This week I was in Russia to deliver two public talks at the Higher School of Economics in St. Petersburg and Moscow. I had the chance to get to know the work of colleagues working with Russian social networks at HSE and the New Economic School in Moscow. I met up with Olessia Koltsova, Iliya Kiriya, Sergey Chernov, Maria Petrova, Ruben Enikolopov, and dozens of great scholars. Overall there is a fascinating push for internationalization in Russian universities, with research being funded from many different sources and scholars being very motivated to produce high-impact scholarship.
In St. Petersburg I talked about the overlapping of users and political communities on Twitter. This is a work coauthored with Cornelius Puschmann and Rodrigo Travitzki and presented at the Hypertext conference earlier this year. We analyzed nearly five hundred hashtags of different topics and languages (mostly English, Spanish and Portuguese tweets) that included a total of over 1 million contributing users. This research offers a glimpse on serial activism – roughly 70% of all users in our dataset tweeted under at least two different hashtags – and political hashtags are particularly prone to overlap with other political hashtags. The next plots show the hashtags clustered in topics and languages.
The Russian audience was particularly interested in the importance of "sockpuppets" reported in this study. Sockpuppeteering refers to a wide range of online identity use for purposes of deception. In this case, we found that politicians, media pundits, and celebrities appeared prominently in the tweets of users that strategically associate them with causes they are most likely unaware of. We found that addressing and mentioning of well-known users and institutions such as @barackobama, @occupywallst, or @cnn is a common strategy and that these accounts did not contribute to the information stream themselves.
Although it is hard to say why exactly this happens, we believe that activists want to recruit or catch the attention of prominent users. In fact, we found many highly-followed accounts that did not post messages with the hashtags but that appeared in the dataset. While the dataset with all users – active and passive – includes over 25 billion non-unique followers, the dataset without sockpuppets includes only 4 billion non-unique followers.
Another interesting (albeit unsurprising) finding was the importance of language in the structure of the network of topics. Hashtags and information streams in English are more closely associated than those in Spanish and Portuguese. Subgroups dedicated to health, politics, entertainment, video games, and sports are also noticeable, with the Occupy movement, Kony2012, and the Spanish Indignados (15M) movement being particularly well connected to each other. These political networks are transnationally followed events that succeed at transcending language barriers.
In Moscow we discussed how newsmaking and newsworthiness are changing due to the growing adoption of social media tools by journalists and audiences. I presented a recent research on the topic published in the journal Sage Open that I coauthored with Gabriela Zago. We described how national differences are reflected by the particular journalistic cultures and audience interests and debated the conflict between what national newspapers editors and audiences deem to be of interest in a number of countries. Perhaps surprisingly, we found that news content in social networks favors items about arts, technology, and particularly opinion pieces, unlike newspapers that are often focused on sports and economics. The chart below shows which platform Twitter users relied on to post links to news articles in the US, UK, Spain, Brazil, and Germany.
I also had the chance to get to know better the Russian social media scene. This is an interesting market because local companies have an edge over multinational corporations in areas like internet search and social networking. Like in China, Google is yet to gain a foothold in the Russian web search business, where over 60% of market share is in the hands of Russian Internet company Yandex. Again confirming the relationship between communities and languages, the search engine reportedly allows for using Russian inflections in search queries.
VKontakte (http://vk.com/) is the second largest social network service in Europe and it appeals particularly to Russian speakers. This poses interesting research opportunities in terms of comparing and contrasting social networks in a cross-national context. I also asked the colleagues I met about networks dedicated to scholarly work (i.e. communities of bloggers like LiveJournal and WordPress, or a social networks dedicated to academic content like Academia and ResearchGate). Unfortunately, we have not been able to identify any significant network dedicated to scholarly work like HASTAC. If you know of a network in Russia that fits the description, make sure to drop me a line in the comment box below or just hit me on Twitter.
- Bastos, M., Puschmann, C., Travitzki, R. (2013). Tweeting Across Hashtags: Overlapping Users and the Importance of Language, Topics, and Politics. Proceedings of 24th ACM Conference on Hypertext and Social Media, 1–3 May 2013, Paris.
- Bastos, M., Zago, G. (2013). Tweeting News Articles: Readership and News Sections in Europe and the Americas. SAGE Open 3 (3).