Facebook Use and Political Participation

Abstract

Some recent studies have illustrated a positive relationship between social media use and political participation among young people. Researchers, however, have operationalized social media usage differently. This article adopts a multidimensional approach to the study of the impact of social media. Focusing on Facebook (FB), the most widely utilized social networking site in Hong Kong, this study examines how time spent on FB, exposure to shared political information, network size, network structural heterogeneity, and direct connection with public political actors relate to young people’s online and offline political participation. Analysis of a survey of university students (N = 774) shows that participation is explained most prominently by direct connection with public political actors, followed by exposure to shared political information. These two variables also mediate the impact of other dimensions of FB use on political participation.

Keywords

social media political participation multidimensional approach network structural heterogeneity network size

Introduction

Since its founding in 2006, Facebook (FB) has grown tremendously from being a high school and college network to being one of the most globally popular social networking sites (SNSs). In the process, it has also become a platform for a wide range of political activities (e.g., Earl & Kimport, 2011; Ward, 2012). At the individual level, some researchers have found a generally positive relationship between social media use and civic and political participation (e.g., Gil de Zúñiga, Jung, & Valenzuela, 2012; Gil de Zúñiga, Puig-I-Abril, & Rojas, 2009; Valenzuela, Park, & Kee, 2009). However, there are also studies coming up with mixed or null findings (e.g., Kushin & Yamamoto, 2010; Vitak et al., 2010; Zhang, Johnson, Seltzer, & Bichard, 2010).

One reason for the differences in research findings on the relationship between social media use and political participation is the lack of a commonly accepted conceptualization and operationalization of FB/SNSs usage. Unlike conventional mass media, “using” FB or other SNSs does not refer merely to exposure to content, it also refers to the active cultivation of networks, the construction of personal profiles, consumption and/or production of shared materials, and commenting on or simply “liking” other people’s postings. FB usage, therefore, can vary along many dimensions. Paying attention to the various aspects of social media use is crucial to understanding the impact of social media.

Based on this premise, this study adopts a multidimensional approach to examining the impact of FB use on political participation among university students in Hong Kong. It aims at contributing to the literature by illustrating the independent impact of various aspects of social media use and how the impact of the various aspects may relate to each other.

Social Media, Political Participation, and Dimensionality of FB Use

Even before social media appeared, scholars have been debating about the Internet’s impact on citizens’ civic and political participation since the Internet’s popularization in the 1990s. The optimists argued that the Internet lowers information costs, enhances people’s sense of efficacy by its interactivity (Chadwick, 2006), and facilitates online mobilization (Earl & Kimport, 2011). The skeptics, in contrast, argued that the Internet presents a “high choice environment” that generates higher levels of audience selectivity (Bennett & Iyengar, 2008). It facilitates the political junkies to stay even closer to politics, but it also allows the uninterested to stay further away. The result is therefore the widening of the existing knowledge and participation gaps among groups of citizens (Davis, 1999; Prior, 2007).

However, some scholars have argued that people’s tendency to consume certain materials based on existing interests and attitudes does not rule out occasional, sometimes accidental, exposure to other materials (Lupia & Philpot, 2005; Mitchelstein & Boczkowski, 2010). The rise of social media has arguably further contributed to the phenomenon of accidental or unintentional exposure to public affairs content because such content are often “pushed” to people by their acquaintances. Although people having strong ties with each other are likely to share similar interests and attitudes, individuals’ social media networks often contain “weak ties” that play important roles in information dissemination (Gil de Zúñiga & Valenzuela, 2011). Politically apathetic individuals can still be connected via social media to acquaintances who are highly interested in public affairs, and the latter could become sources of occasional political information and messages.

