Sage Journals: Discover world-class research

Abstract

Amid growing concerns over the contentious tenor of online political discourse, scholars have begun to recognize that the social contexts and affordances provided by social media may present indirect pathways from online political discussion to offline political participation. Less work has addressed how users’ motivations for using social media might influence such dynamics. In this study, we use two-wave panel survey data collected in the United States to test the possibility that online cross-cutting discussion involving political disagreement can encourage users to share political information on social media, which in turn can increase their offline political engagement. We also test how specific motivations for using social media (i.e., political engagement, relationship maintenance, and self-promotion) moderate the amount users share political information on social media when engaged in conversations involving political disagreement. Our results find that increased online cross-cutting political discussion indirectly affects offline political participation through the influence of social media political information sharing. We also observe that this indirect effect is stronger for users who are motivated to use social media for either political engagement or relationship maintenance (but not self-promotion) purposes. Our findings advance one route from online political disagreement to offline political action, which can impact both politically and nonpolitically motivated social media users.

Keywords

cross-cutting discussion social media information sharing politics relationship maintenance

Talking politics on social media can be an ugly business. The modern online information environment is littered with political disagreement, as evidenced by heated Facebook posts and fiery online news comment sections alike (Coe, Kenski, & Rains, 2014). The rancor that often characterizes online political talk in part stems from discussion between users with differing political views. Political disagreement in online contexts is now commonplace, as the structures of popular online social media sites weaken social boundaries and allow contact between users with conflicting political views (Brundidge, 2010).

The popular press points to the contentious state of online political talk as a key contributor to our divisive political climate (e.g., Miller, 2014). Scholars similarly have sought to understand the consequences of so-called online cross-cutting discussion (OCCD), in which individuals engage in conversation with those who hold opposing political viewpoints (Brundidge, 2010; Wojcieszak & Mutz, 2009). Theories of deliberative democracy suggest that through discussion, citizens engage in productive debate, form opinions about issues of democratic significance, and directly contribute to democratic governance (Chambers, 2003). There is empirical evidence that cross-cutting discussion promotes political tolerance and improves the quality of political views (Mutz, 2002b; Price, Cappella, & Nir, 2002). Yet, there is also work suggesting that cross-cutting discussion ultimately discourages citizens from expressing themselves and decreases offline participation in the political process (Hampton et al., 2014; Mutz, 2002a).

While findings regarding the consequences of online cross-cutting discussion are mixed, there remains the possibility that the unique affordances and social contexts found on social media may indirectly lead users from online political disagreement to offline political participation. Examples of such indirect pathways are bountiful in the social media literature, where online behavior is shaped by complex motivations to seek information, to build and sustain relationships, and to manage one’s identity (see Zhang & Leung, 2014). As users navigate the treacherous waters of online political conversation, they are guided by their personal motivations for using social media and by the composition of their networks (see Papacharissi, 2010). This shapes how they use the expressive affordances of social media, which in turn may influence their willingness to take more costly offline political action. The ability to share news or political information on social media in particular embodies a potent expressive tool for responding to political disagreement, as it entails both an elaboration of a personal political belief and a public commitment to a specific point of view. There is growing evidence that political information sharing not only allows users to respond to those with whom they disagree or disseminate their point of view (Hasell & Weeks, 2016), but it also drives further online and offline political participation (Boulianne, 2015).

This study investigates the possibility that in the presence of political disagreement, users may become increasingly motivated to share political information on social media and subsequently more likely to participate in politics offline. Inevitably, the way users respond to OCCD will be influenced by the motivations that brought them to social media in the first place (see Quan-Haase & Young, 2010). Accordingly, we also explore how individual motivations for using social media may influence the magnitude of the proposed indirect pathway between OCCD and offline political participation through social media political information sharing (SMPIS).

We make a contribution to previous research in at least three ways. First, we propose and test a theoretical model that helps clarify the current scholarly debate on the consequences of OCCD. By demonstrating a pathway from online political disagreement to offline political participation, we find support for one context in which OCCD can have a positive effect on political participation. Second, we use panel data to strengthen the largely cross-sectional evidence of the causal link between sharing political information on social media and offline political engagement (Boulianne, 2015). Finally, we find that users’ motivations for using social media for specific purposes fundamentally determine the way they respond to online political disagreement. By examining not only users who come to social media to engage in politics, but also those who are motivated by social and self-focused goals, we help clarify who stands to benefit from our proposed theoretical model.

Is Political Disagreement Good for Democracy?

The impact of cross-cutting discussion on democratic participation is contested, with scholars arguing that it either facilitates political participation (e.g., Scheufele, Hardy, Brossard, Waismel-Manor, & Nisbet, 2006) or discourages political participation (e.g., Mutz, 2002a, 2006; Valenzuela, Kim, & Gil de Zúñiga, 2012). On one hand, discussions involving disagreement can increase opinion uncertainty, leaving individuals less confident about their own views and less willing to participate further (Mutz, 2002a, 2006). On the other hand, political disagreement can spur active participation in politics by facilitating political learning (Scheufele, Nisbet, Brossard, & Nisbet, 2004), increasing tolerance for diverging viewpoints (Mutz, 2002b), and helping citizens refine their own political views (Price et al., 2002).

More recently, research has examined cross-cutting discussion on social media, alternately finding that it either increases the heterogeneity of discussion networks (Kim, Hsu, & de Zúñiga, 2013) or makes exposure to differing political viewpoints less likely (Bakshy, Messing, & Adamic, 2015; Himelboim, McCreery, & Smith, 2013). While prior research variously suggests that social media either encourages or limits exposure to political difference, less work has specifically addressed how the social contexts and affordances found on social media might shape citizens’ responses to the political disagreement they encounter online. In this study, we test the possibility that sharing political information on social media may be one strategy for responding to online cross-cutting discussion.

Political Information Sharing as a Response to Political Disagreement

Long a key affordance of social media, sharing has become an increasingly important form of political expression and news dissemination (Gil de Zúñiga, Jung, & Valenzuela, 2012; Weeks & Holbert, 2013). However, the term has suffered from a lack of conceptual clarity and has been used differently by scholars (Kümpel, Karnowski, & Keyling, 2015; Lampinen, 2015). For the purposes of this study, we use the term “social media political information sharing” (SMPIS) to refer to a wide range of behaviors, which allow social media users to share information or views about politics and current affairs with either personal contacts or the wider public. In this sense, SMPIS is a related concept with political expression (see Skoric, Zhu, Goh, & Pang, 2016), but more precisely refers to activities in which concrete political information or views are communicated (as opposed to the signaling of political opinion through up-voting or liking content).

This type of political information sharing abounds on sites like Facebook and Twitter in the form of posts or tweets (Gottfried, Barthel, Shearer, & Mitchell, 2016). Despite its prevalence, political information sharing on social media is also a risky act with potential consequences for users’ reputations and relationships. Thorson (2013) found that college students often avoid sharing political content on Facebook for fear that they will be misunderstood or be unfairly judged. Recent work on hostile social media opinion climates has similarly found that the presence of counter-attitudinal information can discourage users from expressing themselves (Hampton et al., 2014). This is consistent with spiral of silence theory, which predicts that individuals are less likely to express their opinion when they believe they are in the minority (Noelle-Neumann, 1991). Spiral of silence theory has been applied in the context of social media to explain how cross-cutting discussion can lead users to perceive their opinions as less dominant within their network and subsequently discourage them from sharing their political views (Gearhart & Zhang, 2015).

