a strategic asset to outline an “enemy,” create a sense of community and a sentiment of “us” versus “them” (Berry & Sobieraj, 2016). Contrary to interpersonal-directed incivility, elite-directed incivility is likely to contain justified opinions, indicating that those making uncivil comments toward politicians are explaining their positions and potentially engaging in substantive discussions.

These patterns suggest that those who target politicians with antinormative intensity when discussing the news online might be exercising their roles as monitorial citizens, publicly criticizing the political sphere (Zaller, 2003). In this situation, incivility might be used as a rhetorical asset to call attention to one’s opinions and to signal dissatisfaction and outrage (Herbst, 2010; Mutz, 2016; Sobieraj & Berry, 2011). This monitorial activity may serve an important purpose, particularly on Facebook, as engagement in the comments section is one of the aspects that influence the algorithmic organization of news feeds (Anspach, 2017). However, it is important to highlight that not all politicians are targeted similarly, and this may have relevant consequences. While this study has not investigated whether elite-directed incivility systematically targeted politicians based on gender or minority status, it is relevant to note that the discrepancies in how these personal characteristics influence abuse can have detrimental democratic consequences (Southern & Harmer, 2021).

These findings have limitations. First, while this study helps understand online incivility by expanding research beyond the US context, its findings cannot be extrapolated to other countries. While it is true that political polarization is a global phenomenon (McCoy et al., 2018), the dynamics of how the public discusses political issues might also be affected by other factors, such as partisanship and political system. Second, this study has focused on identifying uncivil expressions and its different functions in discourse based on targets and other contextual characteristics, and the discussion of whether or not incivility is inherently problematic is based on these characteristics, not actual effects. Future research needs to investigate how different targets of incivility may yield different effects on participants and bystanders of online discussions (Searles et al., 2020). Finally, this study only analyzed textual elements, not considering other frequently used media sources on social media, such as images, graphics interchange formats (GIFs), or emoticons. Future research needs to systematically investigate the role of visual forms of communication in online political talk, more broadly, and in uncivil discourse.

Conclusion

Political talk online is often considered toxic due to the presence of incivility. However, uncivil rhetoric is not, in itself, an indicator that online discussions are merely a “shouting match.” By examining the targets of uncivil discourse, this study has sought to unveil how different contextual and discursive characteristics help explain the conditions in which elite-directed and interpersonal-directed incivility arise in comments on Facebook and news websites.

In interpreting the findings, it is relevant to remember that uncivil comments on political news online are not necessarily characterized by genuine discussions, and might simply indicate citizens’ reactions to a story, or their eagerness to express their opinions publicly (Reagle, 2016). In this vein, the use of uncivil rhetoric when talking about politicians may indicate that this behavior is perceived as acceptable by those commenting online (Hmielowski et al., 2014)—which does not come as a surprise when uncivil rhetoric is routinely adopted by both political actors and by the media (Herbst, 2010; Mutz, 2016; Sobieraj & Berry, 2011). It is important to recognize, however, that incivility toward political elites may have detrimental consequences for democracy, such as increasing polarization, dissatisfaction, and distrust toward politicians (Druckman et al., 2019), and more research is needed to examine the consequences of this type of rhetoric in how citizens perceive politicians, as well as to what extent such consequences may be affected by personal characteristics such as race and gender (Searles et al., 2020; Southern & Harmer, 2021).

While people may target others with uncivil comments when discussing political topics online, it may be the case that, differently than in face-to-face conversations, these personal attacks are not necessarily perceived as offensive by its recipients. Moreover, the finding that exchanges in reply threads are less likely to contain interpersonal-directed incivility is an indicator that meaningful and substantive debates may take place in the comments section. If one believes that democracies may benefit from heated political exchanges—even when those are uncivil (Papacharissi, 2004; Schudson, 1997)—online political talk cannot be dismissed because of the presence of uncivil remarks. Understanding both the contextual features around interpersonal-directed and elite-directed incivility, as well as the discursive characteristics around these behaviors, allows scholars to better theorize the role of incivility as a rhetorical asset in online political conversation instead of dismissing or questioning the potential value of online political talk.