Empirically, several studies have indeed found a generally positive relationship between social media use and civic/political engagement. However, a closer look into the studies would show that researchers have been using different independent variables in their analyses, and the findings may differ depending on the variables employed. Baumgartner and Morris (2010) adopted the conventional approach by measuring exposure to SNSs in terms of days per week. They found that exposure to SNSs related significantly to political expression online, but not to offline political participation such as voting, signing petitions, or calling a politician. Similarly, Gil de Zúñiga, Puig-I-Abril, and Rojas (2009) measured blog usage by asking respondents whether they had “read blogs” and “created a blog that other people can read.” They found that blog use related to online but not offline political participation.

Gil de Zúñiga, Jung, and Valenzuela (2012) also found that general SNS use, measured in time per day, did not relate to social capital, civic participation, online political participation, and offline political participation. What explains participation in Gil de Zúñiga et al.’s (2012) analysis are SNS use for news (i.e., the extent to which people regarded SNS as helpful to keeping them informed about current affairs and the local community) and online network size (i.e., individuals’ estimation of the number of people they talked to via the Internet).

Gil de Zúñiga et al.’s (2012) findings are very understandable. People can use social media for a wide range of purposes. Frequencies of usage per se, hence, may not enhance political participation. Rather, usage specifically related to politics or public affairs is more likely to promote participation. Park, Kee, and Valenzuela (2009) have shown a similar pattern of findings. Besides, social media is by definition places where people connect with others. The number and characteristics of these connections are likely to matter. These ideas underline Ellison, Steinfield, and Lampe’s (2007) attempt to develop the FB intensity index. The index contains items about time spent, number of friends, and attitudes and feelings toward FB. Ellison et al. found that FB intensity significantly relates to amount of bridging and bonding social capital, which are generally considered as antecedents of civic and political participation. Later studies have employed the index (or similarly constructed indices) and more directly demonstrated the significant relationships between FB use and political participation (Valenzuela et al., 2009; Vitak et al., 2010).

The FB intensity index allows researchers to take into account multiple aspects of FB usage while maintaining the parsimony of the analysis. The overall composite index approach is certainly useful for some purposes. Yet, it also has several limitations. First, while the items constituting Ellison et al.’s (2007) index had high levels of interitem reliability, there are inevitably aspects of social media use that were omitted. It is not sure if the omitted aspects, such as network heterogeneity, would be highly correlated to the items constituting the index. If not, it means that the omitted aspects may need to be treated separately. Second, a differentiated approach, that is, to separate the different aspects of usage from each other, would allow one to discern which aspects of social media use actually matter with regard to what outcomes. Third, a differentiated approach would also allow researchers to examine how one aspect of usage may mediate the impact of other aspects.

This study, therefore, adopts the differentiated approach. Specifically, it focuses on FB usage and examines time spent, exposure to shared political information, network size, network structural heterogeneity, and connection with public political actors. Of course, these five do not capture all possible dimensions of FB usage, but they cover sheer usage frequencies, political communication, and network characteristics.

This study hypothesizes that four of the five aspects would relate directly and positively to political participation. As pointed out earlier, several studies have shown a null relationship between sheer social media usage frequencies and offline political participation (Baumgartner & Morris, 2010; Gil de Zúñiga et al., 2012). While people may be accidentally exposed to political materials on FB through frequent usage, exposure to political information, instead of time spent, should be the variable that relates directly to participation. Therefore, we do not set up a main effect hypothesis for time spent, but a hypothesis on the impact of exposure to shared political information on participation. This expectation is consistent with existing studies’ findings (Gil de Zúñiga et al., 2012; Rojas & Puig-i-Abril, 2009).

Gil de Zúñiga et al. (2012) found that online network size significantly predicts political participation. When a person connects with more people, the range and amount of political materials the person receives is likely to be larger. There should also be a higher likelihood for the person to encounter a call for action.

