Analysis of Social Network Measures with Respect to Structural Properties of Networks

Social Network Analysis (SNA), the study of social interactions within a group, spans many different fields of study, ranging from psychology to biology to information sciences. Over the past half century, many analysts outside of the social science field have taken SNA concepts and theories and have applied them to an array of networks in the hope of formulating mathematical descriptions of the relations within the network of interest. More than 50 measures of networks have been identified across these fields; however, little research has examined the findings of these measures for possible relationships. This thesis tests a set of widely accepted SNA measures for correlation and redundancies with respect to the most accepted network structural properties: size, clustering coefficients, and scale-free parameters. The goal of the thesis is to investigate the SNA measures\u27 ability to discriminate and identify different actors in a network. As a result, the study not only identifies high correlation amongst many of the measures, it also aids analysts in identifying which measure best suits a network with specific structural properties, and the measure\u27s efficiency for a given analysis goal