Yet research also suggests that because most partisan social media users have networks populated by predominantly politically like-minded contacts and that algorithms tend to show them slightly more like-minded content (Bakshy et al., 2015), such users are less likely to be exposed to dissonant political views. Encounters with political disagreement make otherwise relatively scarce dissonant political information and views more salient. When faced with increased levels of counter-attitudinal information, users are likely to experience cognitive dissonance as they try to reconcile their political views with others (Donsbach & Mothes, 2013; Festinger, 1962). We argue that within politically homogeneous networks, information sharing may be one effective tool for reducing the dissonance created by online cross-cutting discussion because sharing re-exposes users to pro-attitudinal information. This contention is supported by research showing that when confronted with dissonant political information, users actively utilize the affordances of social media to seek opinion-reinforcing information (Knobloch-Westerwick & Meng, 2011). Sharing may allow users to expose themselves to opinion-reinforcing information through (1) the process of content selection that comes before sharing and (2) feedback from like-minded individuals in response to shared information. Prior to sharing political information, users must seek opinion-reinforcing content and process it themselves (Lee & Ma, 2012), a process which may itself provide the psychological benefit of reducing dissonance. There is also evidence that by expressing their political views on social media, users receive opinion-reinforcing feedback from their networks, thereby reassuring themselves of their existing political beliefs (Cho, Ahmed, Keum, Choi, & Lee, 2016).

This research suggests that sharing can be an effective means for reducing the dissonance created by political disagreement. Accordingly, we argue that OCCD may, in fact, be associated with increased social media political information sharing as users seek opinion-reinforcing information and support from politically like-minded individuals in their networks. We hypothesize the following:

H1. Online cross-cutting discussion (Wave 1 [W1]) will be positively related to social media political information sharing (W1).

The Role of Motivations for Using Social Media

People come to social media with different goals and needs (Park, Kee, & Valenzuela, 2009), which suggests that the relationship between OCCD and SMPIS may be different depending on the motivations of individual users. Motives for social media use have been investigated using the theoretical framework of uses and gratifications, which predicts that people actively use media to satisfy certain needs (Quan-Haase & Young, 2010). Previous literature has identified three primary types of motives for news sharing on social media: altruistic, social, and self-serving (Kümpel et al., 2015). Users with altruistic motives share news content on social media to distribute information to others in their social network (boyd, Golder, & Lotan, 2010). Users with social motives share news in order to socialize and interact with others in their social network (Lee & Ma, 2012). Finally, users with self-serving motives share news content on social media to receive attention and improve their reputations (boyd et al., 2010; Lee & Ma, 2012).

While motives for sharing news on social media have been rather extensively examined (e.g., Skoric et al., 2016), it remains less clear whether different motives for social media use moderate the hypothesized positive relationship between OCCD and SMPIS. Building upon Kümpel et al. (2015) and other prior literature, we consider three motives for using social media relevant to our theoretical model: (1) engaging in social issues and politics, (2) maintaining relationships, and (3) self-promotion.

First, for people who already use social media to engage in social issues and politics, the link between OCCD and SMPIS should be stronger. Such users are likely to be politically interested, and therefore, encounters with disagreements via OCCD should produce greater cognitive dissonance (Donsbach & Mothes, 2013; Festinger, 1962). In order to reduce cognitive dissonance, these users should be motivated to confirm and reinforce their prior beliefs and to affirm their political identity through information sharing (Green, Palmquist, & Schickler, 2004). Along these lines of thinking, we advance the following hypothesis:

H2. The relationship between online cross-cutting discussion (W1) and social media political information sharing (W1) will be stronger for individuals who are motivated to use social media to engage in social issues and politics (W1).

How might nonpolitical motivations for using social media influence the likelihood that users will share when faced with political disagreement? The maintenance of relationships with family, friends, and acquaintances has been identified as a dominant motivation for using social media (Ellison, Vitak, Gray, & Lampe, 2014). When such users encounter high levels of political disagreement online, sharing political information may offer a way to build and maintain relationships. For example, on Twitter, users were found to share news articles that opposed their own views only if these articles matched their followers’ political views (An, Quercia, Cha, Gummadi, & Crowcroft, 2014). For these Twitter users, sharing news articles that opposed their views might have served as a means of socializing and maintaining relationships. However, it is also possible that those who use social media for relationship maintenance may interpret political disagreement as threatening to their relationships and view sharing as a behavior that may further disrupt their social interactions (Thorson, 2013). Indeed, some Twitter users tend to avoid discussing controversial topics and instead prefer to share information on subjects that are safe for all possible audiences (Marwick & boyd, 2011).

Social media has also become an important means of impression management and self-promotion (Ellison, Heino, & Gibbs, 2006; Gentile, Twenge, Freeman, & Campbell, 2012). In an effort to appeal to broad audiences, social media users emphasize or de-emphasize certain aspects of their identity, depending on the feedback they receive (Marwick & boyd, 2011). When individuals who use social media for self-promotion are exposed to political disagreements online, they may share political information as a way of drawing attention to themselves or eliciting reactions from their social networks. Being viewed as politically knowledgeable or opinionated may project a desired image for these users (Gil de Zúñiga et al., 2012). However, it is also possible that sharing is too reputationally risky, particularly under circumstances where political disagreement signals the potential for conflict between users. With these competing possibilities, we advance the following research question:

RQ1. Does the relationship between online cross-cutting discussion (W1) and social media political information sharing (W1) vary for individuals who are motivated to use social media for either relationship maintenance or self-promotion purposes?

Political Information Sharing and Offline Political Participation

As digital technologies have become more prevalent in people’s lives, there have been concerns that Internet use will diminish traditional, offline participation in politics (e.g., Putnam, 2000). Others have similarly argued that social media in particular will negatively impact offline political participation because it allows for low-cost and low-involvement forms of political expression that enable people to feel involved even if they are not, thereby limiting their actual participation (Gladwell, 2010). However, these worries may be overly pessimistic as research suggests that using the Internet and social media is often associated with offline political participation (see Boulianne, 2009, 2015);

Much of the work examining how social media relate to offline political engagement focuses on how low-involvement forms of political expression within these sites can encourage, rather than diminish, traditional forms of participation (Vaccari et al., 2015). Political information sharing on social media is considered to be a low-involvement or low-threshold form of political expression because the physical cost of engaging in this behavior is relatively low (Chadwick, 2009; Vaccari et al., 2015). Although little physical effort is required on the part of a social media user, sharing political information within these platforms can facilitate offline political participation. As Pingree (2007) argues, political self-expression can have positive spillover effects into other areas of citizenry because it enables the expresser to elaborate on and attend to the message, which can motivate them to engage politically. By disseminating political information, sharers may further convince themselves of the need to go beyond the act of self-expression on social media and take offline political action (Rojas & Puig-i-Abril, 2009). Furthermore, posting political information online may help individuals learn more about politics, coordinate political events, and find ways to participate, all of which should result in more participation offline (Kwak, Williams, Wang, & Lee, 2005). This contention is supported by empirical evidence, which demonstrates that engaging in various forms of political expression, including sharing and posting political content on social media, is associated with greater political participation offline (Gil de Zúñiga, Molyneux, & Zheng, 2014; Vaccari et al., 2015). Based on existing theory and empirical findings, we offer the following hypothesis:

H3. Social media political information sharing (W1) will be positively related to offline political participation (Wave 2 [W2]).