Nevertheless, size is not the only network characteristic relevant to participation. This study examines two other aspects of social networks that can influence political participation. The first is network structural heterogeneity. Communication researchers have shown that people embedded in more structurally heterogeneous networks are more likely to participate in community and political affairs (e.g., Scheufele, Nisbet, Brossard, & Nisbet, 2004).¹ This is because people located in such networks tend to be exposed to a wider range of information and viewpoints, and they are likely to be more integrated into their local community. We believe that these arguments about the impact of heterogeneity of offline networks are also applicable to the impact of online social networks.

Second, FB may generate participation by linking people to a range of public figures in the field of politics. We label these people “public political actors.” They include politicians in representative bodies, officials in the executive branch of the government, prominent grassroots movement activists, and academic and media commentators. Young people may become “friends” and/or develop other linkages with them through FB. Admittedly, the exact meanings of “being a friend” with a public political actor may vary from case to case. Sometimes, adding a public political actor as a friend may entail very little interaction. The relationship may be largely “parasocial.” Yet in other cases, grassroots activists or media commentators may indeed interact with their FB “friends” regularly. But no matter how much and what type of interaction there is, the FB accounts of public political actors, generally speaking, are likely to be important sources of political information and viewpoints. Some of the public political actors, such as politicians, activists, and even commentators, may also directly mobilize citizens to act via social media. Connections with these accounts, therefore, should enhance participation.

In summary, the four main effect hypotheses of FB use are as follows:

Hypothesis 1: Exposure to shared political information via FB relates positively to participation.

Hypothesis 2: FB network size relates positively to participation.

Hypothesis 3: FB network structural heterogeneity relates positively to participation.

Hypothesis 4: FB connection with public political actors relates positively to participation.

Besides the main effects, this study is also interested in how some aspects of social media use may mediate the effects of other aspects of usage. Specifically, network structural heterogeneity and connection with public political actors may mediate the impact of network size on participation. Individuals are likely to be most closely connected with similar others. However, when a network expands, the chance is that the network would also become more heterogeneous. Similarly, public political actors are unlikely to be among the first group of “friends” that ordinary people connect with via FB. People are more likely to connect with public political actors only when their networks expand. These arguments imply that network size may enhance participation through enhancing network structural heterogeneity and the chance of having connections with public political actors. This constitutes Hypothesis 5:

Hypothesis 5: FB network structural heterogeneity and direct connection public political actors via FB mediate the impact of FB network size on political participation.

In addition, exposure to shared political information via FB may mediate the impact of the three network characteristics examined. As argued earlier, network size, structural heterogeneity, and connection with public political actors could all enhance the chance of people encountering political information on FB, which in turn leads to participation. Finally, although sheer time spent is not hypothesized to affect participation directly, spending more time on FB may heighten the chance of people encountering all kinds of information and materials, including political materials, on the platform. Therefore, time spent may have an indirect impact on participation through exposure to shared political information. The above arguments lead to Hypotheses 6 and 7:

Hypothesis 6: Exposure to shared political information mediates the impact of network characteristics on participation.

Hypothesis 7: Time spent on FB has an indirect relationship with participation via exposure to shared political information.

Background, Method, and Data

This study reports on an exploratory study of Hong Kong university students’ FB usage and political behavior. FB is the most popular SNSs in the city. According to the commercial research company Socialbakers, there are 4.07 million FB users in Hong Kong by 2012.² The survey was conducted between March 19 and 23, 2012. Given the lack of a comprehensive sampling frame, sampling proceeded by first selecting four universities randomly of the nine established universities in the city. Within each university, research assistants distributed questionnaires at the two most important “public places” (the main library and the main student canteen). The questionnaires were distributed in an early afternoon and an early evening time slot on each of the 5 days. The design gave rise to 80 distribution time points. The assistants were instructed to obtain 10 interviews from each time point by walking along a predetermined path and invite every 10th student to participate. Respondents filled out the questionnaires by themselves. Despite being nonprobabilistic, the design tries to enhance the representativeness of the sample through spreading the time and place of participant recruitment and procedures minimizing interviewers’ selection biases. The sampling design aimed at getting 800 respondents, but 26 questionnaires were incomplete, leaving the final sample size at 774.