Supplemental Material

sj-docx-1-sms-10.1177_20563051211008827 – Supplemental material for More Than Just Shouting? Distinguishing Interpersonal-Directed and Elite-Directed Incivility in Online Political Talk

Supplemental material, sj-docx-1-sms-10.1177_20563051211008827 for More Than Just Shouting? Distinguishing Interpersonal-Directed and Elite-Directed Incivility in Online Political Talk by Patrícia Rossini in Social Media + Society

Footnotes

Acknowledgements

The author would like to thank Professor Jennifer Stromer-Galley for her invaluable guidance and feedback on this project, the two anonymous reviewers for their thoughtful suggestions, and Thais Choucair for her help with content analysis. The author is responsible for any remaining issues.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported in part by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior—Brasil (CAPES)—Finance Code 001.

ORCID iD

Patrícia Rossini

Supplemental Material

Supplemental material for this article is available online.

Notes

Author Biography

Patrícia Rossini (PhD, Federal University of Minas Gerais) is a Derby Fellow in Communication and Media at the University of Liverpool, United Kingdom. Her research interests include online political talk, with emphasis on uncivil and intolerant discourse, misinformation, deliberation, and political campaigning.

References

Anderson

A. A.

Brossard

Scheufele

D. A.

Xenos

M. A.

Ladwig

(2014). The “nasty effect”: Online incivility and risk perceptions of emerging technologies: Crude comments and concern. Journal of Computer-Mediated Communication, 19(3), 373–387. https://doi.org/10.1111/jcc4.12009

Anspach

N. M.

(2017). The new personal influence: How our Facebook friends influence the news we read. Political Communication, 34(4), 1–17. https://doi.org/10.1080/10584609.2017.1316329

Barnidge

(2015). The role of news in promoting political disagreement on social media. Computers in Human Behavior, 52, 211–218. https://doi.org/10.1016/j.chb.2015.06.011

Berry

Sobieraj

(2016). The outrage industry: Political opinion media and the new incivility. Oxford University Press.

Black

L. W.

(2009). Listening to the city: Difference, identity, and storytelling in online deliberative groups. Journal of Public Deliberation, 5(1), 4.

Boczkowski

P. J.

Mitchelstein

Walter

(2012). When burglar alarms sound, do monitorial citizens pay attention to them? The online news choices of journalists and consumers during and after the 2008 U.S. election cycle. Political Communication, 29(4), 347–366. https://doi.org/10.1080/10584609.2012.722173

Borah

(2014). Does it matter where you read the news story? Interaction of incivility and news frames in the political blogosphere. Communication Research, 41(6), 809–827. https://doi.org/10.1177/0093650212449353

Chen

G. M.

(2017). Online incivility and public debate: Nasty talk (1st ed., 2017 ed.). Palgrave Macmillan.

Chen

G. M.

(2017). Online political discourse: Exploring differences in effects of civil and uncivil disagreement in news website comments. Journal of Broadcasting & Electronic Media, 61(1), 108–125. https://doi.org/10.1080/08838151.2016.1273922

10.

Chen

G. M.

Muddiman

Wilner

Pariser

Stroud

N. J.

(2019). We should not get rid of incivility online. Social Media + Society, 5(3), 2056305119862641. https://doi.org/10.1177/2056305119862641

11.

Chen

G. M.

Y. M. M.

(2016). Third-person perception of online comments: Civil ones persuade you more than me. Computers in Human Behavior, 55(Part B), 736–742. https://doi.org/10.1016/j.chb.2015.10.014

12.

Coe

Kenski

Rains

S. A.

(2014). Online and uncivil? Patterns and determinants of incivility in newspaper website comments. Journal of Communication, 64(4), 658–679. https://doi.org/10.1111/jcom.12104

13.

Coles

B. A.

West

(2016). Trolling the trolls: Online forum users constructions of the nature and properties of trolling. Computers in Human Behavior, 60, 233–244. https://doi.org/10.1016/j.chb.2016.02.070

14.

Connolly-Ahern

Ahern

L. A.

Bortree

D. S.

(2009). The effectiveness of stratified constructed week sampling for content analysis of electronic news source archives: AP newswire, business wire, and PR newswire. Journalism & Mass Communication Quarterly, 86(4), 862–883. https://doi.org/10.1177/107769900908600409

15.

Davies

Chandler

(2013). Online deliberation design: Choices, criteria, and evidence. arXiv Preprint. http://arxiv.org/abs/1302.5177

16.

Donath

Boyd

(2004). Public displays of connection. BT Technology Journal, 22(4), 71–82.

17.

