Does social media use have a positive or negative impact on civic engagement? The cynical “slacktivism hypothesis” holds that if citizens use social media for political conversation, those conversations will be fleeting and vapid. Most attempts to answer this question involve public opinion data from the United States, so we offer an examination of an important case from Mexico, where an independent candidate used social media to communicate with the public and eschewed traditional media outlets. He won the race for state governor, defeating candidates from traditional parties and triggering sustained public engagement well beyond election day. In our investigation, we analyze over 750,000 posts, comments, and replies over three years of conversations on the public Facebook page of “El Bronco.” We analyze how rhythms of political communication between the candidate and users evolved over time and demonstrate that social media can be used to sustain a large quantity of civic exchanges about public life well beyond a particular political event.


Social media have become an important part of modern political campaigning. Campaign managers mine them for data. Citizens and civic groups use a plethora of platforms, such as Twitter, Facebook, and reddit, to talk about politics and engage with civil society groups and political leaders.1 Candidates and political parties are also using social media to manage their public image in communications with journalists and the interested public.2 While many factors affect whether or not political actors adopt social media, the vast majority today are at least actively trying to integrate social media into their campaigns.3

The problem is that despite the technological advancements, most political and activist groups are still in the dark on how best to mobilize people.4 The main difficulty arises because citizens’ decisions about how much to participate in a political cause depend on how they perceive the efforts of the political candidate or organization.5 But the norms on social media platforms are continually shifting. As a result, people are constantly changing how they interpret the social media content and online efforts of others. It can thus be very difficult and time consuming for political groups to keep up with new technology and to predict the outcomes of sharing certain types of media and content.

This complexity has forced many political organizations and candidates to limit how much they use social media and who among their organization or group can use social media.6 Politicians usually prefer to have a point person in charge of their party’s social media strategy and even their own personal accounts. However, even when the point person finally understands how to mobilize citizens, it is not easy to transfer that knowledge to others. As a result, most politicians are very cautious about how they use social media and how much they interact with their online audiences.7 This has hindered and limited our understanding of which political strategies work best to mobilize and engage with citizens.

Our understanding of social media and elections has also been bounded by the fact that most of the research has focused on the United States. In this relatively advanced democracy, social media have become a tool for some forms of civic engagement and political expression. Among the advanced democracies, this seems to have resulted in only modest forms of activism, such as petition signing or sharing political content from affinity groups over networks of family and friends.8 Therefore, research on social media and civic engagement in U.S. politics has often sought to test, directly or indirectly, the “slacktivism hypothesis.” We define the slacktivism hypothesis as the supposition that if Internet or social media use increases, civic engagement declines.

The argument that the use of social media has mostly negative consequences for public life begins with evidence that most of the content shared over social media is rarely about politics, and even when it is about politics it consists of short messages shared by people with short tempers in short conversations.9 Citizens rarely use social media for substantive political conversations, and such conversations are often anemic, uncivil, or polarizing. Overall, online political conversations are relatively rare occurrences in comparison to the other kinds of things people do on the Internet on a daily basis.10

When they do occur, moreover, during major political events such as candidate debates, social media users will use digital platforms to learn about and interact with politics, but they tend to acquire new knowledge that is favorable to their preferred candidate.11 Recent work has found that while many U.S.-based activist organizations believe that they are creating stronger communities and dialogues with their public through social media content, this rarely translates to significant mobilization with regard to public events, consumer activism, or grassroots lobbying.12

Researchers have demonstrated that social media use causes people to turn their social networks into “filter bubbles” that diminish the chance of exposure to new or challenging ideas. In other words, social media allow us to create homophilous networks.13 For example, massive amounts of Twitter data have been used to classify users by party affiliation and homophily in the United States, with results indicating that average Democrats tend to be more homophilous than average Republicans, unless the users classified as Republican follow major Republican leaders.14 Ultimately, public debates over social media may do little more than promote ephemeral engagement without translating to offline political impact.15 When social media actions do have offline impacts, they are usually the same kinds of low-quality, high-volume actions that advocacy and political groups have long used to gain notoriety and news headlines for their organizations.16