The key variables for the analysis are operationalized as follows:

Political participation: The survey asked the respondents whether they had participated in the year before the survey in 13 forms of political activities, including 6 online activities and 7 offline activities. The 13 items included the major forms of political activities such as petitioning, protests, contacting political representatives and/or officials, and volunteering for political organizations. Voting was not included though, since many of the respondents had not reached the voting age in the last major election in Hong Kong. The respondents simply stated Yes or No for each activity. A respondent’s score on online participation is the number of yes responses he or she gave to the 6 online participation items (M = 0.93, SD = 1.45, α = .76), and his or her offline participation score is the number of “yeses” given to the 7 offline participation items (M = 1.16, SD = 1.63, α = .76).

Time spent was measured by asking the respondents to report on the amount of time they spent on FB daily using a 0–6 scale, with 0 meaning not using at all, 1 meaning 1–30 min, and 6 meaning 151 min or above (M = 3.00, SD = 1.71).

Exposure to shared political information was the average of respondents’ estimation, registered by a 4-point scale (0 = never, 3 = very frequently), of how frequently their FB friends shared “information or commentaries on public affairs” and “information or commentaries on policy and political issues” (α = .92, r = .85. M = 1.63, SD = 0.82).

Network size was measured by asking the respondents to state the rough number of friends linked to their FB accounts. Respondents who had more than one FB account were asked to report the number for their most frequently used accounts. We capped the variable at 1,000 to resolve the problem of having outliers who have extraordinarily large number of friends. The resulting variable had a mean of 458.18 (SD = 274.36).

Network structural heterogeneity was measured by three questions. The first question asks the respondents whom they would include as FB friends. The answers are (1) only acquaintances, (2) people whom one has met only once or twice, and (3) even strangers. The second question asks the respondents to choose one of the three statements to describe their FB friends: (1) most of them are of similar background, (2) many of them are from different backgrounds, but still a significant portion are from similar background, and (3) they come from a variety of backgrounds. The third question asks the respondents to estimate the proportion of their FB friends who were of similar age. The answers were 1 = 0–20%, 2 = 21–40%, 3 = 41–60%, 4 = 61–80%, and 5 = 81–100%. The three questions did not require the respondents to have precise assessments of their networks’ characteristics. The respondents should be capable of generating estimations in response to the questions. The answers of the three questions were standardized and then averaged to derive the index of network structural heterogeneity (α = .37, M = 0.0003, SD = .67).³

Connection with public political actors was measured by asking the respondents if their FB friends included (1) District or Legislative Councilors, (2) social movement activists, (3) media commentators, (4) academics, and (5) government officials.⁴ On each item, the respondents could indicate none, one to two, or a few or more. The answers were summed up to form a 0–10 index (M = 1.66, SD = 2.13, α = .76).

Control variables include three demographics (gender, semester at school, socioeconomic status), political knowledge (measured by 12 factual knowledge items), internal, external, and collective efficacy (each measured by agreement with two 5-point Likert-type scaled statements),⁵ and support for democratization (measured by agreement with two 5-point Likert-type scaled statements).

Findings and Analysis

The analysis began by examining the bivariate relationships among the various aspects of FB usage. The results show that the five dimensions of FB usage generally correlate with each other positively and significantly. Only two correlations are statistically insignificant—that between network structural heterogeneity and time spent and that between network structural heterogeneity and exposure to shared political information. However, the sizes of the correlations are not overwhelmingly strong. The largest correlation coefficient—for the relationship between connection with public political actors and exposure to shared political information—is only 0.308. The lack of stronger correlations among the dimensions of FB usage supports the argument for a differentiated approach.

Multiple regression was then conducted to examine the main effects Hypotheses 1 to 4, which state that exposure to shared political information, network size, network structural heterogeneity, and connections with public political actors affect participation. The independent variables included all the controls and the five aspects of FB usage. Online and offline participation are used as two separate dependent variables.