Druckman

J. N.

Gubitz

S. R.

Lloyd

A. M.

Levendusky

M. S.

(2019). How incivility on partisan media (de)polarizes the electorate. Journal of Politics, 81(1), 291–295. https://doi.org/10.1086/699912

18.

Eveland

W. P.

Hively

M. H.

(2009). Political discussion frequency, network size, and “heterogeneity” of discussion as predictors of political knowledge and participation. Journal of Communication, 59(2), 205–224. https://doi.org/10.1111/j.1460-2466.2009.01412.x

19.

Facebook. (2016, April 19). 102 milhões de brasileiros compartilham seus momentos no Facebook todos os meses [102 million Brazilians share their moments on Facebook every month]. Facebook for Business. https://pt-br.facebook.com/business/news/102-milhes-de-brasileiros-compartilham-seus-momentos-no-facebook-todos-os-meses

20.

Fletcher

Nielsen

R. K.

(2018). Are people incidentally exposed to news on social media? A comparative analysis. New Media & Society, 20(7), 2450–2468. https://doi.org/10.1177/1461444817724170

21.

Freelon

D. G.

(2010). Analyzing online political discussion using three models of democratic communication. New Media & Society, 12(7), 1172–1190. https://doi.org/10.1177/1461444809357927

22.

Friess

Eilders

(2015). A systematic review of online deliberation research. Policy & Internet, 7(3), 319–339.

23.

Gervais

B. T.

(2015). Incivility online: Affective and behavioral reactions to uncivil political posts in a web-based experiment. Journal of Information Technology & Politics, 12(2), 167–185. https://doi.org/10.1080/19331681.2014.997416

24.

Gil de Zúñiga

Molyneux

Zheng

(2014). Social media, political expression, and political participation: Panel analysis of lagged and concurrent relationships. Journal of Communication, 64(4), 612–634. https://doi.org/10.1111/jcom.12103

25.

Habermas

(1996). Between facts and norms: Contributions to a discourse theory of law and democracy. MIT Press.

26.

Halpern

Gibbs

(2013). Social media as a catalyst for online deliberation? Exploring the affordances of Facebook and YouTube for political expression. Computers in Human Behavior, 29(3), 1159–1168. https://doi.org/10.1016/j.chb.2012.10.008

27.

Herbst

(2010). Rude democracy: Civility and incivility in American politics. Temple University Press.

28.

Hmielowski

J. D.

Hutchens

M. J.

Cicchirillo

V. J.

(2014). Living in an age of online incivility: Examining the conditional indirect effects of online discussion on political flaming. Information, Communication & Society, 17(10), 1196–1211. https://doi.org/10.1080/1369118X.2014.899609

29.

Jamieson

K. H.

Volinsky

Weitz

Kenski

(2015). The political uses and abuses of civility and incivility. In Kenski

Jamieson

K. H.

(Eds.), The Oxford handbook of political communication (1st ed.). Oxford University Press. http://www.oxfordhandbooks.com/view/10.1093/oxfordhb/9780199793471.001.0001/oxfordhb-9780199793471

30.

Janssen

Kies

(2005). Online forums and deliberative democracy. Acta Política, 40(3), 317–335.

31.

Johnson

G. F.

Black

L. W.

Knobloch

K. R.

(2017). Citizens’ initiative review process: Mediating emotions, promoting productive deliberation. Policy & Politics, 45(3), 431–447. https://doi.org/10.1332/030557316X14595273846060

32.

Junger

Keyling

(2018). Facepager was made for fetching public available data from Facebook, Twitter and other JSON-based APIs [Python]. https://github.com/strohne/Facepager/. (Original work published 2012)

33.

Kenski

Coe

Rains

S. A.

(2020). Perceptions of uncivil discourse online: An examination of types and predictors. Communication Research, 47(6), 795–814. https://doi.org/10.1177/0093650217699933

34.

Kim

Wyatt

R. O.

Katz

(1999). News, talk, opinion, participation: The part played by conversation in deliberative democracy. Political Communication, 16(4), 361–385. https://doi.org/10.1080/105846099198541

35.

Maia

R. C. M.

Rezende

T. A. S.

(2016). Respect and disrespect in deliberation across the networked media environment: Examining multiple paths of political talk. Journal of Computer-Mediated Communication, 21(2), 121–139. https://doi.org/10.1111/jcc4.12155

36.

McCoy

Rahman

Somer

(2018). Polarization and the global crisis of democracy: Common patterns, dynamics, and pernicious consequences for democratic polities. American Behavioral Scientist, 62(1), 16–42. https://doi.org/10.1177/0002764218759576

37.

