AINL 2016: Bodrunova, Blekanov, Maksimov
Download as PPT, PDF1 like620 views
The document discusses the structural conditions for inclusive Twitter discussions, exploring the roles of influencers and the dynamics of public discourse during inter-ethnic conflicts across various countries. It examines the disparity in influence metrics, proposing a distinction between 'marketing' and 'deliberative' influencers, and highlights the complexities of comparing discussions in conflict versus calm periods. The analysis reveals that national patterns dominate influencer behavior, with fake accounts skewing nationalistic narratives, while traditional metrics do not always align with social network analysis results.
![Research design: conflict and ‘calm’ periods
Case Conflict period ‘Calm’ period
Russia Biryulyovo warehouse migrants
[killer’s name] people’s gathering
Biryulyovo + migranty +
snowballed keywords
Germany 1) Pegida
2) Koln pegida nopegida
+ more than 10 directly related keywords
1) Pegida + migranten +
snowballed keywords
2) Pegida + migranten +
snowballed keywords
USA Ferguson Ferguson
(+BlackLivesMatter)
France 1) charliehebdo + jesuischarlie /
jenesuispascharlie
2) AttaquesParis (+ over 20 keywords)
CharlieHebdo + migration +
snowballed keywords
AttaquesParis + migration +
snowballed keywords
Belgium AttaquesBruxelles AttaquesBruxelles + migration
+ snowballed keywords
UK Brexit (+directly related keywords) Brexit + migrants + snowballing](https://image.slidesharecdn.com/nsyn6ooxqo4bzxmwarie-signature-d44f3dc49ced19d32cf58a94b539231fc918cb1ad2eec178432f13789e2c54f8-poli-161116133322/85/AINL-2016-Bodrunova-Blekanov-Maksimov-7-320.jpg)
AINL 2016: Bodrunova, Blekanov, Maksimov
- 1. Measuring influencers in Twitter ad-hoc discussions: Active users vs. internal networks in the discourse on Biryuliovo bashings in 2013 Saint Petersburg State University Svetlana S. Bodrunova Ivan S. Blekanov Alexey Maksimov School of Journalism and Mass Communications St. Petersburg State University
- 2. Research premises: societal cleavages in public discussions 1. Public discussion is highly uneven in regard to: - who speaks (institutional players and media are privileged) - who is listened to (again, distorted by media) 2. Media-constructed public sphere: - media as ‘junctions’ of the public sphere - media as gatekeepers of public agendas 3. Network(ed) public discussions on conflicts: - ad-hoc publics | short-term vs. long-term discussions - tracing discussion structure | ‘echo chambers’ vs. crossroads - defining influencers | known disparities + new divisions Saint Petersburg State University
- 3. Research premises: societal cleavages in public discussions Question 1: What are structural conditions for us to speak of an inclusive Twitter discussion? What is an inclusive Twitter discussion? Question 2: Do we see any universal patterns in the structure and influencers in similar discussions across cultures? How do we compare? Saint Petersburg State University
- 4. Twitter as a communicative milieu: optimism vs. pessimism • OPTIMISTS: Twitter as the milieu of platform-limited horizontal communication with a big news alerts potential • PESSIMISTS: Twitter as a de-politicized space for gaming, dating, and chats (Fuchs 2014) • STRUCTURE OF DISCUSSIONS is disputable in terms of what it is and what it tells when we deal with SNA-traced ad-hoc publics (Goncalez-Bailon; Bruns&Burgess 2011; Papacharizzi 2015; Bruns&Highfield 2016) • INFLUENCERS are viewed as key structural elements of power and impact distribution in networked discussions. Aspects like dynamics of influencer status, its linkage to user trust (Liu et al. 2015), discussion topicality (Kelly et al. 2012) etc. are discussed, BUT there is no shared understanding of what an influencer actually is Saint Petersburg State University