For the present purpose, the important results are whether the various aspects of FB usage significantly explain participation in the full model. The results support Hypothesis 1, as exposure to shared political information has a significant impact on both online and offline participation (β = .122 and .083, p < .001 and .01, respectively). The results also support Hypothesis 4. In fact, connection with public political actors is very strongly related to both online and offline participation (β = .349 and .401, respectively, p < .001 in both cases).

The findings partially support Hypothesis 3. Network structural heterogeneity significantly predicts offline, but not online, participation (β = .082 for offline participation, p < .01). Hypothesis 2 is not supported at all, however. Network size does not relate significantly to offline and online participation in the full model. Moreover, consistent with our argument, sheer time spent on FB also does not significantly relate to both offline and online participation.

We turned to structural equation modeling to test the mediating effects hypotheses. We construct the initial model based on the main effect hypotheses and Hypotheses 5 to 7. Offline and online participation were used as separate dependent variables. The results on the initial model show that network size does not have direct effects on exposure to shared political information and both types of participation. Time spent also does not have direct effects on participation. These findings are consistent with the regression results. They simply mean that there is no evidence supporting the existence of those relationships. As the relationships do not exist, the parameters can be removed to simplify the model and potentially improve the model’s goodness of fit. Nevertheless, in order to keep the model for online and offline participation consistent, we remove a parameter only if, in the initial analysis, it is insignificant in both the model for online participation and that for offline participation.

Figure 1 summarizes the results of the revised model. The goodness of fit of the model still falls slightly short of being optimal for both online and offline participation (root mean square error of approximation = .074, .072, respectively; comparative fit index = .938, .949, respectively; χ² = 31.58, 30.24, respectively, p < .05 for both). Nevertheless, for the present purpose, the more important findings are the significance of the mediating relationships. As shown in Figure 1, network size directly and substantially leads to network structural heterogeneity and connection with public political actors, which in turn relate significantly to offline participation, though only connection with public political actors leads significantly to online participation. Sobel test calculations confirm that connection with public political actors significantly mediates the impact of network size on both types of participation (z = 96.90 and 110.34, respectively, p < .001 in both cases). Network structural heterogeneity also significantly mediates the impact of network size on offline participation (z = 10.09, p < .001). Hypothesis 5 is largely supported.

Figure 1.

Structural equation model of the impact of Facebook use on political participation.

Hypothesis 6, which hypothesizes that the three network characteristics would indirectly affect participation through exposure to shared political information, receives weaker support. First, as stated above, network size simply does not relate significantly to exposure to shared political information. Second, for both online and offline participation, the associations between network structural heterogeneity and exposure to shared information are significant but negative in sign (β = −.084, p < .05 in both cases), which contradict our expectations. Nevertheless, connection with public political actors relates positively to exposure to shared political information. The Sobel test calculations confirm that exposure to shared political information mediates part of the effect of connections with public political actors on both online and offline participation (z = 12.95 and 11.69, respectively, p < .001 in both cases).

Finally, time spent on FB relates significantly to exposure to shared political information, which in turn affects both online and offline participation. The Sobel test confirms the indirect effect of time spent on FB on political participation through exposure to shared political information is statistically significant (z = score 5.33 and 5.23 for online and offline participation, respectively, p < .001 in both cases). Hypothesis 7 is supported.

Discussion

The analysis confirms most of the hypotheses. When main effects are concerned, respondents with more structurally heterogeneous FB networks, who had connections with public political actors through FB, and who were exposed to shared political information through FB were more likely to have participated in political activities. Respondents who had larger FB networks also participated in politics more frequently, though the impact of network size is completely mediated by network structural heterogeneity and connection with public political actors. Moreover, exposure to shared political information also mediates part of the impact of connection with public political actors on participation. Besides, sheer time spent on FB also has a significant indirect effect on participation through exposure to shared political information.

