Sociology Social Network Analysis
by
Martin Everett, Nick Crossley, Elisa Bellotti
  • 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.

    Save Citation »Export Citation »E-mail Citation »

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

    Find this resource:

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

      Save Citation »Export Citation »E-mail Citation »

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

      Find this resource:

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

        Save Citation »Export Citation »E-mail Citation »

        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.

        Find this resource:

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

          Save Citation »Export Citation »E-mail Citation »

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

          Find this resource:

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

            Save Citation »Export Citation »E-mail Citation »

            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.

            Find this resource:

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

              DOI: 10.1017/CBO9780511815478Save Citation »Export Citation »E-mail Citation »

              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.

              Find this resource:

              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)90007-8Save Citation »Export Citation »E-mail Citation »

                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).

                Find this resource:

                • 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-FSave Citation »Export Citation »E-mail Citation »

                  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.

                  Find this resource:

                  • 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.

                    Save Citation »Export Citation »E-mail Citation »

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

                    Find this resource:

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

                      DOI: 10.17730/humo.35.3.10215j2m359266n2Save Citation »Export Citation »E-mail Citation »

                      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.

                      Find this resource:

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

                        DOI: 10.1016/0378-8733(87)90009-8Save Citation »Export Citation »E-mail Citation »

                        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.

                        Find this resource:

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

                          DOI: 10.1016/0378-8733(86)90003-1Save Citation »Export Citation »E-mail Citation »

                          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.

                          Find this resource:

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

                            DOI: 10.1146/annurev.so.16.080190.002251Save Citation »Export Citation »E-mail Citation »

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

                            Find this resource:

                            • 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.2004.10.001Save Citation »Export Citation »E-mail Citation »

                              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.

                              Find this resource:

                              Network Measures

                              A network comprises a set of nodes and a set (or sets) of ties connecting certain of those nodes. Defined thus it has a number of measurable (and sociologically meaningful) properties. These are typically defined at one of five levels: (1) the whole network, (2) subgroups, (3) individual nodes, (4) dyads, (5) triads. All of the main introductory textbooks outline and explain the most frequently used of these measures. Scott 2000 offers a largely “maths-free” account, informed by substantive sociological interests. Wasserman and Faust 1994, by contrast, offers a more mathematically based (sociologically focused) account, going to great lengths to explain the conventions typically employed in mathematical formalizations of network ideas in a clear way for the uninitiated reader. Borgatti, et al. 2013 combines exposition of basic measures with a practical, “how to” guide to the Ucinet software package, and De Nooy, et al. 2005 does the same for the Pajek software package. Key properties at the whole network level include: density (number of ties expressed as a proportion of the potential number), components (a component is a subset of nodes any of which has a path connecting it to any of the others), and clustering. Measuring components is also a way of detecting subgroups within a network. This can be useful if we are interested in diffusion because the absence of a path between components prevents flow. Other subgroup measures include some focused upon cohesion and others, discussed further in the section on Roles and Positions, focused upon equivalence of network position. Nodes acquire properties in virtue of their position within a network. The main example of this is centrality. Nodes can be more or less central within a network. There are many definitions of centrality, each capturing a different way in which a node might be central. The four main types are: degree, betweenness, closeness, and eigenvector centrality. Freeman 1978–1979 is a classic early discussion of different types of centrality. Borgatti 2005 offers a much needed clarification of differences in their application and usefulness. Dyadic and triadic properties are particularly important in statistical approaches to social network analysis. Typically analysts survey every possible pairing of nodes in a network or every possible combination of three nodes, recording their pattern of ties. The seminal paper on this approach and its rationale is Holland and Leinhardt 1975. Snijders 2011 brings these more statistically-informed ideas up to date.

                              • Borgatti, S. 2005. Centrality and network flow. Social Networks 27:55–71.

                                DOI: 10.1016/j.socnet.2004.11.008Save Citation »Export Citation »E-mail Citation »

                                An important paper which explores the implications of different types of centrality and encourages researchers to reflect upon these implications when selecting centrality measures in their projects.