In the early stages of the slacktivism debate, there appeared to be only a few very specific cases of movements originating from the Internet that both successfully mobilized people and achieved public policy goals, and often these cases involved a narrow range of technology access issues.17 In more recent years the distinction between online and offline political action has evaporated, such that modern political candidates need to be savvy with multiple technology platforms and many kinds of campaigns spend significant resources on data analytics.18 There are now multiple examples of traditional social movements that have scored impressive victories through their effective use of social media, as well as new social movements that have originated online and become stable civil society actors.19 And complicating all of this is the growing problem of algorithmic control over social media messaging: automated programs can be used to activate citizens or to discourage their engagement.20

The argument against the slacktivism hypothesis is that political engagement over social media is always in addition to, not a replacement for, whatever citizens would normally be doing in their political lives.21 There are important public conversations occurring over social media that grow especially intense during important political events. For example, research has found that social media use helps people build their political identity and community awareness, which even results in financial contributions to relevant civil society groups.22 Indeed, social media, like other Internet-based communications, tend to supplement our intake of information about politics, elections, and public policy, and allow people to be more omnivorous in their information diets. Such “political omnivores” still rely on major broadcast media for information but regularly depend on the Internet for interactivity about politics.23

There is evidence that young adolescents’ use of social media—in conjunction with the intent to participate and the consumption of television news—creates a virtuous circle of civic engagement.24 Most research has consequently focused primarily on small-scale surveys or interview studies. Additionally, it has proven difficult to actually measure levels of civic engagement over social media platforms, especially because there is usually not one central social media site that citizens and politicians use. Being exposed to online activism might influence individual decisions on subsequent civic actions, such as signing a petition or donating to charity, but it is not clear that the use of social media for political conversation results in more sophisticated voters or an increased probability of voter turnout.

While researchers debate the slacktivism hypothesis in the context of the United States and advanced democracies, there are good reasons to expect the relationship between media use and civic engagement in international contexts to be different. The first reason is straightforward for scholars of international studies: there is such a great variety of regime types and political institutions around the world that we should not expect evidence from the United States to hold in many other contexts. The second is a more specific observation from comparative media systems research: the world outside the United States produces, consumes, and regulates political news and information in very different ways.25

Most of the world thinks the Internet is Facebook, and a significant amount of the time many users spend online is actually spent on the Facebook platform.26 The Internet that activists, citizens, and voters use to consume political information in the United States and many advanced democracies is experienced through a browser on a personal computer, and increasingly on a smart phone. In regions of the world where data plans are more expensive and bandwidth more erratic, people instead use SMS (text messaging) or the built-in applications that come with less expensive cellphones. As a result, Facebook has become synonymous with the Internet for most people around the world.

However, the vast majority of the research on slacktivism has been conducted in the United States, and international case studies on the topic demonstrate, sensibly, that the relationship between social media diffusion and civic engagement is complex. Some scholars have demonstrated that political leaders rarely use the most interactive platforms for fear of losing control of the content they produce and the messages they craft.27 Morozov has argued that many digital media platforms are unable to sustain the attention of people who offer a few clicks of support through online petitions but have little energy for the kinds of political engagement that take time or involve personal risk.28 He draws from several international examples of social movements that may have failed because of their dependence on information technologies, but his is not the most systematic analysis.