From Disagreement to Action Through Sharing: A Theoretical Model

The reviewed literature suggests the possibility that OCCD may in fact facilitate offline political participation through the influence of social media political information sharing. While no work to date has explicitly tested such an indirect pathway, there is ample evidence that certain patterns of online political behavior can result in increased offline political participation (e.g., Kwak et al., 2005; Vaccari et al., 2015). Accordingly, we put forth a theoretical model, which predicts an indirect pathway from online disagreement to offline action through social media political information sharing (Figure 1) and hypothesize the following:

H4. Online cross-cutting discussion (W1) will be indirectly positively related to offline political participation (W2) through social media political information sharing (W1).

Figure 1.

Theoretical Model.

We also test the possibility that this predicted indirect pathway varies depending on users’ motivations for using social media. As there is no prior work on this effect, we pose the following research question:

RQ2. Does the indirect effect of online cross-cutting discussion (W1) on offline political participation (W2) through social media political information sharing vary depending individuals’ motivations for using social media (engagement in social issues and politics, relationship maintenance, or self-promotion)?

Method

Sample

This study uses data from a two-wave national online survey conducted in the United States. Online respondents were recruited by the survey research company, Ipsos, which was also responsible for all data collection. The sample was drawn from a pre-recruited panel of approximately 1 million households who agreed to be contacted periodically to take part in online surveys. Quotas for age and gender were set in order to yield a final sample that would be reasonably reflective of the national population. While this quota sampling method is distinct from conventional probability sample procedures, it has been found to yield comparable data (Putnam, 2000). W1 of the survey was conducted in October, 2012 during the general campaign period of the 2012 US presidential election. This first wave had a 7.2% response rate with 1,250 respondents completing the survey out of 17,381 individuals who were invited to participate. This response rate is comparable to those reported by major research organizations during the same period (Pew Research Center, 2012). Invitations to participate in W2 were sent to all W1 respondents in November 2012. In total, 950 respondents completed the survey, resulting in a 76% retention rate. The final sample features a demographic profile that resembles national population figures reported in US Census Bureau’s 2012 American Community Survey (ACS) for variables, including median age of individuals 18 years or older (ACS = 45–54, W1 = 47), percentage of females (ACS = 51.4%, W1 = 52%), and median household income (ACS = US$50,000–US$74,999, W1 = US$$60,000–US$64,999). The only exception is median educational attainment for those aged 25 years or older, which is higher in our W1 sample (college diploma) than in the ACS (some college). W2’s sample demographic characteristics are comparable to the W1 sample, with the exception of median age, which is higher in W2 (53 years) compared to W1 (47 years).

Because our study concerns motivations and behaviors specific to social media, we further limit the analyses and descriptive statistics reported below to respondents who reported using social media such as Facebook and Twitter during W1 (n = 861, 68.9%) and who were retained in the W2 sample. The final sample includes 594 respondents.

Measures

Online Cross-cutting Discussion

To assess the frequency of online conversation involving political disagreement, an index was created using two measures, which asked respondents how frequently in the past month they (1) “discussed politics online with others who opposed the candidate they favored” and (2) “had conversation online about politics or social issues that involved disagreement.” Both items were measured on a six-point scale ranging from “none” to “everyday.” The resulting index serves as our measure of online cross-cutting discussion (OCCD) (W1; M = 1.50, standard deviation [SD] = 0.96, r = .75).

Social Media Political Information Sharing

In order to account for how often respondents engaged in both directed and general political information sharing on social media during W1, they were asked how frequently in the past month they (1) “reached out to friends, family, or acquaintances individually through social networking sites (social media) to share information or views about politics and current affairs” and (2) “posted information or views on social media that many people could see to share information or views about politics and current affairs.” These measures were assessed using the same six-point scale as the discussion measures and combined into an index of social media political information sharing (SMPIS) (W1; M = 1.91, SD = 1.33, r = .81).

Offline Political Participation

To tap into the criterion variable of offline political participation, five measures were employed, which asked respondents how frequently they (1) “attended a political meeting, rally, or speech”; (2) “worked for a candidate or a political party”; (3) “contacted a public official or a political party?”; (4) “encouraged others to vote”; and (5) “contributed money to a candidate or a political party.” Each measure was assessed using the same six-point scale as previous frequency measures and then combined into two indices for offline political participation (one for each wave). Offline political participation (W2) was used as our dependent variable (M = 1.60, SD = 0.72, α = .76), while offline political participation (W1) was used to control for baseline offline political participation (M = 1.50, SD = 0.68, α = .78).

Social Media Motives

Previous research has established a variety of motivations for using social media (e.g., Kümpel et al., 2015). This study looks at the moderating role of three specific motivations—political engagement, relationship maintenance, and self-promotion—each of which has been shown to positively influence online and offline social capital and civic engagement (Park et al., 2009). In order to assess users’ motivations, we asked respondents whether they had used social media, such as Facebook and Twitter, for specific purposes. This allows us to measure the motivational component of respondents’ social media use, which we treat as a reasonable assessment of their motivations.¹ The items were assessed using a six-point scale ranging from “definitely disagree” to “definitely agree.” Motivations for using social media for engagement in politics and social issues were assessed using an index of four measures: (1) “To exchange information on public affairs and politics,” (2) “To discuss news and public affairs,” (3) “To increase awareness about important issues,” and (4) “To advocate for a social cause” (W1; M = 2.71, SD = 1.44, α = .91). Motivations for using social media for relationship maintenance were assessed using an index of two measures: (1) “To stay in touch with others” and (2) “To maintain relationships with others in my network” (W1; M = 4.93, SD = 1.26, r = .77). Finally, motivations for using social media for self-promotion were assessed using an index of two measures: (1) “To impress others with my personal feats or hidden talent” and (2) “To keep people updated on my public accomplishments” (W1; M = 2.27, SD = 1.39, r = .76).

Control Variables

We also include a number of control variables in our analyses, which are either theoretically related to the independent and mediating variables or are known to influence the dependent variable. Specifically, we aim to control for the possibility that any relationships we observe are merely due to overall use of the Internet or social media. Given that those who participate in politics online may be more likely to engage in offline political participation (Boulianne, 2009), we control for overall online political participation. This measure was constructed by creating an index of how often in the past month respondents engaged in six online political behaviors (e.g., started an online petition, clicked a link to join a group online) on the same six-point frequency scale as previous measures (W1; M = 1.64, SD = 0.70). To account for the possibility that SMPIS might simply be reflective of an overall tendency to share on social media, we control for the frequency of social media personal information sharing by asking respondents how frequently in the past month they used social media to “share information or views about personal matters.” This item was measured on the same six-point frequency scale (W1; M = 2.44, SD = 1.43). Research also suggests that both sending and receiving political information may be related in regards to their influence on subsequent offline participation (Vaccari et al., 2015). We therefore control for social media political information reception using two measures asking respondents how often in the past month they received political information either directly or indirectly. Each item was measured on the same six-point frequency scale, and both were combined into an index of social media political information receiving (W1; M = 3.18, SD = 1.62, r = .53).