                                Find this resource:

                                • Borgatti, S. P., and M. G. Everett. 2006. A graph theoretical perspective on centrality. Social Networks 28:466–484.

                                  DOI: 10.1016/j.socnet.2005.11.005Save Citation »Export Citation »E-mail Citation »

                                  An interesting theoretical reflection upon different types of centrality, their relative strengths and weaknesses, and the respective situations in which they might be most appropriate for use.

                                  Find this resource:

                                  • Borgatti, S. P., M. G. Everett, and J. C. Johnson. 2013. Analyzing social networks. Los Angeles: SAGE.

                                    Save Citation »Export Citation »E-mail Citation »

                                    A good introduction to social network analysis (SNA) and to the Ucinet software package, written by the writers of the software. Clear and comprehensive.

                                    Find this resource:

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

                                      Save Citation »Export Citation »E-mail Citation »

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

                                      Find this resource:

                                      • Faust, K. 2006. Comparing social networks: Size density and local structure. Metodoloski zvezki 3.2: 185–216.

                                        Save Citation »Export Citation »E-mail Citation »

                                        An interesting reflection on whole network measures (including the triad census) and their limitations, which also illustrates and discusses the issues involved when different networks are compared.

                                        Find this resource:

                                        • Freeman, L. 1978–1979. Centrality in social networks I: Conceptual clarification. Social Networks 1:215–239.

                                          DOI: 10.1016/0378-8733(78)90021-7Save Citation »Export Citation »E-mail Citation »

                                          This is a seminal paper addressing the issue of centrality in networks and defining several of its key types.

                                          Find this resource:

                                          • Holland, P. W., and S. Leinhardt. 1975. Local structure in social networks. In Sociological methodology. Edited by D. Heise, 1–45. San Fransisco: Jossey-Bass.

                                            Save Citation »Export Citation »E-mail Citation »

                                            The seminal paper on the idea of the triad census which explains both what it entails and the rationale for its use.

                                            Find this resource:

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

                                              Save Citation »Export Citation »E-mail Citation »

                                              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.

                                              Find this resource:

                                              • Snijders, T. 2011. Statistical models for social networks. Annual Review of Sociology 1–53.

                                                Save Citation »Export Citation »E-mail Citation »

                                                A clear introduction to the way in which statistical issues have been brought to bear in social network analysis.

                                                Find this resource:

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

                                                  DOI: 10.1017/CBO9780511815478Save Citation »Export Citation »E-mail Citation »

                                                  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.

                                                  Find this resource:

                                                  Roles and Positions

                                                  Early advocates of social network analysis (SNA) often equated social network structure with social structure more generally. Later writers, Nadel 1956 in particular, challenged this, however, arguing that social structure is a network of roles or positions (each occupied by a potentially large number of nodes) rather than of individual nodes. This begged the question whether SNA routines could be devised which identified such roles/positions and remodeled networks on this basis. Lorrain and White 1971 laid the foundation for doing this with their concept, “structural equivalence,” which identifies positions and their incumbents in a network. Any two nodes are structurally equivalent when they have ties (of the same type) to the same alters. Strictly they should have exactly the same ties to exactly the same alters. In practice, however, two nodes are usually deemed “structurally equivalent” when they have a similar profile of ties. White, et al. 1976 build upon this, and Boorman and White 1976 upon them, to explore ways of remodeling networks (blockmodeling) on this basis. This work has been very influential but, as Scott 2000 explains, the concept of “structural equivalence” upon which it is based, though it perhaps adequately operationalizes the idea of “position,” does not really capture what sociologists mean by “role.” Incumbents of the same position do not necessarily enjoy ties to the same alters. Indeed this is very unlikely. This problem has inspired various efforts to devise other, better ways of defining “equivalence.” Key among the alternatives is the concept of “regular equivalence.” Stephen Borgatti and Martin Everett have played a central role in the development of the regular equivalence concept and their paper on the topic, Borgatti and Everett 1992a, offers a particularly useful exposition, as well as suggesting important extensions and developments of it. They add further clarity in Borgatti and Everett 1992b, which compares the two conceptions of equivalence, drawing out and discussing their respective relevance for (different) concepts of “position.” Doreian, et al. 2005 offers the most up-to-date review of discussions on equivalence and blockmodeling, discussing several further revisions and refinements.

