The social network analysis perspective is a mode of analysis that unites one of sociology’s fundamental objects of study, the intersection between the individual and society, with modern analysis techniques. Specifically, social network analysis is rooted in Moreno’s sociometric analysis techniques, the “Harvard structuralist” school, and the Manchester anthropologist tradition (Keim, 2011). Whereas traditional quantitative projects are suited for comparing individuals and populations, social network analysis techniques allow the researcher to study specific relationships between actors, both directly and indirectly, in order to answer research questions that consider social relationships as a primary concern, such as research into socialization phenomena.
Because social network analysts consider networks as opposed to comparing individuals or groups, the structure of the data they analyze consequently changes. Traditional research often focuses on measurements about these individuals or groups, taken in a vacuum that ignores their interactions. Conversely, social network analysts primarily study data directly relevant to the interaction of two actors. Specifically, this is known as relational data, and is typically understood as the data that specifies any two nodes’ relation to each other. This could be as simple as stating that a relationship between two actors exists, but could also specify the direction of that relationship, or other qualities such as whether or not a relationship is antagonistic.
In a sense, relational data can be seen as similar to descriptive statistics in traditional studies, but quantitative social network analyses can go a little further. Descriptive analysis techniques can also be applied to a social network representation, allowing for measurements of centrality, distance between nodes, and clustering. Predictive analysis will, of course, demand an appropriate research question, but, given an adequate representation of the social network, researchers will find that social network analysis yields significant predictive power.
Take, for example, the Richmond Police Department. Police departments across the country have employed a variety of predictive techniques in order to police more effectively, but, according to reporting by The Economist (2010), the RPD has included social network analysis within their toolkit. According to the article, Richmond police officers have to come to maintain relational data on suspects and their acquaintances. Furthermore, it is reported that they mine social media for data, especially in relation to predicting potential locations of parties. Of course, critics suggest that the end result of these techniques’ application has done nothing but reinforce problematic patterns of stereotyping and over-policing, but, according to the article, RPD and other departments employing similar predictive techniques claim to be able to reduce police labor-hours and, consequently, costs.
Keim, S. (2011). The Social Network Perspective. In Social Networks and Family Formation Processes (pp. 19–30). Hackensack NJ: VS Research.
Mining social networks: Understanding the social web. (2010, September). The Economist. Retrieved from http://www.economist.com/node/16910031