The literature on international communication is vast, but there are lessons to draw from scholars tackling the study of contemporary political communication. Close study of authoritarian regimes from Azerbaijan to China has revealed that social media can be an important means of conducting political conversations, but that doing so depends on the censorship interests and fabrication activities of ruling elites.29 Bailard has demonstrated through a natural experiment that digital media use during elections in Tanzania and Bosnia resulted in increased levels of civic engagement, negligible impact on voter turnout, and higher rates of voter cynicism.30 Moreover, social media use is an important part of the causal explanation for the shape and character of a growing number of major international public protests, including those in Chile and Hong Kong, where Facebook and Twitter use for news proved to be important predictors of protest participation, even holding constant factors like post-materialist values and political ideology.31 A long-term study of trust in institutions in seven countries across Asia has revealed that social media are particularly important in raising civic engagement in countries where other media options are meager.32 An early analysis of four advanced democracies revealed that fears that the Internet has a broad negative effect on social capital and political participation were unfounded.33

The bloom of pro-democracy protests may also undermine the slacktivism hypothesis. Social movements are always causally complex phenomena. However, activists and protest leaders say social media were essential to the organization of the protests, government leaders try to respond over social media or censor it because they believe it to be integral to the protest, journalists report on the particular dynamics of social media use in their coverage of events, and scholars, in hindsight, find it difficult to develop an analytical narrative about events without discussing the role of social media.34

The Case of “El Bronco”

Finding international cases that help us test the slacktivism hypothesis is a challenge because there are few political leaders, in either authoritarian or democratic regimes, whose careers have been built on the savvy use of social media. The evidence on social media and slacktivism is also encumbered by the fact that in every one of the media systems discussed, social media are only a small part of the communications strategy for political leaders.

Until recently, social media campaigning was mostly about reaching journalists and other policymakers, rather than a broad public.35 Most media systems have a dominant means by which citizens get news and information about politics, and usually it is television, radio, or newspapers. The Internet, and social media in particular, provides a secondary source of media, though interestingly it is—among all the possibilities—the most commonly chosen secondary medium.36 In other words, citizens often get most of their news and information either from the television, radio, or newspaper, and then check sources, poll their friends and family, or do additional research on the Internet.

This too may be changing. In 2015, Jaime Rodríguez Calderón, a Mexican independent candidate also known as “El Bronco,” was elected governor in the Mexican state of Nuevo León on the basis of a campaign that treated traditional media with disdain and actively engaged with the electorate through Facebook and Twitter.37 Unlike most candidates, El Bronco did not pay for TV ads. Even street advertisements, very common in Mexican elections, were done by community supporters rather than through a centralized campaign organization.

In this paper, we analyze the entirety of online interactions between citizens and El Bronco, who, primarily through Facebook, was able to run as an independent and defeat his closest rival by nearly 25 percentage points.38 Nuevo León is a northern state in Mexico with around 5 million inhabitants, the second highest development scores in the country, and the second highest rates of Internet penetration in the country. Close to 60 percent of households have access to the Internet, comparable to states like Mississippi in the United States.39

The case of El Bronco is valuable precisely because it is an extreme case of social media use during a competitive election. It highlights the notable outcome of an independent candidate winning an election through dedicated social media engagement with voters. This major subnational election provides a unique opportunity to answer an important research question: what is the impact of social media use on civic engagement during the campaign season and beyond election day itself?

Data Collection and Analysis

To study the long-term patterns of political engagement over social media, we collected three years of posts, comments, and replies from the Facebook fan page of Jaime Heliodoro Rodríguez Calderón. Using the platform’s application program interface (API) we collected 5,708 posts from the candidate and 71,446 from citizens, 20,045 comments from the candidate and 458,544 from citizens, and 31,527 comment replies from the candidate and 171,577 from citizens. This method captured all activity between November 2012, when Bronco’s social media presence was created, and April 2016, almost one year after he took office.

Our analysis began with the collection of many kinds of data, from many people, generated over many kinds of devices. We created a series of online bots to query the Facebook API and collect data on posts, images, comments, and likes from Bronco’s fan page. We used MongoDB, a non-relational database, to help manage the process of saving large volumes of data generated by our queries and to help reconcile the changes over the years in how Facebook organizes data about posts and user devices.

These data capture political conversations initiated either by the candidate or by a Facebook user. They consist of posts on the politician’s fan page, user comments, and subsequent replies. To help explain the structure of this data, Figure 1 presents the organization of a typical conversation initiated by the candidate. The candidate makes a post that attracts a comment from a citizen. The candidate may then comment on what the citizen has written, and subsequently there can be quite a long thread of discussion through replies.

