Yuri Rykov, National Research University Higher School of Economics
The study has investigated the network structure and user behavior of online communities based on the same SNS platform from the perspective of the '90-9-1 rule' for participation inequality (Nielsen, 2006) and heavy-tailed distributions of network properties (Barabasi, 2002). According to this theory online community members are significantly differentiated by their posting behavior, and that may lead to extremely unequal distribution of valuable resources as attention and social capital. According to a theory the structure and users' interaction patterns within online communities vary depending on the platform technical features, temporal structure, external contexts (e.g. language), participants characteristics and group purposes (Baym, 1998; Preece, 2001). Studies devoted to discussion communities on different platforms show there is an association between structural patterns of discussion networks and subjects or topics of communication (Adamic, Zhang, Bakshy, & Ackerman 2008; Gonzalez-Bailon, Kaltenbrunner, & Banchs, 2010; Smith, Rainie, Shneiderman, & Himelboim, 2014). Therefore, the study of communities' functioning and structure in comparative perspective is of interest. We suppose SNS-based online communities from different spheres of social life probably differ from each other by participant's networking and communication behavior and therefore by inequality scores (Gini indexes, graph centralization) even basing on the same platform. In this particular research we focus on online communities form different spheres of social life and with different functions respectively: fan communities - cultural consumption and entertainment, professional communities - knowledge sharing and job seeking, social movement communities - civil society goals achievement.
What are the differences between 'friendship' networks of these three types of online communities? How participation and network inequalities change across different types of online communities? How purposes of online communities determine forms of inequality and hierarchy within?
An empirical object are online groups in the most popular Russian SNS - VK.com. Sample includes 55 groups (vary in size from 5,000 to 34,000 users) equally corresponding to three exploring types of communities: fan communities (e.g. musicians fans), professional communities (e.g. IT specialists, engineers) and social movement communities (e.g. urban, LGBT movements). The data was available through API and was collected automatically by special software. Each group dataset includes: 1) complete data from group's 'wall' and discussion boards including users' activity rates; 2) the metadata of all participants (gender, age, geographical location, etc); 3) the data on 'friend' relationships existing among community participants. Nodes in the network are users participating in online groups. Ties are 'friend' relationships between them. To analyze data we use social network analysis methods and statistics (linear models, ANOVA).
Fan networks have lower density and are less filled with ties, comparing to other groups. Fan networks have significantly more connected components and graph clusters, a higher value of Gini index for betweenness centrality distribution, indicating a greater fragmentation of fan communities compared with other. It means participants are less likely to use fan groups to networking with like-minded individuals and form a collective social capital. Professional communities have the largest share of posting users and lowest Gini index for posted messages distribution that indicates more participatory behavior of users in content creation and knowledge sharing (is in consistent
with Adamic, Zhang, Bakshy, & Ackerman 2008). The highest scores for betweenness centralization indicates that despite the wide participation professional networks stay highly fragmented and clustered. The possible explanation is that the majority of active users from professional groups are "answer persons" with disconnected neighbors in their ego-networks (Welser, Gleave, Fisher, & Smith, 2007). Social movement networks are the most dense and the most internally connected, comparing to others, may be because the collective action require the cooperation between participants. This finding is consistent with some previous research (Gruzd & Tsyganova, 2015). Despite solidarity and cohesion these networks are the most centralized and unequal by degree centrality. Thus, online communities are used by movement activists to accumulate group-level social capital, but larger inequality emerges on the individual-level social capital.
Adamic, L. A., Zhang, J., Bakshy, E., & Ackerman, M. S. (2008). Knowledge sharing and yahoo answers: everyone knows something. In Proceedings of the
17th international conference on World Wide Web (pp. 665–674). ACM. Retrieved from http://dl.acm.org/citation.cfm?id=1367587
Barabasi A. (2002) Linked: The New Science of Networks. Cambridge, Perseus Books Group.
Baym, N. K. (1998). The Emergence of On-Line Community. In S. G. Jones (Ed.), Cybersociety 2.0: Revisiting Computer-Mediated Communication and Community (pp. 35–68). Thousand Oaks, CA: SAGE.
Gonzalez-Bailon, S., Kaltenbrunner, A., & Banchs, R. E. (2010). The structure of political discussion networks: a model for the analysis of online deliberation. Journal of Information Technology, 25(2), 230–243. http://doi.org/10.1057/jit.2010.2
Gruzd, A., & Tsyganova, K. (2015). Information Wars and Online Activism During the 2013/2014 Crisis in Ukraine: Examining the Social Structures of Pro- and Anti-Maidan Groups. Policy & Internet, 7(2), 121–158. http://doi.org/10.1002/poi3.91
Nielsen J. (2006) Participation Inequality: The 90-9-1 Rule for Social Features. Retrieved February 15, 2016, from https://www.nngroup.com/articles/participation-inequality/
Preece, J. (2001). Sociability and usability in online communities: Determining and measuring success. Behaviour & Information Technology, 20(5), 347–356. http://doi.org/10.1080/01449290110084683
Smith, M. A., Rainie, L., Shneiderman, B., & Himelboim, I. (2014, February 20). Mapping twitter topic networks: From polarized crowds to community clusters. Pew Research Internet Project. Retrieved from http://www.pewinternet.org/2014/02/20/part-2-conversational-archetypes-six-conversation-and-group-network-structures-in-twitter/
Welser, H. T., Gleave, E., Fisher, D., & Smith, M. (2007). Visualizing the signatures of social roles in online discussion groups. Journal of Social Structure, 8(2), 564–586.