Previous research suggests that SMPIS is influenced by both respondents’ political interest and level of political news consumption (Hasell & Weeks, 2016). We therefore control for political interest by asking respondents their agreement with the statement, “I am very interested in politics,” on a six-point scale (1: “definitely disagree” to 6: “definitely agree”; M = 3.42, SD = 1.56). To assess traditional media use, respondents were asked how often they used daily newspapers, national nightly news, and local television news programs. Responses to each item were measured on a five-point scale, ranging from “never” to “very often,” and then summed as an index of traditional media use (M = 3.18, SD = 1.02, Cronbach’s α = .67).

Next, we control for constructs known to more broadly influence willingness to engage in politics, including political efficacy (Caprara, Vecchione, Capanna, & Mebane, 2009) and strength of partisanship (Greene, 1999). Political efficacy was measured by asking respondents to report the extent of their agreement with the statement, “I think I can have a significant impact on how things are going in politics” on a six-point scale (1: “definitely disagree,” 6: “definitely agree”; M = 3.42, SD = 1.56). To assess the strength of partisanship, respondents were asked to report their partisan affiliation as either: Strong Democrat, Moderate Democrat, Independent, Moderate Republican, Strong Republican, or Other. Strength of partisanship was computed by recoding partisan affiliation (1 = no party affiliation [N = 179], 2 = moderate Republicans/Democrats [N = 243], 3 = strong Republicans/Democrats [N = 172]; M = 1.99, SD = 0.77). Finally, age, gender, and education were included as demographic control variables.

Results

In order to test the theoretical model proposed in Figure 1, we conducted three moderated mediation analyses, each using a discrete social media use motive (i.e., engagement, relationship maintenance, or self-promotion) as a moderator. All models were tested using Model 7 of the SPSS macro PROCESS, which employs ordinary least-squares path analyses and computes point estimates with bias-corrected bootstrapped confidence intervals (Hayes, 2013).

The mediation portion of the model tests the indirect effect of OCCD (W1) on offline political participation (W2) through SMPIS (W1). Additionally, we examine the extent to which the relationship between OCCD and SMPIS is moderated by each given motive for using social media. To strengthen our test of the causal link between SMPIS (W1) and offline political participation (W2), we use panel data, which allow us to effectively assess the change in offline political participation by controlling for W1 levels of participation (Eveland & Thomson, 2006). In each model, we control for demographic characteristics, relevant political attitudes and behavior, social media–specific behaviors, and other social media use motives. Per Hayes’ (2013) recommendation, we report unstandardized regression coefficients and include change in R² as a measure of effect size for key variables (provided as a note in each table).²

To test whether OCCD is positively related to SMPIS (H1), we begin by assessing the relationship between all predictor variables and SMPIS prior to the addition of interaction terms (Table 1, first column).³ OCCD is positively related to SMPIS, b = .35 (.05), p < .001, indicating that those who engage in more frequent online discussion involving political disagreement are also more likely to share political information on social media. This provides support for our first hypothesis.

Table 1.

Summary of Regression Analyses for Moderated Mediation Model Using Engagement SM Motives (W1) as the Moderator.

	SM political information sharing (W1) (mediator)		Offline political participation (W2) (criterion)
	b (SE)	b (SE)	b (SE)
SM political information sharing (W1)	–	–	.04 (.02)^#
Online cross-cutting discussion (W1)	.35 (.05)***	.07 (.11)	.01 (.03)
Engagement SM motives (W1)	.24 (.03)***	.14 (.05)**	–
Online cross-cutting discussion (W1) × engagement SM motives (W1)	–	.07 (.02)**	–
Relationship maintenance SM motives (W1)	−.08 (.03)**	−.08 (.03)**	−.01 (.02)
Self-promotion SM motives (W1)	−.03 (.03)	−.03 (.03)	.02 (.02)
Offline political participation (W1)	−.04 (.07)	−.06 (.07)	.69 (.04)***
Online political participation (W1)	.18 (.07)**	.17 (.07)**	.03 (.04)
SM personal information sharing (W1)	.18 (.03)***	.18 (.03)***	−.002 (.02)
SM political information receiving (W1)	.25 (.03)***	.25 (.03)***	−.01 (.02)
Age (W1)	.001 (.002)	.001 (.002)	−.0003 (.001)
Sex (W1) (male = 1, female = 2)	−.23 (.07)***	−.24 (.07)***	.01 (.04)
Education (W1)	−.04 (.04)	−.04 (.04)	.03 (.02)
Political interest (W1)	.02 (.03)	.02 (.03)	.01 (.02)
Political self-efficacy (W1)	.05 (.03)^#	.05 (.03)^#	.04 (.02)*
Traditional media use (W1)	.07 (.04)^#	.06 (.04)^#	.01 (.02)
Strength of partisanship (W1)	.004 (.04)	−.002 (.04)	−.02 (.03)
Constant	−.14 (.28)	.30 (.32)	.11 (.16)
Adjusted R²	.65	.66	.60
Residual SE	.79 (df = 578)	.78 (df = 577)	.45 (df = 578)
F statistic	74.57*** (df = 15; 578)	71.40*** (df = 16; 577)	59.28** (df = 15; 578)

SM: social media; SE: standard error.

Unstandardized coefficients reported. SEs in parentheses. N = 594. In models predicting SM political information sharing (W1), ΔR² for online cross-cutting discussion (W1) = .034, ΔR² for online cross-cutting discussion (W1) × engagement SM motive (W1) = .005 (all ps < .05). In model predicting offline political participation (W²), ΔR² for SM political information sharing (W1) = .002 (marginally significant, p < .1).

p < .1; *p < 0.05; **p < 0.01; ***p < 0.001 (two-tailed).

In this model, engagement motives are positively related to SMPIS, b = .24 (.03), p < .001, while relationship maintenance motives are negatively related to SMPIS, b = −.08 (.03), p < .01. Individuals who are motivated to use social media to engage in politics are, in general, more likely to share political information, while those motivated to use social media to build and maintain relationship are less likely to share. Self-promotion motives were not a significant predictor of SMPIS, suggesting that the extent to which individuals use social media to promote themselves has no direct influence on their sharing behavior.

Next, we address whether the three social media use motives (engagement, relationship maintenance, and self-promotion) moderate the relationship between OCCD and SMPIS (H2 and RQ1). In the model testing engagement motives as the moderator, we find a significant and positive interaction, b = .07 (.02), p < .01, indicating that the relationship between OCCD and SMPIS is strengthened as individuals are increasingly motivated to use social media to engage in politics and social issues (Table 1, second column). Figure 2 plots this interactive effect using the Johnson–Neyman technique and shows that the effect increases in a linear fashion with the 95% confidence intervals crossing zero only at the lowest levels of engagement motives. This provides support for our second hypothesis that the relationship between OCCD and SMPIS is stronger for those who report greater motivations for using social media to engage in social issues and politics.

Figure 2.

Conditional Effect of Online Cross-cutting Discussion (OCCD) (W1) on Social Media Political Information Sharing (SMPIS) (W1) as a Function of Engagement Motives.