Figure 2 presents the organization of a political conversation initiated by a citizen. The citizen makes a post on the politician’s page and possibly triggers engagement when the candidate comments on the post. The citizen may then reply and trigger an extended exchange with the candidate. The posts, comments, and replies can be viewed by any user, and other users can subsequently generate their own posts.

To study the communication patterns of the candidate and citizens, we plotted the moving averages of numbers of posts, comments, and replies. We use a fifteen-day statistical moving average to allow us to display a smooth, more interpretable trend line over the three-year period.

El Bronco became an independent politician when he left the Institutional Revolutionary Party (PRI) and, in November 2012, created a Facebook fan page from which to develop his new political profile. At the time, he claimed that his break from the PRI and disavowal of major broadcast media in favor of social media were of themselves important markers of his new public identity. Through much of 2013 the amount of traffic on his social media platform was limited, but by 2014 his hardline stance against organized crime and government corruption found a growing audience. Moreover, he was clearly gaining popularity many months before the election, and by June 2014 he was writing multiple posts, comments, and replies each day, a pace of social media engagement that he was able to sustain for the following two years.

Figures 3 and 4 plot the moving average number of posts, comments, and replies over the entire period. Even though the official campaign period for the office of state governor only lasted three months, it is clear that civic engagement on the fan page peaked well beforehand. Indeed, the analytical narrative for this data should have three chapters: the pre-campaign period of November 2011–February 2015, the official campaign period of March–May 2015, and the time in elected office from June 2015 to the present. In these figures, many of the peaks in political conversation over social media occurred during big moments in Mexican and international politics, and many of the valleys coincide with public holidays.

Figure 4 reveals that the pace of public conversation, both around politics and the candidate, was increasing in terms of posts, comments, and replies well before the formal campaign period began. By May 2014, the public rates of contribution had already hit levels that were sustained for many months thereafter. By that date, the candidate was also posting original content at a pace that he sustained right through to taking office.

To further summarize the trends in social media use and civic engagement, Table 1 presents the weekly average levels of activity before, during, and after the official campaign period. In the lead-up to the campaign, the candidate doubled his pace of posting to social media, making 12 contributions per week, on average, during the campaign. Once in office, El Bronco tended to engage the public with comments and replies to user posts more than through his own original posts. The number of original posts each week returned to the pre-campaign level, his rate of commenting on user posts was higher than in either of the two previous periods, and the frequency of his replies to users diminished very noticeably. Overall, once in office, the candidate’s social media use diminished relative to that of his active period of campaigning.

Table 1 also reveals a transition in the way the public engaged with the candidate on Facebook. In the months before the formal campaign period, users engaged with El Bronco primarily through replies to each other’s comments. During the campaign, users primarily engaged through fresh new comments rather than replies. After the campaign, the pace of social engagement remained high but the distribution among different modes of engagement became more balanced: citizens generated a more equal ratio of posts, comments, and replies.

Users drove large numbers of conversations over social media, and during the peak of the election period they were reacting to the candidate’s messages and queries with 1,100 comments a week. Most interestingly, public engagement with state-level political issues did not evaporate after the official campaign period. If the slacktivism hypothesis were true, we might expect public engagement on this platform to decline over time, and to decline quickly after election day when public conversations move on to other topics. While the pace of social media engagement after the campaign diminished in comparison to that of the campaign period, it remained much higher than that of the period before the campaign. In other words, El Bronco’s page became a sustainable mode of public conversation as users continued to post, comment, and reply with much more frequency than before the campaign.

Clearly there was some kind of positive feedback between the candidate’s energetic social media use and users’ willingness to engage in politics on the Facebook platform. The final step in this analysis is to estimate the degree to which the number of posts, comments, and replies from users drove El Bronco’s use of social media. Similarly, what types of interaction with the candidate drove users to stay engaged beyond election day?