- 5. Detecting influencers ‘Absolute’ figures SNA metrics Activity metrics Number of original user posts within the discussion (Ntweets) Number of users involved into the discussion by a given user, that is, commented or retweeted by a given user (In-Degree) Connectivity metrics Number of received interactions, that is, retweets and comments combined, by the given user (Nrecom) 1. Number of users involving the given user into the discussion by commenting or retweeting his/her tweets (Out-Degree). 2. Betweenness centrality (BC). 3. Pagerank centrality (PR). • Three types of divisions in literature: - ‘marketing’ vs. ‘deliberative’ influencers: conceptual difference - measured based on activity vs. based on connectivity - measured by ‘absolute’ figures vs. by SNA metrics
- 6. Research design: cases and their comparability We look at inter-ethnic (inter-national / inter-racial) conflicts. Each case: - has a violent component (killing / terrorist attack / harassment) - has a ‘majority / minority’ component - is an ‘iceberg tip’ of a wider societal cleavage Cases include: -RUSSIA: Biryuliovo bashings -GERMANY: PEGIDA demonstrations, attacks in Koln -USA: Ferguson riots -FRANCE: attack to Charlie Hebdo, Paris attacks -BELGIUM: Brussels attacks -UK: Brexit and the migration issues in it
- 7. Research design: conflict and ‘calm’ periods Case Conflict period ‘Calm’ period Russia Biryulyovo warehouse migrants [killer’s name] people’s gathering Biryulyovo + migranty + snowballed keywords Germany 1) Pegida 2) Koln pegida nopegida + more than 10 directly related keywords 1) Pegida + migranten + snowballed keywords 2) Pegida + migranten + snowballed keywords USA Ferguson Ferguson (+BlackLivesMatter) France 1) charliehebdo + jesuischarlie / jenesuispascharlie 2) AttaquesParis (+ over 20 keywords) CharlieHebdo + migration + snowballed keywords AttaquesParis + migration + snowballed keywords Belgium AttaquesBruxelles AttaquesBruxelles + migration + snowballed keywords UK Brexit (+directly related keywords) Brexit + migrants + snowballing
- 8. Saint Petersburg State University Influencers by the number of tweets
- 9. Saint Petersburg State University Influencers by various metrics: RUSSIA, Biryulyovo Green are media, light green – Twitter media Yellowish are the accounts of ‘angry commenters’ of oppositional stand Orange and light-orange are nationalists and ‘patriots’ Blue are eyewitnesses, grey – spam/fake
- 10. Saint Petersburg State University Influencer strategies: RUSSIA, Biryulyovo In-Degree, not number of received comments, leads to high betweenness and pagerank centralities: a network of users who comment and retweet each other. In the end, they start shaping the discussion
- 11. Conclusions Saint Petersburg State University 1. Direct comparisons of web graphs and influencer lists do not make much sense; but we could compare is to what extent absolute metrics correlate to SNA metrics – to search for more universal activity-to-connectivity patterns. 2. So far, analysis of top posting users in calm and conflict periods shows that national patterns, quite expectedly, dominate… 3. …except for the fact that fake accounts are created in both Russia and Germany to promote nationalistic content. 4. In case of Russia, number of received comments / retweets does not seem to lead to becoming a discussion center by SNA metrics, unlike in previous research. 5. ‘Marketing’ and ‘deliberative’ influencers co-exist – those who post and get a lot of comments are not those who form a high-pagerank network. The latter may have group impact, rather than individual influence. 6. Understandably, a strategy for becoming a high-betweenness user and high- pagerank user seems the same – posting a lot and commenting various users.
Editor's Notes
- #3: Affective publics
- #4: Affective publics
- #11: Твиттер существенно отличается от, например, русского Фейсбука, где доминируют оппозиционные, бизнес-СМИ и массмедиа альтернативной повестки дня. The following media groups can be distinguished in Tweeter: pro-governmental, neutral, anti-governmental/ independent media. In our case study the first segment dominates over the third in terms of activity, with 1.87 times more tweets in this case: the leaders are «Lifenews» (165 tweets) and the "Voice of Russia" (67 tweets).