When relationship maintenance motives are analyzed as the moderator, we find a significant and positive interaction, b = .09 (.03), p < .01 (Table 2, second column). It is important to note that while the direct relationship between relationship maintenance motives and SMPIS is negative, the interaction term is positive. In other words, those who are more motivated to use social media to build and maintain relationships are in general less likely to share political information, but the relationship between OCCD on SMPIS is strengthened as individuals are increasingly motivated to use social media for relationship maintenance purposes. Figure 3 plots the interactive effect using the Johnson–Neyman technique and shows that the effects increase in a linear fashion with the 95% confidence intervals crossing zero for values of relationship maintenance motives at approximately the midpoint of the six-point scale. The moderating effect of relationship maintenance motives is similar to that of engagement motives, but appears to be limited to values above the midpoint of the scale. There is no moderating effect found for self-promotion motives, b = .02 (.02), p = n.s., suggesting that being motivated to use social media to promote one’s self has no influence on the relationship between OCCD and SMPIS. This addresses our first research question. In the absence of a significant interaction between OCCD and self-promotion motives, the remainder of our results omits the model where self-promotion is used as a moderator.

Table 2.

Summary of Regression Analyses for Moderated Mediation Model Using Relationship Maintenance SM Motives (W1) as a Moderator.

	SM political information sharing (W1) (mediator)		Offline political participation (W2) (criterion)
	b (SE)	b (SE)	b (SE)
SM political information sharing (W1)	–	–	.05 (.02)*
Online cross-cutting discussion (W1)	.35 (.05)***	−.14 (.19)	.01 (.03)
Relationship maintenance SM motives (W1)	−.08 (.03)**	−.20 (.05)***	–
Online cross-cutting discussion (W1) × relationship maintenance motives (W1)	–	.09 (.03)**	–
Engagement SM motives (W1)	.24 (.03)**	.23 (.03)***	−.01 (.02)
Self-promotion SM motives (W1)	−.03 (.03)	−.03 (.03)	.03 (.02)
Offline political participation (W1)	−.04 (.07)	−.02 (.07)	.69 (.04)***
Online political participation (W1)	.18 (.07)*	.19 (.07)**	.03 (.04)
SM personal information sharing (W1)	.18 (.03)**	.18 (.03)***	−.002 (.02)
SM political information receiving (W1)	.25 (.03)**	.25 (.03)***	−.01 (.02)
Age (W1)	.001 (.002)	.0003 (.002)	−.0003 (.001)
Sex (W1) (male = 1, female = 2)	−.23 (.07)**	−.24 (.07)***	.01 (.04)
Education (W1)	−.04 (.04)	−.03 (.04)	.03 (.02)
Political interest (W1)	.02 (.03)	.02 (.03)	.01 (.02)
Political self-efficacy (W1)	.05 (.03)^#	.05 (.03)^#	.04 (.02)*
Traditional media use (W1)	.07 (.04)^#	.07 (.04)^#	.01 (.02)
Strength of partisanship (W1)	.004 (.04)	.004 (.04)	−.02 (.03)
Constant	−.14 (.28)	.30 (.32)	.09 (.15)
Adjusted R²	.65	.66	.60
Residual SE	.79 (df = 578)	.78 (df = 577)	.45 (df = 578)
F statistic	74.57*** (df = 15; 578)	71.20*** (df = 16; 577)	59.35** (df = 15; 578)

SM: social media; SE: standard error.

Unstandardized coefficients reported. SEs in parentheses. N = 594. In models predicting SM political information sharing (W1), ΔR² for online cross-cutting discussion (W1) = .034, ΔR² for online cross-cutting discussion (W1) × relationship maintenance motive (W1) = .004 (all ps < .05). In model predicting offline political participation (W2), ΔR² for SM political information sharing (W1) = .003 (all ps < .05).

p < .1; *p < 0.05; **p < 0.01; ***p < 0.001 (two-tailed).

Figure 3.

Conditional Effect of Online Cross-cutting Discussion (OCCD) (W1) on Social Media Political Information Sharing (SMPIS) (W1) as a Function of Relationship Maintenance Motives.

We next examine whether SMPIS (W1) is positively related to offline political participation (W2) (H3).⁴ SMPIS (W1) is a positive predictor of offline political participation in both the model using engagement motives as a moderator, b = .04 (.02), p < .1, and the model using relationship maintenance motives as a moderator, b = .05 (.02), p < .05 (although the effect in the model using engagement motives as a moderator is only marginally significant). This provides mixed evidence for our third hypothesis, indicating that sharing political information on social media generally leads to increased offline political participation (see Tables 1 and 2, third column).⁵ This relationship persists despite controlling for offline political participation (W1), which is unsurprisingly a strong predictor across models, bs = .69–.70 (.02–.04), ps < .01.

Finally, we test two moderated mediation models using the two moderators, engagement motives and relationship maintenance motives, which were found to influence the relationship between OCCD and SMPIS. These tests determine whether OCCD (W1) indirectly influences offline political participation (W2) through SMPIS (W1) and whether this mediated effect varies depending on levels of each moderator (H4 and RQ2). In the first model using engagement motives as the moderator, a test of the unstandardized indirect effects was computed using 10,000 bootstrapping samples and bias-corrected 95% confidence intervals (Table 3). We find indirect effects at all levels of engagement motives, with point estimates of .07 (.006), at 1 SD below the mean, .011 (.007) at the mean and .016 (.009) at 1 SD above the mean. The indirect effects of OCCD on offline political participation through SMPIS are greater for individuals who are more motivated to use social media to engage with social issue and politics.

Table 3.

Conditional Indirect Effects of Online Cross-Cutting Discussion (OCCD) (W1) on Offline Political Participation (W2) Through SM Political Information Sharing (SMPIS) (W1) at Values of Engagement SM Motives.

Engagement SM motives	Point estimate	95% CI
1.27 (−1 SD)	.007 (.006)	.000 to .024
2.71 (Mean)	.011 (.007)	.001 to .028
4.15 (+1 SD)	.016 (.009)	.001 to .035

SM: social media; CI: confidence interval.

Path estimates are unstandardized coefficients. Indirect effects based on 10,000 bootstrapping samples with 95% bias-corrected confidence intervals.

All control variables included in the model. N = 594.

For the second model using relationship maintenance motives as the moderator, a test of the unstandardized indirect effects shows positive effects at all levels of relationship maintenance motives, with point estimates of .09 (.006) at 1 SD below the mean, .016 (.008) at the mean, and .020 (.011) at 1 SD above the mean (Table 4). Similar to the first model using engagement motives, the indirect effects of OCCD on offline political participation through SMPIS are greater for individuals who are more motivated to use social media for relationship maintenance purposes. These findings address our second research question, indicating that OCCD can lead to offline political participation indirectly through SMPIS and that this effect is moderated by both engagement and relationship maintenance motives (but not self-promotion motives).

Table 4.

Conditional Indirect Effects of Online Cross-cutting Discussion (OCCD) (W1) on Offline Political Participation (W2) through SM Political Information Sharing (SMPIS) (W1) at Values of Relationship Maintenance SM Motives.