To answer this question, we conducted a regression analysis to model the relationship between citizens’ direct interaction with politicians over social media and their continued engagement with political issues on the same platform. To do so, we conducted a repeated measure logistic regression analysis to see whether direct interaction with El Bronco predicted subsequent engagement on his fan page. We included the participation measures of posting on the page, commenting on a post, and replying to a comment on the page as independent variables in the model.

Our analysis indicates that the candidate’s engagement with and replies to individual citizens is a significant, positive predictor of how likely those citizens will go on to post, comment, and reply on the page (β=6.66, p < 0.00 for comments; β=0.91, p < 0.00 for replies). The inverse relationship is a little weaker, and a particular citizen’s engagement with El Bronco does not raise the likelihood that he will post, comment, or reply in an engaging way with that citizen (β=0.47, p < 0.00 for comments; β=0.08 for posts; and β=0.14, p <0.002 for replies). Receiving direct online communication from citizens rarely motivates a candidate to reply.

Overall, these results suggest that direct and personal interactions with politicians can lead citizens to participate more in public conversations about politics and policy. It appears that the more that the candidate directly interacts with citizens, the more citizens participate in politics online, and this is true for Facebook posts, comments, and replies. The same is true for how the candidate engages with citizens: the more the candidate receives replies and comments from citizens, or posts that involve them, the more the candidate will comment, reply to, and share with the community.

Conclusion

This case is a valuable source of evidence of the role social media can have in extended forms of civic engagement. In this Mexican state election, a candidate developed a political communication strategy focused primarily on social media. He won, and his public Facebook page became the key information infrastructure supporting public policy conversations among citizens well beyond election day. This case is significant because it is one of the first clear cases of a political candidate successfully using social media to win elected office and sustain public conversation. There are certainly other cases of failed campaigns and of social media conversations that were neither civil nor sustained. But our findings are consistent with the growing research consensus that political campaign managers themselves see social media as a way for candidates to engage with voters.40

Does social media use have negative consequences for civic engagement? It turns out that in an election where a political candidate and citizens are comfortable using social media, the impact is positive for both kinds of political actors. Candidates find their supporters and learn about their constituents; citizens engage with candidates and come to expect engagement even after election day when candidates take office. This challenges the slacktivism hypothesis, which holds that civic engagement over social media does little more than promote ephemeral engagement and does not translate to the offline world.41

Too often various forms of digital activism are dismissed as slacktivism, closing off inquiry into the ways in which people blend media use and political conversation into their daily lives. Evaluating the slacktivism hypothesis requires that we appreciate the range of social media affordances, media platforms, and usage patterns that define contemporary political culture.42 The supposition that if social media increases civic engagement decreases can certainly be evaluated both in terms of the quality and quantity of public interactions—in this article we evaluate the hypothesis with evidence about the volume and frequency of interactions. Anecdotally, we observed that the page began as a forum for political positioning and argument before the campaign and transitioned to a platform for people to submit public grievances and requests after the campaign. Further research could examine the qualities of particular posting formats or the political sophistication and purpose of contributions.

Based on the evidence collected here, if both political leaders and citizens use the same social media platform, in the context a democratic election, we may expect some positive engagement outcomes. The data reveal that candidate engagement with citizens on the Facebook fan page had a positive effect, resulting in continued platform use. Moreover, continued citizen participation had a positive impact on the politician. His rhythm of engagement was not evenly sustained throughout the campaign period, but it remained noticeably high even after he took office and presumably had more direct access to journalists and broadcast media.

As always, case selection and methodology provide both strengths and limits to generalization. There are many ways to evaluate the slacktivism hypothesis, and here we take a step by investigating changes in the frequency of contributions. In the future, it will be important to study message qualities and the range of other informational, social, political, psychological, or technological factors that may moderate the relationship between social media use and civic engagement. Candidates for elected office in other countries, using other social media platforms, will face different capacities and constraints. Moreover, a wide range of counterfactuals can be imagined. If El Bronco had lost the election and left politics, would the citizens who had been drawn into political conversations on Facebook have continued to actively engage in politics on that platform? Would they have remained engaged but moved off to another page? While it is possible to analyze the rhythm of political conversation, the data do not reveal which users actually voted or who they voted for.

