Sociology Social Network Analysis
by
Martin Everett, Nick Crossley, Elisa Bellotti
  • LAST REVIEWED: 28 November 2016
  • LAST MODIFIED: 28 November 2016
  • DOI: 10.1093/obo/9780199756384-0184

Introduction

Social network analysis (SNA) is a methodology for capturing, storing, visualizing and analyzing relational data; that is, data concerning relations between specified entities (e.g., individuals, organizations, nations) and patterns of connection within populations of such entities. As such it stands in contrast to most standard social scientific approaches, which typically focus upon the attributes of such entities (although attributes can be included in SNA). Interest in social relations, their properties and effects, stretches back to the origins of social science and indeed even further back to the earliest social philosophies, and the origins of SNA itself can be traced back at least as far as the 1930s. The perspective has enjoyed a huge boost in recent years, however, not least as advances in technology have increased the computing power routinely available to social scientists. There are a number of good software packages today which offer users a comprehensive range of powerful analytic routines. This review covers: (1) data collection, (2) network measures, (3) roles and positions, (4) ego networks, (5) statistical methods, (6) mixed methods, (7) social capital, (8) small worlds, (9) large, complex and multimode data, (10) visualization, and (11) cohesive subgroups/community detection.

  • Borgatti, S. P., M. G. Everett, and J. C. Johnson. 2013. Analyzing social networks. Thousand Oaks, CA: SAGE.

    A good introduction to SNA and to the Ucinet software package, written by the writers of the software. Clear and comprehensive.

  • De Nooy, W., A. Mrvar, and V. Batagelj. 2005. Exploratory social network analysis with Pajek. New York: Cambridge Univ. Press.

    A good introduction to SNA and to the Pajek software package, written by the writers of the software. Again, clear and comprehensive.

  • Kadushin, C. 2012. Understanding social networks. Oxford: Oxford Univ. Press.

    A strong introduction which differs in focus form the others listed here in the attention it devotes to linking SNA with substantive concerns, findings, and concepts from the wider social science literature.

  • Robins, G. 2015. Doing social networks research. London: SAGE.

    A state of the art introduction which is distinctive in the attention that it gives to data gathering and research design.

  • Scott, J. 2000. Social network analysis: A handbook. London: SAGE.

    A clear introduction to many of the ideas underlying SNA and many basic measures. The book makes a point of sidestepping mathematical formulations and offers a chapter on the history of the approach.

  • Wasserman, S., and K. Faust. 1994. Social network analysis: Methods and applications. Cambridge, UK: Cambridge Univ. Press.

    DOI: 10.1017/CBO9780511815478

    The standard reference text on social network analysis. Big (857 pp.) and comprehensive. In contrast to Scott this book makes a point of explaining the various mathematical formulations and formalization typically employed in SNA. It does so very clearly and without assuming much prior knowledge.

Data Collection

In social network analysis data can be collected for whole networks or egonets (see Ego Networks). When we analyze whole networks we identify a relevant population and, as far as possible, conduct a census survey of all members of that population, seeking to establish the existence or not of a relevant tie between each pair of nodes in that population. In addition to whole networks, network analysts sometimes analyze two-mode networks. In a two mode network we have two different types of nodes and ties exist only across these two types, not within them. For example, people attending events. An ego-network is the network of contacts (alters) that form around a particular individual or node (ego), sometimes with data regarding ties between alters. There are three main methods of data collection in social network analysis: name generators, which can be used for both whole networks and egonets; position generators and resource generators, which are best suited for egonet analysis. Generators can be embedded in various data collection tools like surveys, interviews, and concentric circles, where people are asked to position their alters in a target where ego lies in the middle, and the closer the circles the stronger the tie to alters. A number of works have introduced various methods for data collection, like Kahn and Antonucci 1980, Burt 1984, Lin and Dumin 1986, Van der Gaag and Snijders 2005, and Krackhardt 1987. Marsden 1990 provides a good summary of the various collection tools, while Killworth and Bernard 1976 and Campbell and Lee 1991 reflect upon aspects of validity and reliability in social networks data collection.

  • Burt, R. 1984. Network items and the general social survey. Social Networks 6:293–339.

    DOI: 10.1016/0378-8733(84)90007-8

    The article discusses a name generator technique for obtaining network data in social surveys, which has then been included in three US general social surveys (GSS 1985/1987/2004).

  • Campbell, K. E., and B. A. Lee. 1991. Name generators in surveys of personal networks. Social Networks 13:203–221.

    DOI: 10.1016/0378-8733(91)90006-F

    The study compares the characteristics of egocentric networks from four famous social network studies, and finds that network size, age and education heterogeneity, and average tie characteristics are most strongly affected by the name generator used, while network composition, and racial and sexual heterogeneity, are more invariant.

  • Kahn, R. L., and T. C. Antonucci. 1980. Convoys over the life course: Attachment, roles and social support. In Life-span development and behaviour. Edited by P. B. Baltes and O. G. Brim, 253–286. New York: Academic Press.

    This is the first empirical research that used the instrument of concentric circles to measure egonets of social support over the life course.

  • Killworth, P., and H. Bernard. 1976. Informant accuracy in social network data. Human Organization 35:269–286.

    DOI: 10.17730/humo.35.3.10215j2m359266n2

    The paper reports results of the first of seven experiments on informant accuracy in the production of social network data, which show that informants are extremely inaccurate. In other words, informants’ reports of their behavior bear little resemblance to their behavior. The paper discusses the implications of the findings for sociometric and network analysis.

  • Krackhardt, D. 1987. Cognitive social structures. Social Networks 9:109–134.

    DOI: 10.1016/0378-8733(87)90009-8

    The article proposes a method to measure whole networks that observes the congruence of the perception of networks people are embedded in. Networks so constructed are called cognitive social structures (CSS) and can be modeled as three-dimensional (N × N × N) network structures.

  • Lin, N., and M. Dumin. 1986. Access to occupations through social ties. Social Networks 8:365–385.

    DOI: 10.1016/0378-8733(86)90003-1

    The article proposes to measure access to occupations through social ties by using an instrument called the position generator. The tool asks the interviewee to indicate if they have a tie to anyone in a range of specified occupations and, if so, the type of involved. The paper shows that individuals in high social positions typically have strong ties to others in those positions while those in lower social positions, where they have such ties at all, have weaker ties.

  • Marsden, P. V. 1990. Network data and measurement. Annual Review of Sociology 16:435–463.

    DOI: 10.1146/annurev.so.16.080190.002251

    The article reviews all the main methods of data collection for social network analysis, including a discussion of data validity and reliability.

  • Van der Gaag, M., and T. Snijders. 2005. The resource generator: Social capital quantification with concrete items. Social Networks 27:1–29.

    DOI: 10.1016/j.socnet.2004.10.001

    The article proposes an instrument for the measurement of social capital in a population, called the Resource Generator, which is constructed with concretely worded items covering general social capital resources. Construction, use, and first empirical findings are discussed for a representative sample (N = 1004) of the Dutch population in 1999–2000.

back to top

Users without a subscription are not able to see the full content on this page. Please subscribe or login.

How to Subscribe

Oxford Bibliographies Online is available by subscription and perpetual access to institutions. For more information or to contact an Oxford Sales Representative click here.

Article

Up

Down