Relationship Maintenance SM Motives	Point estimate	95% CI
3.67 (−1 SD)	.009 (.006)	.001 to .027
4.93 (Mean)	.016 (.008)	.001 to .034
6.00 (+1 SD)	.020 (.011)	.002 to .045

SD: standard deviation; SM: social media; CI: confidence interval.

Path estimates are unstandardized coefficients. Indirect effects based on 10,000 bootstrapping samples with 95% bias-corrected confidence intervals.

All control variables included in the model. N = 594.

Discussion

This study explores the possibility that in the increasingly contentious world of online political discussion, the unique affordances of social media provide an indirect pathway to traditional forms of offline political participation. Our results support the proposed theoretical model, demonstrating that increased levels of online cross-cutting discussion lead to increased offline political participation indirectly through the act of social media political information sharing. Furthermore, we observe that the association between OCCD and SMPIS is strengthened by both political engagement motives and relationship maintenance motives for social media use.

Reflecting the mixed findings regarding the outcomes of cross-cutting discussion, we find no direct link between OCCD and offline political participation (Gearhart & Zhang, 2015; Hampton et al., 2014). Instead, OCCD is associated with increased political information sharing. This relationship suggests that sharing is a behavior that may uniquely help individuals cope with the threats presented by political disagreement. As users select and consume political content in the process of sharing, they are provided an opportunity to expose themselves to pro-attitudinal information and receive positive feedback from politically like-minded contacts.

Another important contribution that this study makes is the finding that the relationship between OCCD and political information sharing is influenced by individuals’ motivations for using social media. Unsurprisingly, individuals who came to social media ready to jump into the political fray, turned more frequently to political information sharing when they encountered political disagreement. For users high in engagement motives, whom we may term “political junkies,” political engagement may be a key reason for using social media. Accordingly, “political junkies” are on the lookout for opportunities to engage in political conversation, and once engaged they are more likely to be affected by political disagreement. This finding is in line with prior research, which demonstrates that more politically involved individuals tend to experience greater dissonance when faced with challenges to their political beliefs and are more likely to engage in behaviors designed to reaffirm their own opinions (e.g., Taber & Lodge, 2006).

Yet our results indicate that other motivations for using social media can have an impact on the relationship between political disagreement and sharing. While users who came to social media to maintain relationships were not explicitly oriented toward engagement in politics, the presence of political disagreement provided an unexpected path to political expression. This may be due to the fact that those high in relationship maintenance motives are incentivized to engage with others in their network in a way that sustains connection. The social media literature highlights the role of relationship maintenance in the building of social capital (Tong & Walther, 2011) and suggests that users often take advantage of the affordances of social media to signal social support (Ellison et al., 2014). We speculate that users with such an outward social orientation could be more willing to listen to others of divergent political views and contribute to the conversation. Those with greater sensitivity to their social context may also experience increased dissonance when political disagreement arises due to the concern over the possibility of relational conflict. While sharing political information is, in general, less likely for those high in relationship maintenance motives, it may become a helpful way to deal with political difference. In this sense, social media sharing is not simply a megaphone for political expression but a method of engaging in social relationships (Gil de Zúñiga et al., 2014; Lee & Ma, 2012). This finding is particularly important given that in our data using social media for relationship maintenance purposes is a more commonly reported motivation for using social media than engagement in politics.

The finding that self-promotion motives did not have a moderating effect on our model suggests that there is a segment of users for which sharing is not an especially important strategy for dealing with political disagreement. The literature on rising levels of narcissism finds that those who are concerned with self-image are less likely to be civically engaged (Twenge, Keith, & Freeman, 2012). While we find no such negative influence of self-promotion motives, it does appear that those preoccupied with their own image may either attend less to conversations involving political disagreement or view sharing as too reputationally risky (Thorson, 2013).

Our results indicate a link between wave 1 SMPIS and wave 2 offline political participation, while controlling for wave 1 levels of offline participation. This adds further causal support to previous evidence of a relationship between online political expression and offline political action (Gil de Zúñiga et al., 2014). Individuals in our sample who engage in political information sharing likely become more engaged in the political debate, feel more empowered to participate, and are exposed to more opportunities to do so (Shah et al., 2012; Vaccari et al., 2015).

Finally, a test of our theoretical model finds that OCCD has an indirect influence on offline political participation through political information sharing. This mediated effect advances our theoretical understanding of how online disagreement can be translated into offline action and of how users’ individual orientation to online social contexts shape how they use the affordances of social media. The moderating role of both political engagement and relationship maintenance motives indicates that the uses and gratifications of social media users can play an important role in the translation of political activity from online to offline contexts (Lee & Ma, 2012). Contrary to work finding that cross-cutting discussion inhibits participation, our results suggest a prosocial effect of disagreement, in that exposure to disagreement can spur political action.

The modest effect sizes we find indicate that our model explains a small proportion of the variance in wave 2 offline political participation, which is expected given the other strong determinants of high-threshold political behavior. The effects we find may have larger implications at the population level. The panel data analyzed in this study add further strength to our causal claims by allowing us to examine the change in offline political participation over time while controlling for a variety of variables, including past political behavior and media use. Our study also has several important limitations. First, we rely on self-report measures to assess respondents’ online behavior. As with any survey, there are a variety of ways in which self-report data can be biased. This may be of particular concern in regards to our measures of social media motives, which ask respondents to report whether or not they used social media for a specific purpose. While we argue that these items are reasonable assessments of specific motivations, they are limited in that they do not ask respondents to directly reflect and report on their motivations. Across self-report measures, we have no reason to believe that a response bias exists that would explicitly affect the test of our theoretical model. Second, our measure of political information sharing does not capture format, content, or intended audience. Given the wide variety of motivations for sharing political information on social media (Hasell & Weeks, 2016; Lee & Ma, 2012), we believe more nuanced investigations of the sharing process are an important direction for future research. Finally, it is possible that political disagreement and expression co-occur and influence each other dynamically on social media. While we analyze OCCD as an independent variable in this model, future research should examine the possibility that political discussion and expression mutually reinforce each other and shape behavior over time.

Political life has become characterized by record levels of partisan polarization (Pew Research Center, 2016) and fears about the increasingly negative nature of political discourse on social media (e.g., Carr, 2016). There may be good reason to be concerned about the role the Internet plays in creating a fragmented and dysfunctional public sphere. However, social media also provides unique social contexts and affordances that give us hope that online political disagreement can in fact prove productive. Our results are particularly encouraging in that even those who use social media to maintain relationships can find themselves more likely to engage in political expression and participate offline. Future work in this area should continue to explore such indirect pathways and incorporate political as well as nonpolitical social media motivations into accounts of how online disagreement can lead to offline action.

Footnotes

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) received no financial support for the research, authorship, and/or publication of this article.

Notes

Author Biographies

Daniel S. Lane (BS, Television, Radio & Film, Syracuse University) is a doctoral candidate in the Department of Communication Studies at the University of Michigan. His research focuses on the ways in which communication on social media can leverage identities and social connections to promote participation in political and social change.

Dam Hee Kim (MA, Journalism & Mass Communication, Korea University) is a doctoral candidate of Communication Studies at the University of Michigan. Her field of research focuses on political communication, with a particular emphasis on the issue of media diversity and democratic citizenship.