But these counterfactuals and caveats do not undermine the argument that when candidates for elected office and the public use social media for political conversation, they can create new patterns of civic engagement that can last for months beyond an election.

Philip N. Howard is a professor of technology, international affairs, and public life at Oxford University.

Saiph Savage is an assistant professor at West Virginia University, where she directs the Human Computer Interaction Laboratory, and is also adjunct faculty at the National Autonomous University of Mexico.

Claudia Flores Saviaga is a PhD student in social computing and crowdsourcing at West Virginia University, research collaborator at the National Autonomous University of Mexico, and technology advisor for the Mexican government.

Carlos Toxtli is a PhD student at West Virginia University, where he conducts research in human computer interaction and social computing.

Andrés Monroy-Hernández is a researcher at Microsoft Research.


Acknowledgements

The authors gratefully acknowledge the support of the National Science Foundation, “EAGER CNS: Computational Propaganda and the Production / Detection of Bots,” BIGDATA-1450193, 2014-16, Philip N. Howard, principle investigator, and the European Research Council, “Computational Propaganda: Investigating the Impact of Algorithms and Bots on Political Discourse in Europe,” Proposal 648311, 2015-2020, Philip N. Howard, principal investigator. Project activities were approved by the University of Washington Human Subjects Committee, approval #48103-EG, and the University of Oxford’s Research Ethics Committee. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Science Foundation or European Research Council.


1 David Karpf, The MoveOn Effect: The Unexpected Transformation of American Political Advocacy (Oxford, UK: Oxford University Press, 2012); Jason Gainous and Kevin M. Wagner, Tweeting to Power: The Social Media Revolution in American Politics (New York, NY: Oxford University Press, 2013); Jonathan A. Obar, Paul Zube, and Clifford Lampe, “Advocacy 2.0: An Analysis of How Advocacy Groups in the United States Perceive and Use Social Media as Tools for Facilitating Civic Engagement and Collective Action,” Journal of Information Policy 2 (2012), 1-25.

2 Philip N. Howard, New Media Campaigns and the Managed Citizen (New York, NY: Cambridge University Press, 2006); Peter Dahlgren, “The Internet, Public Spheres, and Political Communication: Dispersion and Deliberation,” Political Communication 22 (2005), 147-62; Kristen Lovejoy and Gregory D. Saxton, “Information, Community, and Action: How Nonprofit Organizations Use Social Media,” Journal of Computer-Mediated Communication 17 (2012), 337-353.

3 Seungahn Nah and Gregory D. Saxton, “Modeling the Adoption and Use of Social Media by Nonprofit Organizations,” New Media & Society 15 (2013), 294-313.

4 Lovejoy and Saxton, “Information, Community, and Action: How Nonprofit Organizations Use Social Media.”

5 Saiph Savage, Andrés Monroy-Hernández, and Tobias Höllerer, “Botivist: Calling Volunteers to Action Using Online Bots” (Proceedings of the 19th ACM Conference on Computer-Supported Cooperative Work & Social Computing, ACM, New York, NY: 2016), 813-822.

6 Jonathan A. Obar, “Canadian Advocacy 2.0: An Analysis of Social Media Adoption and Perceived Affordances by Advocacy Groups Looking to Advance Activism in Canada,” Canadian Journal of Communication 39 (2014), 211-233.

7 Savage, Saiph, and Andrés Monroy-Hernández, “Participatory Militias: An Analysis of an Armed Movement’s Online Audience” (Proceedings of the 18th ACM Conference on Computer Supported Cooperative Work & Social Computing, ACM, New York, NY: 2015), 724-733.

8 Rachel Gibson, Philip Howard, and Stephen Ward, “Social Capital, Internet Connectedness and Political Participation: A Four-Country Study” (paper prepared for the International Political Association in Quebec, Canada, 2000).

