More Than Virality: Online Sharing of Controversial News With Activated Audience

Abstract

In contrast to earlier studies that focus on virality, we examine how individuals share controversial news online by accounting for audience features. The results from surveys of 400 individuals suggest that they are more likely to share highly controversial news (a) on open, asymmetrical social media or (b) with communication partners with a low frequency rather than a high frequency, but are more likely to share moderately controversial news (c) on closed, symmetrical social media. This study implies that interpersonal motives determine the online path of controversial news and that media environments play unique roles in moderating and spreading controversies in society.

Keywords

online news sharing news controversiality imagined audience audience feature

With the grand transition from “push” media to “pull” media, followed by the convergence into social media as an integrated communication, user sharing drives the consumption and spread of news. As individuals take active roles in mass spread of online news to their audience, online news sharing integrates mass and interpersonal communication processes. Understanding online news sharing is imperative for the mass media industries as well as communication scholars who aim to understand online public attention dynamics and develop a general theory of human communication across mass and interpersonal communication (Flanagin, 2017; Neuman, 2016; Walther & Valkenburg, 2017).

Many studies have explained online news sharing behavior as a news diffusion process, taking an informational perspective and focusing on the types of news content gaining the greatest dissemination online (e.g., negative, useful, controversial, credible, novel, and/or interesting content; Bobkowski, 2015; Cappella, Kim, & Albarracín, 2015; Chen & Berger, 2013; Choi & Toma, 2014; Ma, Lee, & Goh, 2014; Trilling, Tolochko, & Burscher, 2017; Vosoughi, Roy, & Aral, 2018). However, the content itself never entirely determines whether people will share it. Sharing news is an interaction between the sharers and their audience, and this implies that individuals will seek to present themselves in ways that match their ideal self-image and please their target audience (Baumeister & Hutton, 1987; Berger & Schwartz, 2011; Goffman, 1959). As such, in addition to news diffusion, news sharing is also an interpersonal behavior that includes consideration of the self and others.

The purpose of this study was to examine how individuals consider their audience’s features in different media environments and share news online for their interpersonal motives. Specifically, this article focuses on one characteristic of news content: news controversiality. Controversial news can be an interesting and important subject because diffusion of and debate on controversial issues are closely related to how individuals want to present themselves to others, as well as how society manages conflicts and solves social problems, which is an essential function of media.

This article first explains how sharing controversial news online may be an act of relational communication. Next, the article conceptualizes “activated audience” and introduces three features of such audience in different media environments that individuals may take into account when sharing controversial news online. We focus on how different media environments activate audiences with different characteristics and how individuals share controversial news differently with the different audiences. In other words, the technological features of media are not the focus of this study; different audience characteristics activated from each media environment are. Drawing from online surveys of online news sharers, this article examines how individuals communicate their public self-images by considering news controversiality and their audience’s features when they engage in news-sharing activities.

The present study makes three contributions to research on online news sharing. First, it interprets the online behavior of sharing controversial news as a relational communication and combines journalism and communication research. Second, this study increases understanding of the management of controversies in different media environments. Finally, this study develops the concept of imagined audience and provides future theoretical directions.

Sharing Controversial News

Individuals, as communicators, “perform” for an audience in everyday social interactions (Baumeister & Hutton, 1987; Berger & Schwartz, 2011; Goffman, 1959). When they share news with others, they consider how they want to present themselves to their audiences through their selectively shared content (Ihm & Kim, 2018). Choosing what to read can be a private decision, but choosing what to share is a public performance that involves one’s audience.

The advent of advanced media venues has provided opportunities for more sophisticated control of self-presentation in news sharing, such as refining and editing the content before sharing it (Walther, 1996, 2011). People can take advantage of online news-sharing opportunities to maintain an ideal self-image and please their audience (Hogg, Terry, & White, 1995; Turner & Onorato, 1999). As such, individuals with a high self-presentation motivation are more likely to share news online as an impression and relationship management strategy (E. M. Kim, Ihm, & Park, 2017). In this light, online news sharing is not only an act of information diffusion, but it is also a strategic act of conveying the self (e.g., one’s identities, thoughts, feelings, and opinions) to others to enhance selfhood and maintain relationships.

Controversial news may play a particular role in this conveyance. The term refers to issues that generate continued conversation and debate as well as intense public argument and disagreement often pertinent to social divisions such as race, gender, or class. On one hand, controversiality makes news more engaging and increases the audience’s interest; the mere fact that many people are paying attention to a controversial news stimulates further interest regardless of the nature of the content. On the other hand, engagement, thus instigated, has a potential to ignite and even enlarge social conflict.

Focusing on controversiality in news is meaningful for this study in two ways. First, talking about controversial news often reveals individuals’ stance regarding the issue, which provides a useful cue to understanding and categorizing them. Therefore, sharing controversial news is closely associated with interpersonal and self-presentational motives.

Second, how a community shares and discusses controversial news, compared with noncontroversial issues, is directly related to what it can achieve in terms of deliberative democracy. Diffusion of controversial news tends to trigger public debate, which can, at least in the short term, exacerbate social division by making the polarization of opposing views more visible. However, in the long run, compromise and consensus cannot be made without debates. As such, the ways in which individuals share controversial news are deeply related to how social problems and conflicts get dissipated or solved. Controversiality is not only a feature of news content that is amenable to broad and quick diffusion (Berger & Milkman, 2012; Cappella et al., 2015). Rather, the act of sharing controversial news forms the core process of deliberative democracy and political conversation (J. Kim, Wyatt, & Katz, 1999).

News sharers have to be careful about controversial news because it may both benefit and risk their self-image more than noncontroversial news. Decisions on what news to share and with whom to share it generally involve assessing the social risks and benefits, as shared news may present the news sharers’ opinions and identities to others (E. M. Kim et al., 2017). Sharing controversial news may pose social risks of disclosing positions, identities, or experiences the sharer would otherwise keep private and projecting a radical image to their audience; it may discomfort those who have opposing views and risk one’s self-image (Baumeister & Hutton, 1987). Sharing controversial news may also generate the social benefits of gaining high attention and response, proving the high social status of knowing interesting information (Vosoughi et al., 2018) and pleasing the audience’s interest and curiosity (Baumeister & Hutton, 1987).

Chen and Berger (2013) tested the impact of such social benefits and risks when sharing controversial news. In their study, participants were generally more likely to share news when its controversiality increased. It stimulated the interest of the audience but did not discomfort it. However, when the level of controversiality increased to reach an extreme level, the intention to share the news fell because sharing became socially burdensome. This result suggests that the most important factor behind the decision to share controversial news is the news sharers’ consciousness of others’ evaluation and control of self-presentation.