Slgi S. Lee (MA, Journalism & Mass Communication, Sung-kyunkwan University) is a doctoral student in the department of Communication Studies at the University of Michigan. Her research interests center on the roles of new media and the Internet in facilitating deliberative democracy.

Brian E. Weeks (PhD, Communication, The Ohio State University) is an assistant professor in the Department of Communication Studies and a Faculty Associate in the Center for Political Studies at the University of Michigan. His research interests include political misinformation and misperceptions, news on social media, and political information sharing online.

Nojin Kwak (PhD, Mass Communication, University of Wisconsin-Madison) is a professor in the Department of Communication Studies and the director of the Nam Center for Korean Studies at the University of Michigan. His research centers on the role of communication media in civic and political engagement.

References

Quercia

Cha

Gummadi

Crowcroft

(2014). Sharing political news: the balancing act of intimacy and socialization in selective exposure. EPJ Data Science, 3, 1–21. doi:10.1140/epjds/s13688-014-0012-2

Bakshy

Messing

Adamic

L. A.

(2015). Exposure to ideologically diverse news and opinion on Facebook. Science, 348, 1130–1132. doi:10.1126/science.aaa1160

Boulianne

(2009). Does internet use affect engagement? A meta-analysis of research. Political Communication, 26, 193–211. doi:10.1080/10584600902854363

Boulianne

(2015). Social media use and participation: A meta-analysis of current research. Information, Communication, & Society, 18, 524-538. doi:10.1080/1369118x.2015.1008542

boyd

Golder

Lotan

(2010, January 5–8). Tweet, tweet, retweet: Conversational aspects of retweeting on Twitter. 2010 43rd Hawaii International Conference on System Sciences, Honolulu, HI. doi:10.1109/HICSS.2010.412

Brundidge

. (2010). Encountering “Difference” in the contemporary public sphere: The contribution of the internet to the heterogeneity of political discussion networks. Journal of Communication, 60(4), 680–700. doi:10.1111/j.1460-2466.2010.01509.x

Campbell

S. W.

Kwak

(2010). Mobile communication and civic life: Linking patterns of use to civic and political engagement. Journal of Communication, 60, 536–555. doi:10.1111/j.1460-2466.2010.01496.x

Caprara

G. V.

Vecchione

Capanna

Mebane

(2009). Perceived political self-efficacy: Theory, assessment, and applications. European Journal of Social Psychology, 39, 1002–1020. doi: 10.1002/ejsp.604

Carr

. (2015, September 2). How social media is ruining politics. Retrieved from http://politi.co/1JLtKs8

10.

Chadwick

(2009). Web 2.0: New challenges for the study of e-democracy in an era of informational exuberance. I/S: A Journal of Law and Policy for the Information Society, 5, 9–41.

11.

Chambers

. (2003). Deliberative democratic theory. Annual Review of Political Science, 6(1), 307–326. doi:10.1146/annurev.polisci.6.121901.085538

12.

Cho

Ahmed

Keum

Choi

Y. J.

Lee

J. H.

(2016). Influencing myself: Self-reinforcement through online political expression. Communication Research, 93650216644020. doi:10.1177/0093650216644020

13.

Coe

Kenski

Rains

S. A.

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

14.

Darlington

R. B.

Hayes

A. F.

(2016). Regression analysis and linear models: Concepts, applications, and implementation. New York, NY: Guilford.

15.

Donsbach

Mothes

(2013). The dissonant self: Contributions from dissonance theory to a new agenda for studying political communication. Annals of the International Communication Association, 36, 3–44. doi:10.1080/23808985.2013.11679124

16.

Ellison

N. B.

Heino

Gibbs

(2006). Managing impressions online: Self-presentation processes in the online dating environment. Journal of Computer-Mediated Communication, 11, 415–441. doi:10.1111/j.1083-6101.2006.00020.x

17.

Ellison

N. B.

Vitak

Gray

Lampe

(2014). Cultivating social resources on social network sites: Facebook relationship maintenance behaviors and their role in social capital processes. Journal of Computer-Mediated Communication, 19, 855–870. doi:10.1111/jcc4.12078

18.

Eveland

W. P.

Thomson

(2006). Is it talking, thinking, or both? A lagged dependent variable model of discussion effects on political knowledge. Journal of Communication, 563, 523–542. doi:10.1111/j.1460-2466.2006.00299.x

19.

Festinger

(1962). A theory of cognitive dissonance. Stanford, CA: Stanford University Press.

20.

Gearhart

Zhang

(2015). “Was it something I said?” “No, it was something you posted!” A study of the spiral of silence theory in social media contexts. Cyberpsychology, Behavior, and Social Networking, 18, 208–213. doi:10.1089/cyber.2014.0443

21.

Gentile

Twenge

J. M.

Freeman

E. C.

Campbell

W. K.

(2012). The effect of social networking websites on positive self-views: An experimental investigation. Computers in Human Behavior, 28, 1929–1933. doi:10.1016/j.chb.2012.05.012

22.

Gil

Zúñiga

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

23.

Gil

Zúñiga

Molyneux

Zheng

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

24.

Gladwell

(2010, October 4). Small change: Why the revolution will not be tweeted. The New Yorker, pp. 42–49.

25.

Gottfried

Barthel

Shearer

Mitchell

(2016, February 4). The 2016 presidential campaign—A news event that’s hard to miss. Pew Research Center’s Journalism Project. Retrieved from http://www.journalism.org/2016/02/04/the-2016-presidential-campaign-a-news-event-thats-hard-to-miss/

26.

Green

D. P.

Palmquist

Schickler

(2004). Partisan hearts and minds: Political parties and the social identities of voters. New Haven, CT: Yale University Press.

27.

Greene

(1999). Understanding party identification: A social identity approach. Political Psychology, 20, 393–403. doi:10.1111/0162-895X.00150

28.

Hampton

K. N.

Rainie

Dwyer

Shin

Purcell

(2014). Social media and the “spiral of silence.” Washington, DC: Pew Research Center.

29.

Hasell

Weeks

B. E.

(2016). Partisan provocation: The role of partisan news use and emotional responses in political information sharing in social media. Human Communication Research, 42, 641–661. doi:10.1111/hcre.12092

30.

Hayes

A. F.

(2013). Introduction to mediation, moderation, and conditional process analysis: A regression-based approach. New York, NY: Guilford Press.

31.

Himelboim

McCreery

Smith

(2013). Birds of a feather tweet together: Integrating network and content analyses to examine cross-ideology exposure on Twitter. Journal of Computer-Mediated Communication, 18, 40–60. doi:10.1111/jcc4.12001

32.

Kim

Hsu

S.-H.

de Zúñiga

H. G.

(2013). Influence of social media use on discussion network heterogeneity and civic engagement: The moderating role of personality traits. Journal of Communication, 63, 498–516. doi:10.1111/jcom.12034

33.

Knobloch-Westerwick

Meng

(2011). Reinforcement of the political self through selective exposure to political messages. Journal of Communication, 61, 349–368. doi:10.1111/j.1460-2466.2011.01543.x

34.

Kümpel

A. S.

Karnowski

Keyling

. (2015). News sharing in social media: A review of current research on news sharing users, content, and networks. Social Media + Society, 1(2), doi:10.1177/2056305115610141

35.