Meltzer

(2015). Journalistic concern about uncivil political talk in digital news media: Responsibility, credibility, and academic influence. The International Journal of Press/politics, 20(1), 85–107. https://doi.org/10.1177/1940161214558748

38.

Miller

P. R.

Bobkowski

P. S.

Maliniak

Rapoport

R. B.

(2015). Talking politics on Facebook: Network centrality and political discussion practices in social media. Political Research Quarterly, 68(2), 377–391. https://doi.org/10.1177/1065912915580135

39.

Moy

Gastil

(2006). Predicting deliberative conversation: The impact of discussion networks, media use, and political cognitions. Political Communication, 23(4), 443–460. https://doi.org/10.1080/10584600600977003

40.

Muddiman

(2017). Personal and public levels of political incivility. International Journal of Communication, 11, 21.

41.

Mutz

D. C.

(2006). Hearing the other side: Deliberative versus participatory democracy. Cambridge University Press.

42.

Mutz

D. C.

(2016). In-your-face politics: The consequences of uncivil media. Princeton University Press.

43.

Neuendorf

K. A.

(2002). The content analysis guidebook (1st ed.). SAGE.

44.

Newman

Fletcher

Schulz

Andi

Nielsen

R. K.

(2020). Reuters Institute Digital News Report (p. 112). Reuters Institute for the Study of Journalism. https://www.digitalnewsreport.org/

45.

Núcleo de Informação e Coordenação do Ponto BR. (2020). ICT households (p. 344). Comitê Gestor da Internet no Brasil. https://www.cetic.br/pt/pesquisa/domicilios/publicacoes/

46.

Zheng

Chen

G. M.

(2018). Twitter versus Facebook: Comparing incivility, impoliteness, and deliberative attributes. New Media & Society, 20(9), 3400–3419. https://doi.org/10.1177/1461444817749516

47.

Papacharissi

(2004). Democracy online: Civility, politeness, and the democratic potential of online political discussion groups. New Media & Society, 6(2), 259–283. https://doi.org/10.1177/1461444804041444

48.

Parkinson

D. J.

Mansbridge

P. J.

(Eds.). (2012). Deliberative systems. Cambridge University Press.

49.

Phillips

(2015). This is why we can’t have nice things: Mapping the relationship between online trolling and mainstream culture. MIT Press.

50.

Reagle

J. M.

Jr. (2016). Reading the comments: Likers, haters, and manipulators at the bottom of the web (Reprint ed.). MIT Press.

51.

Riffe

Lacy

Fico

(2005). Analyzing media messages: Using quantitative content analysis in research (2nd ed.). Routledge.

52.

Rösner

Krämer

N. C.

(2016). Verbal venting in the social web: Effects of anonymity and group norms on aggressive language use in online comments. Social Media + Society, 2(3), 205630511666422. https://doi.org/10.1177/2056305116664220

53.

Rossini

(2019). Disentangling uncivil and intolerant discourse. In Boatright

Shaffer

Sobieraj

Young

D. G.

(Eds.), A crisis of civility? Contemporary research on civility, incivility, and political discourse (pp. 142–157). Routledge.

54.

Rossini

(2020). Beyond incivility: Understanding patterns of uncivil and intolerant discourse in online political talk. Communication Research. Advance online publication. https://doi.org/10.1177/0093650220921314

55.

Rossini

Stromer-Galley

(2019). Citizen deliberation online. In Suhay

Grofman

Tresel

(Eds.), Oxford handbook of electoral persuasion (pp. 690–712). Oxford University Press.

56.

Rossini

Stromer-Galley

Zhang

(2021). Exploring the relationship between campaign discourse on Facebook and the public’s comments: A case study of incivility during the 2016 US presidential election. Political Studies, 69(1), 89–107. https://doi.org/10.1177/0032321719890818

57.

Rowe

(2015). Civility 2.0: A comparative analysis of incivility in online political discussion. Information, Communication & Society, 18(2), 121–138. https://doi.org/10.1080/1369118X.2014.940365

58.

Santana

A. D.

(2014). Virtuous or vitriolic: The effect of anonymity on civility in online newspaper reader comment boards. Journalism Practice, 8(1), 18–33. https://doi.org/10.1080/17512786.2013.813194

59.