                                                  • Boorman, S., and H. White. 1976. Social structures from multiple networks. II: Role structures. American Journal of Sociology 81:1384–1446.

                                                    DOI: 10.1086/226228Save Citation »Export Citation »E-mail Citation »

                                                    The final installment of White’s early work on blockmodeling which concludes his early reflections on the method.

                                                    Find this resource:

                                                    • Borgatti, S. P., and M. G. Everett. 1992a. Regular blockmodels of multiway, multimode matrices. Social Networks 14:91–120.

                                                      DOI: 10.1016/0378-8733(92)90015-YSave Citation »Export Citation »E-mail Citation »

                                                      This paper elaborates upon the concept of regular equivalence, extending its use beyond straightforward, single-mode networks to more complex and multi-mode cases.

                                                      Find this resource:

                                                      • Borgatti, S. P., and M. G. Everett. 1992b. Notions of positions in social network analysis. Sociological Methodology: 1–35.

                                                        DOI: 10.2307/270991Save Citation »Export Citation »E-mail Citation »

                                                        Premised on the observation that the concept, “position,” is often confused in empirical social network analysis, this paper distinguishes different definitions of “position” (focusing chiefly upon the distinction between structural and regular equivalence), explores the different results yielded when one or the other is chosen in the context of empirical work, and makes recommendations for their use upon this basis.

                                                        Find this resource:

                                                        • Doreian, P., V. Batagelj, and A. Ferligoj. 2005. Positional analysis of sociometric data. In Models and methods in social network analysis. Edited by P. Carrington, J. Scott, and S. Wasserman, 77–97. Cambridge, UK: Cambridge Univ. Press.

                                                          DOI: 10.1017/CBO9780511811395.005Save Citation »Export Citation »E-mail Citation »

                                                          The most up-to-date, state of the art discussion of methods and models for positional analysis.

                                                          Find this resource:

                                                          • Lorrain, F., and H. White. 1971. Structural equivalence of individuals in social networks. The Journal of Mathematical Sociology 1.1: 49–80.

                                                            DOI: 10.1080/0022250X.1971.9989788Save Citation »Export Citation »E-mail Citation »

                                                            The classic early paper in which the concept of structural equivalence is introduced and discussed.

                                                            Find this resource:

                                                            • Nadel, S. 1956. The theory of social structure. London: Cohen and West.

                                                              Save Citation »Export Citation »E-mail Citation »

                                                              Not a work of social network analysis, as such, but a key theoretical inspiration for much early network analytic work on roles and positions.

                                                              Find this resource:

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

                                                                Save Citation »Export Citation »E-mail Citation »

                                                                A good general introduction to social network analysis, relevant here because Scott nicely draws out the failings of the concept of “structural equivalence” insofar as it is supposed to capture and operationalize the concept of “role.”

                                                                Find this resource:

                                                                • White, H., S. Boorman, and R. Breiger. 1976. Social structure from multiple networks I: Blockmodels and roles and positions. American Journal of Sociology 81:730–780.

                                                                  DOI: 10.1086/226141Save Citation »Export Citation »E-mail Citation »

                                                                  This paper to some extent follows on from the Lorrain and White paper, moving from the idea of structural equivalence to the method of blockmodeling (in its early and classic form).