9 James E. Katz, Michael Barris, and Anshul Jain, The Social Media President: Barack Obama and the Politics of Digital Engagement (Oxford, UK: Springer, 2013).

10 Adrienne L. Massanari and Philip N. Howard, “Information Technologies and Omnivorous News Diets over Three U.S. Presidential Elections,” Journal of Information Technology & Politics 8 (2011), 177-198.

11 Shelley Boulianne, “Social Media Use and Participation: A Meta-Analysis of Current Research,” Information, Communication & Society 18 (2015), 524-538.

12 Chao Guo and Gregory D. Saxton, “Tweeting Social Change: How Social Media Are Changing Nonprofit Advocacy,” Nonprofit and Voluntary Sector Quarterly 43 (2014), 57-79; Lovejoy and Saxton, “Information, Community, and Action: How Nonprofit Organizations Use Social Media.”

13 Eli Pariser, The Filter Bubble: How the New Personalized Web Is Changing What We Read and How We Think (New York, NY: Penguin Books, 2012).

14 Elanor Colleoni, Alessandro Rozza, and Adam Arvidsson, “Echo Chamber or Public Sphere? Predicting Political Orientation and Measuring Political Homophily in Twitter Using Big Data,” Journal of Communication 64 (2014), 317-332.

15 Henrik Serup Christensen, “Political Activities on the Internet: Slacktivism or Political Participation by Other Means?” First Monday 16 (2011); Dana Rotman et al., “From Slacktivism to Activism: Participatory Culture in the Age of Social Media” (CHI’11 Extended Abstracts on Human Factors in Computing Systems, ACM, 2011), 819-822.

16 David Karpf, “Online Political Mobilization from the Advocacy Group’s Perspective: Looking Beyond Clicktivism,” Policy & Internet 2 (2010), 7-41.

17 Yochai Benkler et al., “Social Mobilization and the Networked Public Sphere: Mapping the SOPA-PIPA Debate,” Political Communication 32 (2015), 594-624; Robert Faris et al., “Score Another One for the Internet? The Role of the Networked Public Sphere in the U.S. Net Neutrality Policy Debate,” SSRN Electronic Journal (2015); Des Freedman et al., Strategies for Media Reform: International Perspectives (Oxford, UK: Oxford University Press, 2016); Elizabeth Dubois and William H. Dutton, “The Fifth Estate in Internet Governance: Collective Accountability of a Canadian Policy Initiative,” Revue Française D’études Américaines 134 (2013), 81-97.

18 Daniel Kreiss, Prototype Politics: Technology-Intensive Campaigning and the Data of Democracy (New York, NY: Oxford University Press, 2016); Rasmus Kleis Nielsen, Ground Wars: Personalized Communication in Political Campaigns (Princeton, NJ: Princeton University Press, 2012).

19 Jessica Beyer, Expect Us: Online Communities and Political Mobilization (Oxford, UK: Oxford University Press, 2014).

20 Saiph Savage and Andrés Monroy-Hernández, “Participatory Militias: An Analysis of an Armed Movement’s Online Audience” (Proceedings of the 18th ACM Conference on Computer Supported Cooperative Work & Social Computing, ACM, 2015), 724-733; Samuel C. Woolley and Philip N. Howard, “Political Communication, Computational Propaganda, and Autonomous Agents — Introduction,” International Journal of Communication 10 (2016), 9.

21 Christensen, “Political Activities on the Internet.”

22 Yu-Hao Lee and Gary Hsieh, “Does Slacktivism Hurt Activism?: The Effects of Moral Balancing and Consistency in Online Activism,” (Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, ACM, New York, NY: 2013), 811-820.

23 Adrienne L. Massanari and Philip N. Howard, “Information Technologies and Omnivorous News Diets over Three U.S. Presidential Elections,” Journal of Information Technology & Politics 8 (2011), 177-198.

24 Sanne Kruikemeier and Adam Shehata, “News Media Use and Political Engagement Among Adolescents: An Analysis of Virtuous Circles Using Panel Data,” Political Communication 34 (2016), 1-22.