Schudson

(1997). Why conversation is not the soul of democracy. Critical Studies in Mass Communication, 14(4), 297–309. https://doi.org/10.1080/15295039709367020

60.

Searles

Spencer

Duru

(2020). Don’t read the comments: The effects of abusive comments on perceptions of women authors’ credibility. Information, Communication & Society, 23(7), 947–962. https://doi.org/10.1080/1369118X.2018.1534985

61.

Sobieraj

Berry

J. M.

(2011). From incivility to outrage: Political discourse in blogs, talk radio, and cable news. Political Communication, 28(1), 19–41. https://doi.org/10.1080/10584609.2010.542360

62.

Southern

Harmer

(2021). Twitter, incivility and “everyday” gendered othering: An analysis of tweets sent to UK members of parliament. Social Science Computer Review, 39(2), 259–275. https://doi.org/10.1177/0894439319865519

63.

Stromer-Galley

(2007). Measuring deliberation’s content: A coding scheme. Journal of Public Deliberation, 3(1), 12.

64.

Stromer-Galley

Bryant

Bimber

(2015). Context and medium matter: Expressing disagreements online and face-to-face in political deliberations. Journal of Public Deliberation, 11(1), 1.

65.

Stromer-Galley

Wichowski

(2011). Political discussion online. In Consalvo

Ess

(Eds.), The handbook of internet studies (Vol. 11, pp. 168–187). Wiley-Blackwell.

66.

Stryker

Conway

B. A.

Danielson

J. T.

(2016). What is political incivility? Communication Monographs, 83(4), 535–556. https://doi.org/10.1080/03637751.2016.1201207

67.

Y.-F.

Xenos

M. A.

Rose

K. M.

Wirz

Scheufele

D. A.

Brossard

(2018). Uncivil and personal? Comparing patterns of incivility in comments on the Facebook pages of news outlets. New Media & Society, 20(10), 3678–3699. https://doi.org/10.1177/1461444818757205

68.

Suler

(2004). The online disinhibition effect. Cyberpsy-chology & Behavior, 7(3), 321–326. https://doi.org/10.1089/1094931041291295

69.

Sydnor

(2018). Platforms for incivility: Examining perceptions across different media formats. Political Communication, 35(1), 97–116. https://doi.org/10.1080/10584609.2017.1355857

70.

Sydnor

(2019). Disrespectful democracy: The psychology of political incivility. Columbia University Press.

71.

Theocharis

Barberá

Fazekas

Popa

S. A.

Parnet

(2016). A bad workman blames his tweets: The consequences of citizens’ uncivil Twitter use when interacting with party candidates: Incivility in interactions with candidates on Twitter. Journal of Communication, 66(6), 1007–1031. https://doi.org/10.1111/jcom.12259

72.

Vaccari

Valeriani

Barberá

Jost

J. T.

Nagler

Tucker

J. A.

(2016). Of echo chambers and contrarian clubs: Exposure to political disagreement among German and Italian users of Twitter. Social Media + Society, 2(3), 2056305116664221.

73.

Vitak

Chadha

Steiner

Ashktorab

(2017). Identifying women’s experiences with and strategies for mitigating negative effects of online harassment. In Proceedings of the 2017 ACM Conference on Computer Supported Cooperative Work and Social Computing (pp. 1231–1245). Association for Computing Machinery. https://doi.org/10.1145/2998181.2998337

74.

Wright

(2012). From “third place” to “third space”: Everyday political talk in non-political online spaces. Javnost—The Public, 19(3), 5–20. http://javnost-thepublic.org/article/2012/3/1/

75.

Wyatt

R. O.

Katz

Kim

(2000). Bridging the spheres: Political and personal conversation in public and private spaces. Journal of Communication, 50(1), 71–92.

76.

Young

A. M.

Battaglia

Cloud

D. L.

(2010). (UN)disciplining the scholar activist: Policing the boundaries of political engagement. Quarterly Journal of Speech, 96(4), 427–435. https://doi.org/10.1080/00335630.2010.521179

77.

Ytre-Arne

Moe

(2018). Approximately informed, occasionally monitorial? Reconsidering normative citizen ideals. The International Journal of Press/Politics, 23(2), 227–246. https://doi.org/10.1177/1940161218771903

78.

Zaller

(2003). A new standard of news quality: Burglar alarms for the monitorial citizen. Political Communication, 20(2), 109–130. https://doi.org/10.1080/10584600390211136

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.03 MB