While controversiality is one of the features of news content, publicly presenting an individual’s opinion on and attention to a controversial issue via news sharing becomes a sophisticated interpersonal process of self-presenting to the audience. The next section explains this interpersonal process of news sharing by exploring how individuals consider and manage their audience when sharing controversial news.

From Imagined Audience to Activated Audience

Both social benefit and risk of sharing controversial news depend on audience. Individuals conceptualize whom their “imagined audience” will be during online communication and present themselves in different ways to match the audience’s expectations and maintain their authenticity (Litt, 2012; Litt & Hargittai, 2016; Marwick & boyd, 2011). Previous studies have described that these imagined audiences are “public,” “broad audience,” “those with similar interests,” “personal ties,” “communal ties,” or “professional ties” (Cook & Teasley, 2011; Litt & Hargittai, 2016; Marwick & boyd, 2011).

However, the imagined audience does not come purely from individuals’ imaginations. If it were a pure concoction of one’s imagination, the imagined audience would have remained an intrapersonal process. In the actual context of news sharing, media features (e.g., technological properties and cultural contexts) may set the boundary and conditions when imagining the audience (Ihm & Kim, 2018; Litt, 2012). For instance, Twitter users may imagine their audience only in an abstract sense because of the technological (e.g., no sophisticated functions for audience separation or management) and cultural (e.g., networked users retweeting postings actively to broader audience) features specific to the medium.

In other words, the variation in technological features and cultural contexts of different media may have different influences on how individuals imagine their audience. Every medium forms a unique universe of communication, and the specific features of media may influence and “activate” a specific audience group with whom individuals share news. This audience group is not superficial or only “imagined,” because specific features of media influence it. Therefore, this study proposes “activated audience” as an extended concept of imagined audience specified by the influences from the media features. Using this concept, this study examines how media features may activate audiences differently and how individuals may account for the activated audience and share controversial news with it.

Activated Audience and News Sharing

This study first explains how individuals may share controversial news differently with different audiences activated in two social media environments: (a) closed and symmetrical social media (CSS) and (b) open and asymmetrical social media (OAS). Research has classified social media into different categories according to the nature of each media, such as the degree of openness regarding who can join the communication (Choi & Lee, 2017) or the degree of symmetry in relationships (Kim & Lee, 2016). This article uses these two aspects of social media to differentiate two types of audiences activated from different social media environments.

First, CSS is one of the most prevalent and popular communication vehicles in the contemporary media environment. As of 2017, 1.82 billion people globally were using CSS (“Most Popular Mobile Messaging Apps,” 2017). In Korea, more than 30 million people—almost three-fifths of Koreans—use KakaoTalk, 4.45 million use Facebook Messenger, and 0.72 million use WeChat (“KakaoTalk,” 2016).

CSS allows communication only among invited users, usually in chat rooms created by one of the users. CSS activates audiences with specific context and characteristics because it allows users to control their audiences in a sophisticated way by inviting them separately into different chat rooms (Waterloo, Baumgartner, Peter, & Valkenburg, 2018). Many CSS users keep multiple chat rooms for various groups of different sizes and contexts, such as family, colleagues from work or project teams, or, say, middle school alumni (E. M. Kim et al., 2017). Everyone has the same opportunity to be the speaker and the audience in CSS chat rooms. In other words, communication on CSS is symmetrical, resembling everyday, face-to-face conversation.

Second, OAS allows for more open and asymmetrical communication than CSS. When a person addresses oneself in one’s OAS, anyone who has relationships with the person (e.g., “followers” on Twitter or “friends” on Facebook) can see the communication. The audience does not need to be invited to a specific chat room because the communication is open to them. As such, the media environment in OAS activates audiences with mixed contexts and characteristics. Audience separation and management in OAS are not as easy as in CSS (Baym & boyd, 2012). OAS users typically interact with both their personal friends and the general public at the same time in the open communication environment; the specific context of each social group (e.g., friends or acquaintances) gets blurred or disappears altogether (i.e., context collapse; Marwick & boyd, 2011). These mixed characteristics from diverse contexts create an “invisible audience” that is hard to assess or control (Litt, 2012). Communication on OAS tends to be more asymmetrical than on CSS. Communication on a person’s OAS page consists of the person addressing the broader audience, rather than a conversation occurring among the person and the audience. On a person’s OAS page, the person is the main speaker.

Due to the different technological features and contexts of CSS and OAS, users may share controversial news differently between the two media. For instance, CSS users may be more cautious about extremely controversial news than OAS users. CSS users generally use the medium for private conversations and relationship management (E. M. Kim et al., 2017; Waterloo et al., 2018) because its technological features allow sophisticated separation and management of different audience groups. Furthermore, individuals are better aware of whom their audiences are and can receive responses and feedback more easily and directly from the audience than can OAS users (Choi & Lee, 2017; Kim & Lee, 2016). As such, CSS users may be more concerned than OAS users with maintaining their relationship with the specific audience and not discomforting them.

Chen and Berger (2013) found that individuals are, by instinct, attracted to controversial news, so they share news more frequently when its controversiality increases. However, their news sharing decreases as the level of controversiality increases further, to the point when the risk of discomforting their audience and ruining their own reputation overrides the benefit of interesting their audience. In the same way, CSS users may also increase their sharing of controversial news as it can benefit them to satisfy the interest of their audience. When the controversiality reaches an extreme level, the risk of discomforting their audience may increase and CSS users may become reluctant and decrease sharing such news to maintain their relationship with their audience in the closed, symmetrical communication environment; it becomes socially burdensome in CSS where the users can directly receive responses from specific audiences of whom they are well aware. Therefore, this article hypothesizes the following:

H1: CSS users’ intention to share news will increase when the controversiality of news increases up to a moderate level, but will decrease when the controversiality reaches extremely high values; there will be an inverted U-shaped relationship between individuals’ intention to share news on CSS and the controversiality of news.

Because of the openness and the asymmetry in communication on OAS (Choi & Lee, 2017; Kim & Lee, 2016), OAS users often take advantage of the medium to diffuse information to a broader audience, including those outside their direct networks (Marwick & boyd, 2011). The invisible, context-collapsed audience on OAS (Litt, 2012; Marwick & boyd, 2011) does not enable sophisticated relationship management as on CSS (E. M. Kim et al., 2017), and thus encourages OAS users to focus more on broader diffusion and influence in the online environment. News consumption can enhance audience members’ knowledge and influence their behaviors (Gil de Zúñiga, Jung, & Valenzuela, 2012), and more interesting news gains more attention, recognition, and influence (Chen & Berger, 2013).

