The Wealth of (Occupation) Networks? Communication Patterns and Information Distribution in a Twitter Protest Network

Can social media facilitate better protest action organization and coordination? This question has been at the forefront of discussion between media pundits and academic scholars. But what networking mechanisms and communication patterns help activists achieve better organization and coordination? This study uses social network analysis to explore students' Twitter use during the 2010 university occupations. It maps the interconnections between Twitter occupation accounts at the peak of the online protest network's activity and examines how this structure benefited the entire network. It identifies (a) how "closely" accounts were connected, (b) which were the best connected, and (c) which acted as brokers. Findings show that the protest groups used the Twitter network structure to establish horizontal, direct lines of communication to efficiently redistribute information. Although some accounts played a larger role in distributing information than others, occupation accounts could continue to receive information from the decentralized network even if well-connected accounts were removed. Keywords: Political activismsocial mediasocial networking sitesTwitterUK university occupations Acknowledgments I would like to thank the Editor and the three anonymous reviewers for the thought provoking and constructive comments from which this article greatly benefited, and my colleagues at the University of Mannheim who read and commented on a previous draft. I am also greatly appreciative of the support of the Alexander von Humboldt Foundation. Notes 1. Data for replication of Tables 1–3 will be deposited within the journal's dedicated space in the Dataverse (http://dvn.iq.harvard.edu/dvn/dv/jitp) two months after publication. 2. Web 2.0 is a Web platform that can accommodate interactive information sharing, interoperability, user-centered design, and collaboration on the World Wide Web, and is characterized by applications such as video-sharing sites, wikis, blogs, and SNS (e.g., Facebook, YouTube, Twitter). Web 2.0 applications emphasize the importance of user participation, openness, and network effects (CitationBekkers et al., 2011). 3. But see Lotan et al. (2011), and also relevant studies by CitationGruzd et al. (2011) and CitationTakhteyev et al. (2012). 4. It must be noted that this approach, and especially searching for specific occupation followers among thousands of others, became extremely time consuming as the number of followers increased. This method was only selected due to the relatively small number of occupations (and, consequently, Twitter accounts) and the urgency of capturing the protest activities from their very beginning. More time-effective approaches for compiling lists of activists have been used (CitationGruzd et al., 2011). Moreover, collecting data in this fashion introduces a validation problem because the data cannot be replicated. There is a limit to how far back you can go on your Twitter timeline and, although one could still observe some of the follower/followed relationships established between occupation accounts (that have remained active since December 2010), it would not be possible to get a reliable picture of how connecting patterns among them looked like on the eve of December 9. Furthermore, although Twitter's open API allows for various possibilities for scholars with developer knowledge to search and collect Twitter data (see CitationBoyd, Golder, & Lotan, 2010), it would not be feasible to determine which and how many followers a potential account had on a particular day and time back in 2010. 5. Although all 38 occupations created a Twitter account, by the evening of the vote, two of them, Royal Holloway and East Anglia universities, were suspended and were, therefore, not included in the analysis.