Beyond measuring and visualising networks, this research paradigm has developed concepts to reflect networks and their structures. For example, in Figure 4.5, node E bridges two different groups together. If node E was removed, there would be a structural hole in the network; there would be no connection between the two groups. These terms help to describe potentially interesting social phenomena taking place in the network.
Network scholars also pay close attention to triangles caused by triadic closure observed in social networks. Both Simmel (1950) and Granovetter (1977) highlight humans from triadic networks. If A and B know each other, and B and C know each other, it is likely that A and C will know each other as well. For those interested in network structures, this means they can compute different types of triadic connections that exist in the network. The triadic census describes the types of three-member networks observed in the data. For example, in Figure 4.5 triangle ABC is complete, while triangle CDE is missing one connection. Triangle ADG has no connections; thus, it is an empty triangle. In directed graphs, the census also accounts for different types of directions to further understand the network structures. For example, social scientists have used different ways that organisations link to each other on the Web (and the types of triangles observed) to understand relationships between these organisationsfind ref