network analysis

Show Summary Details

Quick Reference

The term network refers to individuals (or more rarely collectivities and roles) who are linked together by one or more social relationships, thus forming a social network. Examples of relationship links include kinship, communication, friendship, authority, and sexual contact. When individuals are represented as a point and links as a line, the pairwise choices or relations can be arrayed in a table (called a sociomatrix) and the network drawn from this information is often referred to as a sociogram, which features centrally in sociometry, an elaborate but simple form of analysis pioneered by J. L. Moreno in the 1940s. The mathematical basis of much network analysis is graph theory.

Initially, much work concentrated on small-group and institutional structures, describing individual points (stars and isolates) and forms of cohesion (clique-detection), but after the 1950s network analysis concentrated more on structural characteristics, such as ‘bridges’ (persons who formed the only link between strongly connected groups), ‘balance’ (the tendency of highly cohesive groups to polarize), and more refined definitions of cliques. In the 1960s and beyond, the analysis of social networks was strongly influenced by developments in mathematical sociology under Harrison White, became much more highly theoretical, and now supports its own journal (Social Networks). White was the focus of a highly productive and innovative group of students and staff at Harvard University in the 1960s and 1970s, and is best known for insisting upon social rather than individualistic concepts (for example, movement of clergy vacancies versus movement of individual clergymen), and for developing block-modelling techniques for studying the ‘structural equivalence’ of network members sharing the same pattern of contacts. Recent developments have been concerned with the issue of multi-dimensional scaling and similar methods to represent networks in social space and to use sophisticated computer techniques to depict these spaces visually.

Three main foci typify work in the area of network analyses. Egocentric networks are rooted in a single individual and depend usually on that individual's report of his or her network (such as, for example, E. Bott's study of the effect of overlap between spouse's networks, Family and Social Network, 1957). Systemic networks are constructed from all the participants in the network, and concentrate on the structure of the network itself, as in Mark Granovetter's identification of the importance of the ‘weak tie’ in obtaining new job information—whereby new information comes not from those who are in one's close circle of interaction, but from those in one's network who have access to different sources (see Getting a Job, 1974). Finally, diffusion studies explore the shape and form of flows within the networks, as in the processes of innovation, rumour, or epidemiological diffusion. For an overview of the field as a whole see John Scott, Social Network Analysis (2nd edn., 2000) and the more advanced discussion in Stanley Wasserman and Katherine Faust, Social Network Analysis (1993). See also balance theory.

http://www.insna.org/ A key site for resources on social network analysis.

Subjects: Sociology.

Reference entries

Users without a subscription are not able to see the full content. Please, subscribe or login to access all content.