As such, the benefit of interesting their audience to have a greater influence on the online environment may override the risk of discomforting their audience on OAS (Chen & Berger, 2013). In other words, for OAS users, the risk of discomforting the “invisible” audience may not matter as much as the benefit of gaining more attention, influence, and recognition in the online environment, considering that gaining recognition and increasing online influence are among the chief motivators in online communication (Shirky, 2010). Therefore, this article hypothesizes the following:

H2: OAS users’ intention to share news will increase when the controversiality of news increases; there will be a positive relationship between individuals’ intention to share news on OAS and the controversiality of the news.

As each medium provides a condition where the activation of audience takes place, individuals respond to specific features of the audience that may affect their interpersonally motivated decision to share controversial news. For instance, individuals consider the size of their audience and their previous relationships with members of the audience when they decide to share news with them (E. M. Kim et al., 2017; Ma et al., 2014). Whereas previous studies on imagined audience have focused on how individuals describe their audience features (e.g., public, those with similar interests, personal ties, and professional ties; Cook & Teasley, 2011; Litt & Hargittai, 2016; Marwick & boyd, 2011), this study pays attention to how individuals react to and take account of audience features. This attempt may explain the more essential mechanism of the relational communication of sharing controversial news and provide further theoretical directions for the online communication process. As such, this study examines three audience features (i.e., size, feeling of closeness, and frequency of communication) that individuals may take into account when sharing controversial news.

Size

Individuals may have different numbers of participants in their CSS chat rooms or followers in their OAS. The size of the audience has been a negative predictor of individuals’ intention to share news online (E. M. Kim et al., 2017; Ma et al., 2014). Individuals censor themselves more when they perceive a larger online audience (Das & Kramer, 2013) because online news sharing stems from an individual’s motivation to maintain an ideal self-image and please the audience (Baumeister & Hutton, 1987; Hogg et al., 1995; Turner & Onorato, 1999). Individuals become more cautious, and even reluctant, to share news online with a larger audience because of the difficulties of pleasing a greater number of people and meeting their diverse expectations (E. M. Kim et al., 2017).

However, individuals may take account of the audience size differently when they share controversial news. While the advantages of a smaller audience—fewer difficulties pleasing it, smaller risk of ruining their self-images (E. M. Kim et al., 2017; Ma et al., 2014)—still apply, they may be more eager to share controversial news with a larger audience to spread the news more broadly for more influence (Ihm & Kim, 2018). The inconsistencies in previous studies suggest a need for more research on this topic. Therefore, this article asks,

RQ1: Is there an interaction effect between audience size and news controversiality on news sharing intention in (a) CSS and (b) OAS?

Strength of ties

Granovetter (1973) explains that “the strength of a tie is a combination of the amount of time, the emotional intensity, the intimacy, and the reciprocal services which characterize the tie” (p. 1361). Marsden and Campbell (1984) also distinguish “two distinct aspects of tie strength, having to do with the time spent in a relationship, and the depth of the relationship” (p. 482). Therefore, this study focuses on both aspects to examine whether the strength of ties with the audience may affect individuals’ online behavior in sharing controversial news.

Most behaviors of sharing information happen with people with whom individuals have strong ties (J. Brown & Reingen, 1987), and sharing news is also more likely to happen between individuals with strong ties (Chu, 2009; E. M. Kim et al., 2017; Ma et al., 2014). The behavior of sharing news may be an active strategy to maintain their close relationships on which individuals place more value. However, individuals may also become more cautious about what they share within their close relationships. Individuals strive to maintain and improve a positive self-image within their close relationships because they care more about impression management with this audience (B. R. Brown & Garland, 1971; Stephen & Lehmann, 2009). Sharing controversial news online may lead to revealing differences between the news sharer and the audience. Individuals tend to strengthen relationships when they share similarities in their interests, attitudes, or experiences (Newcomb, 1961), so exposing such differences poses a threat to close relationships. As such, individuals’ intention to share news online within these close relationships may decrease when news controversiality increases. These studies suggest that how news controversiality and strength of ties among conversation partners interact with each other is inconclusive. Therefore, we ask,

RQ2: Is there an interaction effect between the strength of ties and news controversiality on news-sharing intention in (a) CSS and (b) OAS?

Method

Sample

We created a survey for this study and conducted a preliminary study in May 2016 to examine its validity. This study involved 10 graduate students in the communication department at a university in Seoul, South Korea. The participants’ comments guided us to revise some confusing phrases.

The sample for the actual study was obtained from general panel data from Insight Korea, a professional research and consulting firm. In May 2016, 400 individuals who answered the qualifying question of whether they had ever shared news (e.g., daily news articles, online news, news on portal websites, news relevant to their interests and life) via both CSS and OAS participated in the online survey. The age range was limited to 20 to 49 because those in this age group have similar media habits in the online sphere (Korea Information Society Development Institute, 2017). In total, 400 people were sampled, equally divided between the genders (i.e., male, female) and three age ranges (i.e., 20s, 30s, and 40s). Only Koreans who live in the Seoul area were chosen for the sample to control for the influence of environmental and social surroundings.

Procedure

We first asked the participants about their uses of and audiences on CSS. The participants were asked to list as many as 10 group chat rooms in which they had been active over the past month on the CSS they use most frequently. A group chat room was operationalized in the survey as “a chat room having at least three participants including yourself.” The participants were asked to give nicknames to each group chat room (e.g., colleagues, friends from university, elementary, alumni). Of the 10 group chat rooms, the participants were asked to choose one room in which they had the most frequent communication. The participants stated the audience features of the room (i.e., frequency of communication, size of audience, and feeling of closeness with the audience group). Following the questions about CSS were questions about the participants’ uses of and audiences on OAS. The participants were asked to choose one OAS that they had used most frequently within the past month and to provide its audience features.

The questions about CSS and OAS were followed by questions about the participants’ intention to share news articles depending on the news content. First, the participants were asked to choose the subject that they read about with the most interest from among six categories (i.e., politics, economics, society, lifestyle, world, and IT/science). This was based on the fact that interest in a news subject may increase the intention to share news (Berger & Schwartz, 2011). One article from the chosen subject, which we selected carefully from a web news portal, Naver, was shown to the participants (see Appendices A and B for the full procedure of selecting the articles and the full list of articles). When the news article was shown on the screen, the participants stated how much they were willing to share it with the chat room and the OAS with which they mentioned having the most frequent communication. The survey was designed to allow the participants to go to the next page only after at least 5 s to make sure that the participants had actually read the articles.

