From a technical point of view, social network analysis (SNA) is the process of investigating social structures through network and graph theories [1]. Essentially, it involves creating a colourful, if a little confusing at first, graph that is structured by nodes and edges. Nodes are individuals within a network and the edges are the relationships between the nodes that connect them. “Great and what does that mean?” I hear you say. Well with this information it is then possible to determine who within a network is important, in terms of information transfer, and who may be a little more peripheral, literally, within the network. It is used in several disciplines from the hard and soft sciences and most things in between. So it is evident that its practical applications could be far reaching from anything to mapping social media networks, friendship and acquaintance networks (great if you want to see how important or unimportant you are to your friendship group) and disease transmission.
To help visualise this there is a simple example of a kite graph below which is commonly used to explain social network analysis [2]:
So what does this mean… this is essentially a graph of who invited who to a party. We can see that Diane is the most central person for invites and has a degree of six as she is directly connected to six other people. If we were looking at this in terms of popularity it would be reasonable to say that Diane is the most popular because she has the most connections whilst Jane has only one direct connection and therefore, sadly for Jane, would be technically the least popular. However, Diane is not the be-all and end-all because if you take a look at Heather we can see she has a degree factor of three (connected directly to three other people). Whilst this is lower than Diane her position is still important. Therefore, if information needed to be conveyed to Ike, and by extension to Jane, this would not occur and Ike and Jane wouldn’t get an invite to the party… things aren’t looking good for Jane. But fortunately, Heather is there with her high betweeness centrality which in lay-person’s terms means she is important for connecting two groups of people together.
Having said this, demonstrating how popular someone is within a group is not that profound. However, when you think of the more complex graphs that can be produced with networks consisting of thousands of people its results can be a lot more useful for indicating influence and presence of smaller sub networks. Although social network analysis with a user-friendly programme such as NodeXL is in its relative infancy it is intriguing to see the range of ways in which it can be used to explore different kinds of networks.
For example, with NodeXL this can even be applied to social media platforms such as Twitter in order to explore how information is dispersed amongst follower-networks etc. This is especially relevant as social media use for businesses and individuals becomes all the more ubiquitous. The future of social network analysis with programmes such as NodeXL is an exciting and developing field for sure.
[1] Otte, E. and Rousseau, R., 2002. Social network analysis:
a powerful strategy, also for the information sciences. Journal of information Science,28(6),
pp.441-453.
[2] Hansen, D., Shneiderman, B. and Smith,
M.A., 2010. Analyzing social
media networks with NodeXL: Insights from a connected world. Morgan
Kaufmann.
