Social Network Analysis
Social media analytics is the process of gathering and analysing data from social networks. (Scott, 2000). Social Network Theory is the study of how people or groups interact with others inside their network. The three types of social networks are ego-centric networks, socio-centric networks, and open-system networks (Borgatti, & Lopez-Kidwell, 2011).
The objective of social network analysis (SNA) is to understand the interactions between each of the members of the network. These connections, called relationships or ties, are at the heart of what this analysis seeks to study and understand. The reasons why the individuals interact and how they interact their level of closeness (Borgatti et al., 2009). SNA provides both qualitative and quantitative data of online learning communities.
Social Network Analysis: GO-GN Insights
Aras Bozkurt used SNA to track digital footprints of online participants and map and visualize online learning community.
“For data collection and analysis, social network analysis, interview, observation and document analysis was used. Research findings were interpreted with the perspectives of connectivism, rhizomatic learning and social network theory.
“According to the demographic findings of the research, learners in connectivist massive open online networks are distributed globally in time and place, many participate from English spoken countries, and 89% of the learners come from low-context cultures while 11% comes from high context cultures. Participants are individuals that are somehow connected to the education field; or students or instructors in higher education. When examined in terms of interaction patterns, unified-tight crowd community pattern was observed in connectivist massive open online course networks. The nodes in this kind of networks have strong connections to one another and significant connections that bridge sub-groups. Learners of this type of networks tend to communicate with each other frequently and share a common interest. These networks are composed of a few dense and/or densely interconnected groups where conversations usually swirl around and increase its density towards the center, involving different people at different times.
“Research findings additionally demonstrated that connectivist learning environments require relatively few hops to communicate and interact with the learning community, and confirmed the theses proposed in the Small World Phenomenon and the Global Village. SNA provides both qualitative and quantitative data of online learning communities. However, it fails to provide phenomenological qualitative data.”
Some researchers collect this phenomenological data separately. For example, in addition to analysing network structures, Katy Jordan held co-interpretive interviews with 18 participants, to understand the significance and construction of their academic social networks.
“My PhD study addressed the question of how academics use dedicated social networks through mixed methods social network analysis. First, an online survey was conducted to gain contextual data and recruit participants (n = 528). Second, ego-networks were drawn up for a sub-sample of 55 academics (reflecting a range of job positions and disciplines). Ego-networks were sampled from an academic SNS and Twitter for each participant. Third, co-interpretive interviews were held with 18 participants, to understand the significance of the structures and how the networks were constructed.
“My methods changed direction (subtly) twice during the course of my PhD. The focus was always on the structure of academic online social networks, but the level at which I looked at the networks changed. Originally I had planned to look at networks at a larger scale – such as the entire UK HE sector on Academia.edu. I changed tack to focus on academics’ individual (personal, ego-) networks instead, for two reasons. First, ethically, it is a lot more sound to capture an ego-network – at this level, you can get the participants’ consent. Second, in order to be able to understand the structures involved. For example, I could see interesting structural features in the OU networks, but network metrics can only tell you so much. By sampling personal networks, the structures could be meaningfully discussed with the participants themselves, in order to understand the significance and characteristics of different network features from their perspective. Combining digital (scraped) data with co-interpretive interviews offers much greater insight into the digital, open practices behind the network structures.”
Useful references for Social Network Analysis: Borgatti & Lopez-Kidwell (2011); Borgatti et al. (2009); Dominguez & Hollstein (2014); Edwards, G. (2010); Hansen, Shneiderman & Smith, (2010); Jordan (2018); Kozinets (2015); Newman (2018); Scott (2000); Wenger, Trayner & de Laat (2011)