Measures

Audience features

The participants were asked to indicate the size of and their strength of ties with the audience they share news with. Size was measured by the number of audience members in the chat room and on the OAS with which they had the most frequent communication, respectively. For strength of ties, the survey first measured the frequency of communication with the audience to measure the quantitative aspect of the strength of ties. Frequency of communication was operationalized as the number of times per month that participants conversed with more than two participants in the chat room on CSS or the number of times per month that participants uploaded posts on OAS.

The survey also measured the feeling of closeness with the audience in the chat room and on the OAS to capture the depth and emotional intimacy of the relationship (Granovetter, 1973; Marsden & Campbell, 1984). Previous studies have measured relationship closeness by asking whether the person is (a) an acquaintance, (b) a good friend, or (c) a very close friend (Marsden & Campbell, 1984) or whether “we have a close relationship” and “I would categorize him/her as one of my strong ties” (Lin & Utz, 2015). These measures have not captured the actual “emotional” intensity of the ties and are based on relationships between individuals. To account for the “emotional intensity” in contrast to frequency of communication, and this study’s context of asking about relationships with the audience group as a whole instead of each individual, we adapted seven 5-point Likert-type scales from previous studies on attachment feelings toward a group (France, Finney, & Swerdzewski, 2010; e.g., “I feel good when I talk with the participants in this chat room/this OAS” and “I feel close to the people in this chat room/this OAS”) and on conventional measures of relationship closeness (Lin & Utz, 2015). These measures (α = .90) evaluated the strength of the participants’ intimacy with the audience over the past year from 1 (highly disagree) to 5 (highly agree).

Perceived controversiality of news content

This study measured the perceived controversiality of news content. When the news article in the chosen category was shown to the participants, the participants indicated how controversial they thought the news article was from 1 (not at all) to 5 (very controversial).

Control variables

Because sociodemographic variables can influence behaviors on media (Gil de Zúñiga et al., 2012), gender, age, education, and monthly household income were controlled in this study (see Table 1 for demographics). Self-presentation motivation can also affect the intention of online news sharing (Ihm & Kim, 2018), so we controlled the participants’ self-presentation motivation as well. The survey participants answered nine 5-point Likert-type scales based on the combined measures of Lee (2013) and Govern and Marsch (2001) about their motivation of self-presentation. Lee’s measures evaluate two aspects of self-presentation motivation based on Baumeister (1982) and Baumeister and Hutton (1987): (a) audience-pleasing and (b) self-constructive motivation. Govern and Marsch’s (2001) criteria examine an audience’s degree of self-awareness or the consciousness of their self-image (α = .84, M = 3.61, SD = 0.53).

Table 1.

Demographic and Socioeconomic Characteristics of the Participants.

	Measure	M	SD
Age	Age	34.32	8.00
Gender	1: Male	50% each
Gender	0: Female	50% each
Education^a	0: Less than undergraduate degree	0.82	0.38
Education^a	1: At least an undergraduate degree	0.82	0.38
Household income^b	0: Less than US$4,000	0.61	0.49
Household income^b	1: At least US$ 4,000	0.61	0.49

We first examined the participants’ education by categorical variables, and the average was 3.82 (3: pursuing an undergraduate degree; 4: completed an undergraduate degree). Because 47% of Koreans have an undergraduate degree (Organisation for Economic Co-operation and Development, 2016), we transformed the education variable into a dummy variable differentiating between individuals who hold “less than undergraduate degree” and those who hold “at least an undergraduate degree.”

We first examined the participants’ household income by categorical variables, and the average was 4.74 (4: US$3,000-US$4,000; 5: US$4,000-US$5,000). Because the average household income in Korea is US$4,034.11 (Korean Statistical Information Service, 2016), we transformed the household income variable into a dummy variable differentiating between individuals who earn “less than US$4,000” and those who earn “at least US$4,000.”

Intention to share news online

When the news item in the chosen category was shown to the participants, the participants indicated how much they would like to share the news item (1: not at all; 5: very much) with (a) the CSS with the most frequent communication and (b) the OAS with the most frequent communication, respectively (see Table 2 for descriptive statistics and correlations of variables).

Table 2.

Descriptive Statistics and Correlations of Variables of CSS and OAS.

	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)	(10)
(1) Self-presentation motif	—
(2) Controversiality	.20*	—
Closed symmetric
(3) Audience size	.08	.08	—
(4) Feeling of closeness	.42*	.10*	−.02	—
(5) Frequency of communication	.07*	.03	.16*	.10*	—
(6) Intention to share news	.30*	.27*	−.05	.27*	−.08*	—
Open asymmetric
(7) Audience size	.06	−.01	—	—	—	—	—
(8) Feeling of closeness	.52*	.20*	—	—	—	—	.19*	—
(9) Frequency of communication	−.16*	.04	—	—	—	—	−.23*	−.18*	—
(10) Intention to share news	.32*	.29*	—	—	—	—	−.01	.35*	−.10*	—
M	3.61	3.48	10.98	3.89	26.30	3.06	95.45	3.53	2.50	2.94
SD	0.53	0.83	13.29	0.67	23.75	1.17	155.88	0.70	1.41	1.19

Note. CSS = closed and symmetrical social media; OAS = open and asymmetrical social media.

Analysis

The study conducted two multiple regressions to examine the influence of perceived news controversiality, audience features, and their interaction terms on the intention of online news sharing on CSS and OAS. Model 1 for both CSS and OAS was a baseline model with all control variables included. Model 2 examined the effect of perceived controversiality of news content. Model 3 examined the effect of audience features. The final model examined the influence of the interaction terms between perceived news controversiality and audience features.

Results

H1 predicted that there will be an inverted U-shaped relationship between individuals’ intention to share news on CSS and the controversiality of news. The final model suggests that individuals are more likely to share controversial news when the controversiality increases (β = .15, p < .01); when news becomes controversial at the excessive level, the intention to share news decreases (β = −.20, p < .01). Thus, H1 was supported (see Table 3).

Table 3.

The Impact of News Controversiality and Audience Features on the Intention to Share News on CSS.