While some of the results are replicating existing findings, this article contributes to the study of the impact of social media on participation by differentiating among a number of dimensions of FB use. As discussed in the theoretical section, some researchers have developed the composite index of FB intensity in order to capture the complexity of FB usage (Ellison, Steinfield, & Lampe, 2007). While the composite index approach is parsimonious and would be a good choice if a study is concerned only with the overall impact of social media use, this study shows that a differentiated approach can give us insights into not only whether but also how social media use matters. Take network size as the example. Past research has shown that network size significantly predicts political participation (Gil de Zúñiga et al., 2012). This study shows that the impact of network size resides in how it leads to network structural heterogeneity and connections with public political actors. In other words, simply connecting to more people may not result in higher levels of participation. Network size matters only when changes in quantity lead to changes in quality, that is, when the increasing number of connections leads to certain changes in network composition.

Network structural heterogeneity is the center of a body of research in the broader political communication literature (e.g., Scheufele et al., 2004). This study confirms the relevance of the notion to the study of online networks and the impact of social media on participation. As expected, respondents who were embedded in more structurally heterogeneous online networks were more likely to engage in offline political activities. Nevertheless, there is an unexpected negative relationship between FB network structural heterogeneity and exposure to shared political information. One plausible reason is linked to the fact that this study examines university students. Being young and educated, these people are likely to be those most active in sharing political information through SNSs. In this case, university students who are mostly connected to similar others (i.e., other university students) would then be more exposed to political information through FB.

Connection with public political actors is a variable seldom examined in the extant literature. This study shows that the influence of this variable is remarkable. We contend that having connections with people such as activists and political representatives via social media would typically result in exposure to a large amount of political information, persuasive messages, or even direct mobilization attempts. The findings thus point to an important way through which social media can influence people politically—through facilitating people to build weak ties with figures within the political arena. In fact, in the present survey, 25.8%, 37.6%, and 18.4% of the university students had at least one political representative, one movement activist, and one media commentator as their FB friends, respectively. These percentages are substantial. The high percentages may be partly due to the social and political environment of Hong Kong: Connections with activists and politicians are quite easy to establish for young people in such a small and densely populated city, and such connections are seemingly having tremendous influence of young people’s political behavior.

The relationship between participation and sheer time spent on FB is also worth discussing. In a multiple regression, time spent does not relate to participation. This finding is consistent with past studies (e.g., Baumgartner & Morris, 2010; Gil de Zúñiga et al., 2012). Nevertheless, the structural equation modeling analysis shows that time spent on FB does have a significant relationship with exposure to shared political information. This relationship is consistent with the argument that many young people can be accidentally exposed to political information in the online environment (e.g., Lupia & Philpot, 2005; Mitchelstein & Boczkowski, 2010). People who are otherwise not too interested in politics may be connected to friends who are interested in public affairs and who would share political information and messages. People may pay attention to such information and messages shared by friends. The relevance of such accidental exposure to political information is confirmed by the structural equation modeling analysis.

To what extent the findings of this article are peculiar to the university student body and to the Hong Kong society? It has already been noted that the contextual background and sample characteristic may indeed matter to the strength and character of certain relationships. The substantial impact of connection with public political actors may be related to the prominence of individual public figures in local politics and the relative ease with which ordinary young people can get in touch with such figures. The impact of network heterogeneity on exposure to shared political information may be affected by the nature of the university student sample. But there are few substantial reasons why the other relationships shown in the study should not be replicable in other places and to the population at large. Future studies can confirm the generalizability of the findings here.

The problem of causality should also be acknowledged here. Some of the relationships in Figure 1 may actually involve causal influence flowing in both directions. For instance, participation may lead to exposure to shared political information via FB, as an individual may become more attentive toward political content after participation. The most plausible scenario is that exposure to shared political information and political participation mutually reinforce each other. The current data set does not allow us to tackle the causality problem. Nevertheless, the analysis should at least have presented useful findings about how the various dimensions of FB usage relate to other each and to participation.