25 Hartmut Wessler et al., “Global Multimodal News Frames on Climate Change: A Comparison of Five Democracies around the World,” The International Journal of Press/Politics 21 (2016), 423-445.

26 James B. Stewart, “Facebook Has 50 Minutes of Your Time Each Day. It Wants More,” New York Times, 5 May 2016.

27 Jennifer Stromer-Galley, Presidential Campaigning in the Internet Age (New York, NY: Oxford University Press, 2014); Philip N. Howard, “Digitizing the Social Contract: Producing American Political Culture in the Age of New Media,” The Communication Review 6 (2003), 213-245.

28 Evgeny Morozov, To Save Everything, Click Here: The Folly of Technological Solutionism (New York, NY: PublicAffairs, 2014).

29 Gary King, Jennifer Pan, and Margaret E. Roberts, “How the Chinese Government Fabricates Social Media Posts for Strategic Distraction, Not Engaged Argument” (2016); Gary King, Jennifer Pan, and Margaret E. Roberts, “How Censorship in China Allows Government Criticism but Silences Collective Expression,” American Political Science Review 107 (2013), 326-343. Katy E. Pearce and Farid Guliyev, “The Affordances of Social Media for a Repressed Opposition against an Entrenched Authoritarian Regime in Azerbaijan,” The Routledge Companion to Social Media and Politics (2015), 235-247.

30 Catie Snow Bailard, Democracy’s Double-Edged Sword: How Internet Use Changes Citizens’ Views of Their Government (Baltimore, MD: Johns Hopkins University Press, 2014); Catie Snow Bailard, “A Field Experiment on the Internet’s Effect in an African Election: Savvier Citizens, Disaffected Voters, or Both?” Journal of Communication 62 (2012), 330-344.

31 Sebastián Valenzuela, Arturo Arriagada, and Andrés Scherman, “The Social Media Basis of Youth Protest Behavior: The Case of Chile,” Journal of Communication 62 (2012), 299-314; Paul Lee, Clement So, and Louis Leung, “Social Media and Umbrella Movement: Insurgent Public Sphere in Formation,” Chinese Journal of Communication 8 (2015), 356-375.

32 Shin Lee, “Digital Democracy in Asia: The Impact of the Asian Internet on Political Participation,” Journal of Information Technology and Politics (2016), 1-21.

33 Gibson, Howard, and Ward, “Social Capital, Internet Connectedness and Political Participation.”

34 Andrés Monroy-Hernández et al., “Narcotweets: Social Media in Wartime” (Proceedings of the Sixth International AAAI Conference on Weblogs and Social Media, 2012), 515-518.

35 Daniel Kreiss, “Seizing the Moment: The Presidential Campaigns’ Use of Twitter during the 2012 Electoral Cycle,” New Media & Society 18 (2016), 1473-1490.

36 Massanari and Howard, “Information Technologies and Omnivorous News Diets over Three U.S. Presidential Elections.”

37 Carin Zissis, “El Bronco Bucks Mexico’s System,” AS/COA, 11 June 2015.

38 Paulina Villegas and Randal C. Archibold, “El Bronco: Blunt, Frequently Vulgar, and Aiming to Run Nuevo León,” New York Times, 24 May 2015.

39 “Índice de Desarrollo Humano Para Las Entidades Federativas, México 2015” (report, UNDP Mexico, 2015); “Estadisticas a Proposito Del Dia Mundial De Internet” (Aguascalientes: INEGI, 2016); Lee Rainie and D’Vera Cohn, “Census: Computer Ownership, Internet Connection Varies Widely across U.S.” (report, Pew Research Center, 2014).

40 Kreiss, Prototype Politics; Nielsen, Ground Wars.

41 Christensen, “Political Activities on the Internet;” Rotman et al., “From Slacktivism to Activism.”

42 Peter Nagy and Gina Neff, “Imagined Affordance: Reconstructing a Keyword for Communication Theory,” Social Media + Society 1 (2015), 1-9.