	Model 1		Model 2		Model 3		Final model
	β	(SE)	β	(SE)	β	(SE)	β	(SE)
Gender	−.06	(.11)	−.03	(.11)	.01	(.11)	.01	(.11)
Age	.04	(.01)	.04	(.01)	.05	(.01)	.03	(.01)
Education	.01	(.16)	.02	(.15)	.02	(.15)	.02	(.15)
Income	−.02	(.12)	−.02	(.12)	−.03	(.11)	−.02	(.11)
Self-presentation motivation	.30**	(.11)	.26**	(.10)	.21**	(.11)	.22**	(.12)
Controversiality			.19**	(.07)	.19**	(.07)	.15**	(.08)
(Controversiality)²			−.19**	(.06)	−.19**	(.06)	−.20**	(.06)
Audience size					−.08	(.00)	−.08*	(.01)
Feeling of closeness					.14**	(.09)	.08	(.11)
Frequency of communication					−.08	(.00)	.03	(.00)
Controversiality × Size							.01	(.01)
Controversiality × Feeling of Closeness							.08	(.10)
Controversiality × Frequency of Communication							−.09	(.00)
(Controversiality)² × Size							−.004	(.01)
(Controversiality)² × Feeling of Closeness							.05	(.08)
(Controversiality)² × Frequency of Communication							−.13*	(.00)
N	400		400		400		400
R ²	.10		.17		.20		.22
ΔR²	.10		.07		.03		.02
F	8.27**		18.56**		4.63**		2.80*

Note. CSS = closed and symmetrical social media.

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

H2 predicted that there will be a positive relationship between individuals’ intention to share news on OAS and the news controversiality. The final model suggests that individuals are more likely to share news on OAS when the news becomes controversial (β = .23, p < .01). The squared term of news controversiality did not have a significant influence on the intention to share news on OAS; the intention to share news does not decrease even when the controversiality reaches an excessive level. Thus, H2 was supported (see Table 4).

Table 4.

The Impact of News Controversiality and Audience Features on the Intention to Share News on OAS.

	Model 1		Model 2		Model 3		Final Model
	β	(SE)	β	(SE)	β	(SE)	β	(SE)
Gender	.06	(.11)	.08	(.11)	.08	(.11)	.10*	(.11)
Age	.16**	(.01)	.16**	(.01)	.14**	(.01)	.12*	(.01)
Education	−.05	(.16)	−.05	(.15)	−.03	(.15)	−.03	(.15)
Income	.04	(.12)	−.03	(.12)	−.06	(.11)	−.05	(.11)
Self-presentation motivation	.31**	(.11)	.26**	(.10)	.14**	(.12)	.15**	(.12)
Controversiality			.23**	(.07)	.20**	(.07)	.23**	(.15)
(Controversiality)²			−.07	(.06)	−.08	(.06)	.05	(.12)
Audience size					−.03	(.00)	−.03	(.00)
Feeling of closeness					.24**	(.10)	.21**	(.19)
Frequency of communication					−.03	(.04)	.16	(.11)
Controversiality × Size							.01	(.00)
Controversiality × Feeling of Closeness							.10*	(.10)
Controversiality × Frequency of Comm.							−.11	(.05)
(Controversiality)² × Size							−.01	(.00)
(Controversiality)² × Feeling of Closeness							.02	(.09)
(Controversiality)² × Frequency of Communication							−.20*	(.05)
N	395		395		395		395
R ²	.13		.19		.23		.25
ΔR²	.13		.06		.04		.02
F	11.43**		14.10**		7.04**		2.82*

Note. OAS = open and asymmetrical social media.

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

RQ1 asked about the influence of the interaction term between audience size and news controversiality on the intention to share news on CSS and OAS. The interaction terms had no significant influence on the intention to share news on either CSS or OAS.

RQ2 asked about the influence of the interaction term between the strength of ties with the audience and news controversiality on the intention to share news on CSS and OAS. The interaction terms between feeling of closeness and both news controversiality variables had no significant influence on the intention to share news on either CSS or OAS. The interaction terms between frequency of communication and the squared term of news controversiality had negative influence on the intention to share news on both CSS (β = −.01, p < .01; see Table 3 and Figure 1) and OAS (β = −.06, p < .01; see Table 4 and Figure 2). Figure 1 suggests that when sharing news of a very high controversiality, almost reaching the level of 5, CSS users were more likely to share the news with those with whom they communicated with a low frequency, rather than a high frequency. Figure 2 also illustrates that when sharing news of a high controversiality, exceeding the level of around 3.5, OAS users were more likely to share the news with communication partners with a low frequency, rather than a high frequency. To sum up, when the controversiality exceeded the moderate level, CSS and OAS users were more likely to share news with an audience with whom their frequency of communication was low than with those with whom their frequency of communication was high.

Figure 1.

Interaction effect between controversiality and frequency of communication on CSS.

Figure 2.

Interaction effect between controversiality and frequency of communication on OAS.

Discussion

While previous studies on news sharing have revealed features of news content that affect news selection and transmission (Berger & Milkman, 2012; Bobkowski, 2015; Cappella et al., 2015; Choi & Toma, 2014; Chen & Berger, 2013 Ma et al., 2014; Trilling et al., 2017; Vosoughi et al., 2018), this study interpreted online news sharing as interpersonal communication and examined how individuals take their audience into account and share controversial news differently on CSS and OAS. The results suggest that both audience and content features affect the decision to share news, and they interact to certain degrees.

First, the test of H1 suggests that individuals’ intention to share news on CSS increases when the news becomes more controversial, but that intention decreases when the news becomes extremely controversial. CSS activates audiences with specific context and characteristics; CSS users take advantage of such natures of the medium for sophisticated relationship management by keeping multiple chat rooms and managing their audiences separately in different chat rooms (Waterloo et al., 2018). As such, CSS users may be more cautious about the social risks of sharing extremely controversial content and ruining their relationships than interesting their audience and gaining more attention. Corresponding with previous research (Chen & Berger, 2013), CSS users seem to share controversial news only up to the point that it can interest the audience without discomforting them. This pattern is close to what we experience in a more natural (unmediated) condition of face-to-face communication.

The result of H2 suggests that OAS users may share controversial news differently from CSS users. Unlike the patterns confirmed in H1, which is consistent with the previous research (Chen & Berger, 2013), there was no negative relationship between the squared term of news controversiality and the intention to share news on OAS. This result shows how individuals deal with the activated audience differently in OAS compared with CSS. Because OAS activates audience with mixed contexts (Baym & boyd, 2012), OAS users do not seem to care as much as CSS users about meeting the expectations of every audience or incurring social risk. They seem more interested in whether they get noticed and in leveraging the connected audience on OAS (Marwick & boyd, 2011) for broader attention and news diffusion.