Kwak

Williams

A. E.

Wang

Lee

(2005). Talking politics and engaging politics: An examination of the interactive relationships between structural features of political talk and discussion engagement. Communication Research, 32, 87–111. doi:10.1177/0093650204271400

36.

Lampinen

. (2015). Deceptively simple: Unpacking the notion of “sharing.” Social Media + Society, 1(1), doi:10.1177/2056305115578135

37.

Lee

C. S.

(2012). News sharing in social media: The effect of gratifications and prior experience. Computers in Human Behavior, 28, 331–339. doi:10.1016/j.chb.2011.10.002

38.

Marwick

A. E.

boyd

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

39.

Miller

C. C

. (2014, August 26). How social media silences debate. The New York Times. Retrieved from http://www.nytimes.com/2014/08/27/upshot/how-social-media-silences-debate.html

40.

Mutz

D. C.

(2002a). The consequences of cross-cutting networks for political participation. American Journal of Political Science, 46, 838–855. doi:10.2307/3088437

41.

Mutz

D. C.

(2002b). Cross-cutting social networks: Testing democratic theory in practice. American Political Science Review, 96, 111–126. doi:10.1017/S0003055402004264

42.

Mutz

D. C.

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

43.

Noelle-Neumann

(1991). The theory of public opinion: The concept of the spiral of silence. Annals of the International Communication Association, 14, 256–287. doi:10.1080/23808985.1991.11678790

44.

Papacharissi

(2010). A networked self: Identity, community, and culture on social network sites. New York, NY: Routledge.

45.

Park

Kee

K. F.

Valenzuela

(2009). Being immersed in social networking environment: Facebook groups, uses and gratifications, and social outcomes. Cyberpsychology & Behavior, 12, 729–733. doi:10.1089/cpb.2009.0003

46.

Pew Research Center. (2012). Social networking sites and politics. Retrieved from http://www.pewinternet.org/2012/03/12/social-networking-sites-and-politics/

47.

Pew Research Center. (2016). Partisanship and Political Animosity in 2016. Retrieved from http://www.people-press.org/2016/06/22/partisanship-and-political-animosity-in-2016/

48.

Pingree

R. J.

(2007). How messages affect their senders: A more general model of message effects and implications for deliberation. Communication Theory, 17, 439–461. doi:10.1111/j.1468-2885.2007.00306.x

49.

Price

Cappella

J. N.

Nir

(2002). Does disagreement contribute to more deliberative opinion? Political Communication, 19, 95–112. doi:10.1080/105846002317246506

50.

Putnam

R. D.

(2000). Bowling alone: The collapse and revival of American community. New York, NY: Simon & Schuster.

51.

Quan-Haase

Young

A. L.

(2010). Uses and gratifications of social media: A comparison of Facebook and Instant messaging. Bulletin of Science, Technology & Society, 30, 350–361. doi:10.1177/0270467610380009

52.

Rojas

Puig-i-Abril

(2009). Mobilizers mobilized: Information, expression, mobilization, and participation in the digital age. Journal of Computer-Mediated Communication, 14, 902–927. doi:10.1111/j.1083-6101.2009.01475.x

53.

Scheufele

D. A.

Nisbet

M. C.

Brossard

Nisbet

E. C.

(2004). Social structure and citizenship: Examining the impacts of social setting, network heterogeneity, and informational variables on political participation. Political Communication, 21, 315–338. doi:10.1080/10584600490481389

54.

Scheufele

D. A.

Hardy

B. W.

Brossard

Waismel-Manor

I. S.

Nisbet

. (2006). Democracy based on difference: Examining the links between structural heterogeneity, heterogeneity of discussion networks, and democratic citizenship. Journal of Communication, 56(4), 728–753. doi:10.1111/j.1460-2466.2006.00317.x

55.

Shah

D. V.

Friedland

L. A.

Wells

Kim

Y. M.

Rojas

Bennett

W. L

. (2012). The personalization of politics: Political identity, social media, and changing patterns of participation. The ANNALS of the American Academy of Political and Social Science, 644(1), 20–39. doi:10.1177/0002716212451428

56.

Skoric

M. M.

Zhu

Goh

Pang

(2016). Social media and citizen engagement: A meta-analytic review. New Media & Society, 18, 1817–1839. doi:10.1177/1461444815616221

57.

Tabachnick

B. G.

Fidell

L. S.

(2012). Using multivariate statistics (6th ed.). Boston, MA: Pearson.

58.

Taber

C. S.

Lodge

. (2006). Motivated skepticism in the evaluation of political beliefs. American Journal of Political Science, 50(3), 755–769. doi:10.1111/j.1540-5907.2006.00214.x

59.

Thorson

(2013). Facing an uncertain reception: young citizens and political interaction on Facebook. Information, Communication & Society, 17, 203–216. doi:10.1080/1369118X.2013.862563

60.

Tong

Walther

J. B.

(2011). Relational maintenance and CMC. In Wright

K. B.

Webb

L. M.

(Eds.), Computer-mediated communication in personal relationships (pp. 98–118). New York, NY: Peter Lang.

61.

Twenge

J. M.

Keith

Freeman

E. C.

(2012). Generational differences in young adults’ life goals, concern for others, and civic orientation, 1966–2009. Journal of Personality and Social Psychology, 102, 1045–1062. doi:10.1037/a0027408

62.

Vaccari

Valeriani

Barberá

Bonneau

Jost

J. T.

Nagler

. . . Tucker

J. A.

(2015). Political expression and action on social media: Exploring the relationship between lower- and higher-threshold political activities among twitter users in Italy. Journal of Computer-Mediated Communication, 20, 221–239. doi:10.1111/jcc4.12108

63.

Valenzuela

Kim

Gil de Zúñiga

(2012). Social networks that matter: Exploring the role of political discussion for online political participation. International Journal of Public Opinion Research, 24, 163–184. doi:10.1093/ijpor/edr037

64.

Weeks

B. E.

Holbert

R. L.

(2013). Predicting dissemination of news content on social media: A focus on reception, friending, and partisanship. Journalism & Mass Communication Quarterly, 90, 212–232. doi:10.1177/1077699013482906

65.

Wojcieszak

M. E.

Mutz

D. C.

(2009). Online groups and political discourse: Do online discussion spaces facilitate exposure to political disagreement? Journal of Communication, 59, 40–56. doi:10.1111/j.1460-2466.2008.01403.x

66.

Zhang

Leung

(2014). A review of social networking service (SNS) research in communication journals from 2006 to 2011. New Media & Society, 17, 1007–1024. doi:10.1177/1461444813520477

From Online Disagreement to Offline Action: How Diverse Motivations for Using Social Media Can Increase Political Information Sharing and Catalyze Offline Political Participation

Abstract

Keywords

Is Political Disagreement Good for Democracy?

Political Information Sharing as a Response to Political Disagreement

The Role of Motivations for Using Social Media

Political Information Sharing and Offline Political Participation

From Disagreement to Action Through Sharing: A Theoretical Model

Method

Sample

Measures

Online Cross-cutting Discussion

Social Media Political Information Sharing

Offline Political Participation

Social Media Motives

Control Variables

Results

Discussion

Footnotes

Declaration of Conflicting Interests

Funding

Notes

Author Biographies

References