Overall, this study illustrates the complexity of FB usage and how examining the various aspects of FB usage separately can further our understanding of the impact of social media. This study does not claim to have developed the best way to characterize the dimensional structure of social media use. Yet the analysis does suggest that, when examining the impact of social media on political attitudes and behavior, one can consider at least usage frequencies, extents of political communication, and network characteristics. Each of the three aspects can be further differentiated into more fine-grained concepts/variables, such as the identification of network structural heterogeneity, network size, and connection with public political actors as the three pertinent network characteristics in this study. This approach should fit many purposes in a wide range of contexts, though researchers can also continue to refine the dimensionality of social media usage through future research.

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) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research article is supported by a General Research Fund grant offered to the second author of the article by the Research Grant Council of the Hong Kong government (project no.: CUHK449011).

Notes

References

Baumgartner

Morris

(2010). MyFaceTube politics: Social networking web sites and political engagement of young adults. Social Science Computer Review, 28(1): 24–44.

Bennett

W. L.

Iyengar

(2008). A new era of minimum effects? The changing foundations of political communication. Journal of Communication, 58, 707–731.

Chadwick

(2006). Internet politics. Oxford, England: Oxford University Press.

Davis

(1999). The web of politics. New York, NY: Oxford University Press.

Earl

Kimport

(2011). Digitally enabled social change. Cambridge, MA: MIT Press.

Ellison

Steinfield

Lampe

(2007). The benefits of Facebook “friends”: Social capital and college students’ use of online social network sites. Journal of Computer-Mediated Communication, 12, 1143–1168.

Gil de Zúñiga, H., 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.

Gil de Zúñiga, H., Puig-i-Abril

Rojas

(2009). Weblogs, traditional sources online and political participation: An assessment of how the internet is changing the political environment. New Media & Society, 11, 553–574.

Gil de Zúñiga, H., Valenzuela

(2011). The mediating path to a stronger citizenship: Online and offline networks, weak ties and civic engagement. Communication Research, 38, 397–421.

10.

Kushin

Yamamoto

(2010). Did social media really matter? College students’ use of online media and political decision making in the 2008 election. Mass Communication and Society, 13, 608–630.

11.

Lee

F. L. F.

(2006). Collective efficacy, support for democracy, and political participation in Hong Kong. International Journal of Public Opinion Research, 18, 297–317.

12.

Lupia

Philpot

T. S.

(2005). Views from inside the net: How websites affect young adults’ political interest. Journal of Politics, 36, 77–82.

13.

Mitchelstein

Boczkowski

P. J.

(2010). Online news consumption research: An assessment of past work and an agenda for the future. New Media & Society, 12, 1085–1102.

14.

Mutz

(2006). Hearing the other side. New York, NY: Cambridge University Press.

15.

Park

N. S.

Kee

K. F.

Valenzuela

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

16.

Prior

(2007). Post-broadcast democracy. New York, NY: Cambridge University Press.

17.

Rojas

Puig-i-Abril

(2009). Mobilizers mobilized: Information, expression, mobilization and participation in the digital age. Journal of Computer-Mediated Communication, 14, 902–927.

18.

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.

19.

Valenzuela

Park

Kee

K. F.

(2009). Is there social capital in social network site?: Facebook use and college students’ life satisfaction, trust, and participation. Journal of Computer-Mediated Communication, 14, 875–901.

20.

Vitak

Zube

Smock

Carr

Ellison

Lampe

(2010). It’s complicated: Facebook users’ political participation in the 2008 election. Cyperpsychology, Behavior, and Social Networking, 14, 107–114.

21.

Ward

(2012). Communication citizenship online. New York, NY: Hampton Press.

22.

Zhang

Johnson

Seltzer

Bichard

(2010). The revolution will be networked: The influence of social networking sites on political attitudes and behavior. Social Science Computer Review, 28, 75–92.