Answering RQ1, there was no interaction effect between the audience size and the controversiality of news on either CSS or OAS. Audience size was a negative predictor of online news sharing in previous studies (E. M. Kim et al., 2017; Ma et al., 2014), but it does not seem to have a moderating effect when sharing controversial news. As mentioned above, studies suggest that individuals may become more eager to share controversial news with both a smaller and a larger audience (B. R. Brown & Garland, 1971; E. M. Kim et al., 2017; Ma et al., 2014; Stephen & Lehmann, 2009). Additional regression analyses on subsamples of participants with small and large audiences suggest that such a contradiction does not exist. The distribution graphs (Supplemental Appendix C, available online), however, reveal that audience size may play an interesting role in sharing controversial news, which provides direction for future studies.

Answering RQ2, the interaction term between the frequency of communication and the squared news controversiality showed a negative effect on both CSS and OAS. The interaction terms with a feeling of closeness did not have any significant influence. The results suggest that the more CSS and OAS users communicated with their audience, the more they became reluctant to share extremely controversial news to maintain their self-image and relationships. Frequent communication partners may not be the same people with whom users feel psychologically comfortable. For many CSS users, their work colleagues, their boss, and their children’s friends’ parents are their most frequent communication partners (“Katalk,” 2017), and it is important to maintain their self-image and relationships with these individuals. OAS users who frequently upload postings and communicate with their audience may also pay much attention to maintaining their online reputation and authenticity (Marwick & boyd, 2011). Some of them even regard themselves as celebrities and their audience as fans. As such, they may be more cautious and strategic about what they share with the frequent communication partners than with communication partners with a low frequency when news controversiality exceeds a moderate level.

To summarize, the results suggest interesting differences in how individuals account for different audiences on CSS and OAS. OAS users were more likely to share controversial news regardless of the degree of controversiality, whereas CSS users were concerned about extreme controversiality. Because CSS activates audiences with specific contexts and characteristics, CSS users seem to manage shared news to maintain their self-image and relationships with care in each chat room. On the contrary, because OAS activates audiences with mixed contexts and characteristics (Litt, 2012; Marwick & boyd, 2011), OAS users seem to pay more attention to the broader audience and influence as long as they do not discomfort their frequent communication partners. Specifically, many relationships on OAS are so weak or fleeting that unless the interactions are frequent they may not be counted as salient “relationships” worthy of deep consideration for relationship management. The differences suggest that as media features in CSS and OAS activate different types of audience groups, each medium seems to cultivate different types of relationships and serve different self-presentational purposes. How media features activate different types of audience groups and how individuals take account of their audience from the activated audiences together determine whether controversial news spreads broadly or disappears.

Interpreting the differences from a news diffusion perspective, rather than from a relational communication perspective, CSS seems to have a self-managing mechanism that when news is too controversial, users curb further sharing and discussion. News diffusion through CSS may cool down by itself more easily than it does on OAS.

The mechanism of cooling down does not seem to kick in with OAS. While CSS may act as a space for managing and modulating controversies via sophisticated control, OAS may act as a space for spreading controversial issues more broadly among members of society, and controversy can mount in the process. How individuals use and understand OAS may be similar to what previous studies have focused on in online news sharing: fast and broad diffusion (Berger & Milkman, 2012; Berger & Schwartz, 2011; Cappella et al., 2015; Chen & Berger, 2013; Choi & Toma, 2014; Ma et al., 2014). However, this dissemination process does not stem from individuals focusing only on the news content (e.g., interesting, negative, or controversial news), as previous studies have suggested; it stems from individuals taking their audience into account and sharing news depending on the audience’s features.

Conclusion

This study interpreted online news sharing as an interpersonal communication and examined how individuals take their audience into account and share controversial news differently on CSS and OAS. Focusing on the fact that news is not only transmitted to but also shared with the audience, this study explained how relational features of audience and different media features may provide explanations beyond the content features (i.e., news controversiality) of online news sharing.

There are several limitations in this study. First, most measures in this study are self-reported, not manifest or objective numbers, such as individuals’ intention to share news. Second, the sample for this study was based on a specific cultural context in Korea. However, the focus of this study was not on revealing specific communication patterns of news sharing in social media but on examining how individuals share controversial news differently as they consider their audience’s features for their interpersonal motives. Koreans’ consideration of their audience’s features and interpersonal motives in online news sharing may still provide meaningful insights to understanding online news sharing. Furthermore, the participants in this study were highly educated because they had to have experiences of sharing news via both CSS and OAS to qualify for the online survey.

Each of these limitations can be considered a direction for future research. For instance, although how individuals perceive their relationships with their audience (i.e., feeling of closeness) or their intention to share news on closed social media can be studied based on their self-reported responses, future studies based on log files (i.e., the most shared news in Twitter) may further strengthen the results of this study. In addition, although sharing news on social media is a universal phenomenon (Ihm & Kim, 2018), conducting multicultural studies may provide broader understanding of online news-sharing behaviors.

This study also leaves a few questions that are worth pursuing in future research. One direction would be to examine audience features other than the three features in this study (i.e., size, frequency of communication, feeling of closeness) that individuals may take into account during online news sharing. Examining how other structural or environmental features may activate audiences differently also remains an important potential topic to develop the concepts of imagined and activated audience.

This study makes three contributions to research on online news sharing. First, this study interprets the online behavior of sharing controversial news as a relational communication and combines journalism and communication research. Many studies have focused on informational aspects of news diffusion (Bobkowski, 2015; Cappella et al., 2015; Choi & Toma, 2014; Vosoughi et al., 2018). While news controversiality represents content features of news sharing, this study showed that relational motivations may determine the spread and subsidence of such content.

Second, this study provides insight into how controversies are managed, intensified, or dissipated in different media environments. Previous studies have mostly explained the online news diffusion without distinguishing different media environments (Berger & Milkman, 2012; Ma et al., 2014; Vosoughi et al., 2018). While controversial news, by its nature, may interest audiences, media environments and relationships cultivated in the environments may affect whether controversial news becomes more or less controversial once posted. There are various types of social media, and an increasing number of new media appear every day, so describing only technological features of different media does not provide novel insights into communication scholarship. This study captures how individuals take into account different audience characteristics activated by different technological features of media in intensifying or dissipating controversial news. This attempt enriches previous studies on news and online communication (Cappella et al., 2015; Chen & Berger, 2013). This study additionally provides practical implications. For instance, social activists who aim to spread information and mobilize individuals should simultaneously account for different social media environments and the degree of controversiality of the content (Bobkowski, 2015; Cappella et al., 2015; Chen & Berger, 2013; Choi & Toma, 2014; Ma et al., 2014; Trilling et al., 2017; Vosoughi et al., 2018).

Finally, this study develops the concept of imagined audience and provides future directions. This study tried to go beyond the concept of imagined audience and descriptions of the audience (Cook & Teasley, 2011; Litt & Hargittai, 2016; Marwick & boyd, 2011). Instead, we suggested using the concept “activated audience” and delved into media features that activate different audiences and audience features that users take into account across media. Calling all social media audiences “imagined” obscures the impact of different media features. The different results between CSS and OAS underline the differences between media features and support the need to further explore the idea of activated audience.

This study interpreted online news sharing as an interpersonal communication and examined how individuals take their audience into account and share controversial news differently on CSS and OAS. Corresponding with the recent scholarly demand for the integration of mass communication and interpersonal communication, this study explained the mass diffusion of controversial news via interpersonal interactions. The paths of controversial news from such interpersonal motives may explain the mechanism of how social problems are dissipated and addressed in society and how media may act as a space for conversation for deliberative democracy in this process (J. Kim et al., 1999).

Supplemental Material

DS_10.1177_1077699019836950 – Supplemental material for More Than Virality: Online Sharing of Controversial News With Activated Audience

Supplemental material, DS_10.1177_1077699019836950 for More Than Virality: Online Sharing of Controversial News With Activated Audience by Eun-mee Kim and Jennifer Ihm in Journalism & Mass Communication Quarterly

Footnotes

Appendix A

Appendix B

Selected News Articles.

Category	News article	No. of responses
Politics	President Park “Will actively push ahead the labor reform despite various challenges”	8,862
Economics	Chairman Woohyun Chung, attacking a security guard, Now Saying “I am Sorry”	1,558
Society	Bride School is an old school . . . No vocation, no marriage.	4,121
Lifestyle	A luxury addict in a red light district, Dior Exposition, Controversy on Looking down on Korean Women	4,786
World	Trump “Will Allow the U.S. Entry of Wealthy Muslims”	866
IT/Science	Murderer AI about to be born . . . development should be banned	612

Acknowledgements

The authors gratefully acknowledge the feedback from Su Jung Kim, Jin Woo Kim, Chankyung Pak, and the anonymous reviewers.

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) declared receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Ministry of Education of the Republic of Korea and the National Research Foundation of Korea (NRF-2018S1A5B8070398). This study was financially supported in part by the Institute of Communication Research, Seoul National University. The present research has been conducted by the Research Grant of Kwangwoon University in 2019.

Supplemental Material

Supplemental material for this article is available online.

ORCID iD

Jennifer Ihm

Author Biographies

Eun-mee Kim is a professor of communication, Seoul National University. Her recent research covers the use of digital media and its social and cultural implications.

Jennifer Ihm is interested in how information and communication technologies (ICTs) can contribute to more community engagement and how online engagement can extend to offline environment ().

References

Baumeister

R. F.

(1982). Self-esteem, self-presentation, and future interaction: A dilemma of reputation. Journal of Personality, 50, 29-45. doi:10.1111/j.1467-6494.1982.tb00743.x

Baumeister

R. F.

Hutton

D. G.

(1987). Self-presentation theory: Self-construction and audience pleasing. In Mullen

Goethals

G. R.

(Eds.), Theories of group behavior (pp. 71-87). New York, NY: Springer.

Baym

N. K.

boyd

(2012). Socially mediated publicness: An introduction. Journal of Broadcasting & Electronic Media, 56, 320-329. doi:10.1080/08838151.2012.705200

Berger

Milkman

K. L.

(2012). What makes online content viral?. Journal of marketing research, 49(2), 192-205.

Berger

Schwartz

E. M.

(2011). What drives immediate and ongoing word of mouth? Journal of Marketing Research, 48, 869-880. doi:10.1509/jmkr.48.5.869

Bobkowski

P. S.

(2015). Sharing the news: Effects of informational utility and opinion leadership on online news sharing. Journalism & Mass Communication Quarterly, 92, 320-345.

Brown

B. R.

Garland

(1971). The effects of incompetency, audience acquaintanceship, and anticipated evaluative feedback on face-saving behavior. Journal of Experimental Social Psychology, 7, 490-502. doi:10.1016/0022-1031(71)90011-4

Brown

Reingen

P. H.

(1987). Social ties and word-of-mouth referral behavior. The Journal of Consumer Research, 14, 350-362. doi:10.1086/209118

Cappella

J. N.

Kim

H. S.

Albarracín

(2015). Selection and transmission processes for information in the emerging media environment: Psychological motives and message characteristics. Media Psychology, 18, 396-424. doi:10.1080/15213269.2014.941112

10.

Chen

Berger

(2013). When, why, and how controversy causes conversation. Journal of Consumer Research, 40, 580-593. doi:10.1086/671465

11.

Choi

Lee

(2017). Trust in open versus closed social media: The relative influence of user-and marketer-generated content in social network services on customer trust. Telematics and Informatics, 34(5), 550-559.

12.

Choi

Toma

C. L.

(2014). Social sharing through interpersonal media: Patterns and effects on emotional well-being. Computers in Human Behavior, 36, 530-541.

13.

Chu

S. C.

(2009). Determinants of consumer engagement in electronic word-of-mouth in social networking sites. International Journal of Advertising, 30, 47-75. doi:10.2501/IJA-30-1-047-075

14.

Cook

E. C.

Teasley

S. D.

(2011, February). Beyond promotion and protection: Creators, audiences and common ground in user-generated media. In Proceedings of the 2011 iConference (pp. 41-47). New York, NY: ACM. doi:10.1145/1940761.1940767

15.

Das

Kramer

(2013). Self-censorship on Facebook. In Proceedings of the Seventh International AAAI Conference on Weblogs and Social Media (pp. 120–127). Washington, DC: Association for the Advancement of Artificial Intelligence.

16.

Flanagin

A. J.

(2017). Online social influence and the convergence of mass and interpersonal communication. Human Communication Research, 43, 450-463. doi:10.1111/hcre.12116

17.

France

M. K.

Finney

S. J.

Swerdzewski

(2010). Students’ group and member attachment to their university: A construct validity study of the University Attachment Scale. Educational and Psychological Measurement, 70, 440-458. doi:10.1177/0013164409344510

18.

Gil de Zúñiga

H. G.

Jung

Valenzuela

(2012). Social media use for news and individuals’ social capital, civic engagement and political participation. Journal of Computer-mediated Communication, 17, 319-336. doi:10.1111/j.1083-6101.2012.01574.x

19.

Goffman

(1959). The presentation of self in everyday life. New York, NY: Doubleday Anchor.

20.

Govern

J. M.

Marsch

L. A.

(2001). Development and validation of the situational self-awareness scale. Consciousness and Cognition, 10, 366-378. doi:10.1006/ccog.2001.0506

21.

Granovetter

M. S.

(1973). The strength of weak ties. American Journal of Sociology, 78, 1360-1380. doi:10.1086/225469

22.

Hogg

M. A.

Terry

D. J.

White

K. M.

(1995). A tale of two theories: A critical comparison of identity theory with social identity theory. Social Psychology Quarterly, 58, 255-269. doi:10.2307/2787127

23.

Ihm

Kim

E. M.

(2018). The hidden side of news diffusion: Understanding online news sharing as an interpersonal behavior. New Media & Society, 20, 4346-4365. doi:10.1177/1461444818772847

24.

Kakaotalk, mobile messenger monthly active users 95%. (2016). Retrieved from http://www.dt.co.kr/contents.html?article_no=2016102502109931033002

25.

Katalk, Katalk, after work, I want to logout. (2017). Retrieved from http://news.chosun.com/site/data/html_dir/2017/09/20/2017092000936.html

26.

Kim

E. M.

Ihm

Park

H. A.

(2017). News sharing as relational communication: Focusing on self-presentation tendency and characteristics of sharing audiences. Korean Journal of Broadcasting and Telecommunication Studies, 31, 114-151.

27.

Kim

Lee

J. K.

(2016). Social media type matters: Investigating the relationship between motivation and online social network heterogeneity. Journal of Broadcasting & Electronic Media, 60(4), 676-693.

28.

Kim

Wyatt

R. O.

Katz

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

29.

Korea Information Society Development Institute (KISDI) (2016). KISDI STAT report 16-24. Available at: https://stat.kisdi.re.kr/MediaStat/Library/Library_detail1.aspx?MENU_ID=233&Division=1&seq=1360 (accessed 5 March 2019).

30.

Korea Information Society Development Institute. (2017). KISDI STAT report 18-03: Media panel. Retrieved from http://m.kisdi.re.kr/mobile/repo/stat_view.m?key1=14266&selectPage=

31.

Korean Statistical Information Service. (2016). Household income statistics. Retrieved from http://kosis.kr/statisticsList/statisticsListIndex.do?menuId=M_01_01&vwcd=MT_ZTITLE&parmTabId=M_01_01#SelectStatsBoxDiv

32.

Lee

(2013). Self-presentation and WOM: Focusing on mediating role of intrinsic and extrinsic motivation. Korean Management Review, 42, 183-211.

33.

Lin

Utz

(2015). The emotional responses of browsing Facebook: Happiness, envy, and the role of tie strength. Computers in Human Behavior, 52, 29-38.

34.

Litt

(2012). Knock, knock. Who’s there? The imagined audience. Journal of Broadcasting & Electronic Media, 56, 330-345. doi:10.1080/08838151.2012.705195

35.

Litt

Hargittai

(2016). The imagined audience on social network sites. Social Media + Society, 2(1), 1-12.

36.

Lee

C. S.

Goh

D. H. L.

(2014). Understanding news sharing in social media: An explanation from the diffusion of innovations theory. Online Information Review, 38, 598-615. doi:10.1108/OIR-10-2013-0239

37.

Marsden

P. V.

Campbell

K. E.

(1984). Measuring tie strength. Social Forces, 63, 482-501.

38.

Marwick

A. E.

boyd

(2011). I tweet honestly, I tweet passionately: Twitter users, context collapse, and the imagined audience. New Media & Society, 13, 114-133. doi:10.1177/1461444810365313

39.

Most popular mobile messaging apps worldwide as of January 2017, based on number of monthly active users (in millions). (2017). Retrieved from https://www.statista.com/statistics/258749/most-popular-global-mobile-messenger-apps/

40.

Neuman

W. R.

(2016). The digital difference. Cambridge, MA: Harvard University Press.

41.

Newcomb

T. M.

(1961). The acquaintance process. New York, NY: Holt, Rinehart, & Winston.

42.

Organisation for Economic Co-operation and Development. (2016). Education at a glance 2016: OECD indicators. Paris, France: Author.

43.

Shirky

(2010). Cognitive surplus: How technology makes consumers into collaborators. New York: The Penguin Press.

44.

Stephen

A. T.

Lehmann

D. R.

(2009). Why do people transmit word-of-mouth? The effects of recipient and relationship characteristics on transmission behaviors (Working paper). New York, NY: Columbia University.

45.

Trilling

Tolochko

Burscher

(2017). From newsworthiness to shareworthiness: How to predict news sharing based on article characteristics. Journalism & Mass Communication Quarterly, 94, 38-60.

46.

Turner

J. C.

Onorato

R. S.

(1999). Social identity, personality, and the self-concept: A self-categorization perspective. In Tyler

T. R.

Kramer

R. M.

John

O. P.

(Eds.), The psychology of the social self (pp. 11-46). New York, NY: Lawrence Erlbaum.

47.

Vosoughi

Roy

Aral

(2018). The spread of true and false news online. Science, 359, 1146-1151. doi:10.1126/science.aap9559

48.

Walther

J. B.

(1996). Computer-mediated communication. Communication Research, 23(1), 3-43. doi:10.1177/009365096023001001

49.

Walther

J. B.

(2011). Theories of computer-mediated communication and interpersonal relations. In Knapp

M. L.

Daly

J. A.

(Eds.), The handbook of interpersonal communication (pp. 443-479). Thousand Oaks, CA: SAGE.

50.

Walther

J. B.

Valkenburg

P. M.

(2017). Merging mass and interpersonal communication via interactive communication technology: A symposium. Human Communication Research, 43, 415-423. doi:10.1111/hcre.12120

51.

Waterloo

S. F.

Baumgartner

S. E.

Peter

Valkenburg

P. M.

(2018). Norms of online expressions of emotion: Comparing Facebook, Twitter, Instagram, and WhatsApp. New Media & Society, 20, 1813-1831.

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

0.52 MB