                                                                  Find this resource:

                                                                  Ego Networks

                                                                  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, and with attribute data for both egos and alters. Egonet data can be collected using various generators. Generators can be embedded in numerous data collection tools like surveys, interviews, and concentric circles, often with a combination of such tools (see Data Collection). A comprehensive text on how to conduct egonet research is Crossley, et al. 2015, while a useful guide on how to analyze egonet data using the software E-NET is contained in Halgin and Borgatti 2012. Bellotti 2008 gives an example of analysis of friendship egonets, while Granovetter 1973 and Lubbers, et al. 2010 explore the function of egonets in processes of searching for jobs and settling into a new country (post migration) respectively. The papers in Mitchell 1969 and the articles of Wellman and Wortley 1990 and Fischer 1982 represent classic studies of egonets in urban settings.

                                                                  • Bellotti, E. 2008. What are friends for? Elective communities of single people. Social Networks 30:318–329.

                                                                    DOI: 10.1016/j.socnet.2008.07.001Save Citation »Export Citation »E-mail Citation »

                                                                    The article is an example of an egonet study that combines name generator and qualitative interviews. It discusses the friendship networks of twenty-three heterosexual singles of Milan and proposes four typologies of egonets derived from structural characteristics and individual narratives.

                                                                    Find this resource:

                                                                    • Crossley, N., E. Bellotti, G. Edwards, J. Koskinen, M. Everett, and M. Tranmer. 2015. Social network analysis for egonets. London: SAGE.

                                                                      Save Citation »Export Citation »E-mail Citation »

                                                                      A comprehensive introduction to egonet analysis, covering: data gathering, a range of analytic approaches (statistical and qualitative as well as more standard graph-theoretic approaches), a range of measures, and background theories.

                                                                      Find this resource:

                                                                      • Fischer, C. 1982. To dwell amongst friends. Chicago: Univ. of Chicago.

                                                                        Save Citation »Export Citation »E-mail Citation »

                                                                        Presents the results of one of the earliest and finest studies of personal networks in an urban context. Set in 1978 in fifty northern California communities, it asked 1,050 adults about their social relations, their communities, their personal backgrounds, and their feelings.

                                                                        Find this resource:

                                                                        • Granovetter, M. 1973. The strength of weak ties. American Journal of Sociology 78:1360–1380.

                                                                          DOI: 10.1086/225469Save Citation »Export Citation »E-mail Citation »

                                                                          The most cited article in sociology, Granovetter’s study explores the importance of weak ties in egonets in the process of getting a job.

                                                                          Find this resource:

                                                                          • Halgin, D. S., and S. P. Borgatti. 2012. An introduction to personal network analysis and tie churn statistics using E-NET. Connections 32.1: 37–48.

                                                                            Save Citation »Export Citation »E-mail Citation »

                                                                            A very useful guide introducing the software E-NET designed specifically for egonet analysis. It presents the basic steps for egonet data collection and reviews key measures of egonet analysis such as size, composition, and structure. It then introduces longitudinal measures of egonet change, including tie churn, brokerage elasticity, and triad change.

                                                                            Find this resource:

                                                                            • Lubbers, M. J., J. L. Molina, J. Lerner, U. Brandes, J. Ávila, and C. McCarty. 2010. Longitudinal analysis of personal networks. The case of Argentinean migrants in Spain. Social Networks 32.1: 91–104.

                                                                              DOI: 10.1016/j.socnet.2009.05.001Save Citation »Export Citation »E-mail Citation »

                                                                              The article discusses and illustrates various approaches for the longitudinal analysis of personal networks (multilevel analysis, regression analysis, and SIENA), and combines the different types of analyses in a study of the changing personal networks of twenty-five Argentineans in Spain interviewed at two time points (with a two-year interval).

                                                                              Find this resource:

                                                                              • Mitchell, J. C., ed. 1969. Social networks in urban situations. Manchester, UK: Manchester Univ. Press.

                                                                                Save Citation »Export Citation »E-mail Citation »

                                                                                A classic collection of essays that mix ethnographic observations with social networks. The chapters by Wheeldon and Boswell report on egonets studies, the first focusing on six leaders who were frequently named by other members of a community in Southern Africa, the second observing the mobilization of personal networks during periods of crisis in the African city of Lusaka.

                                                                                Find this resource:

                                                                                • Snijders, T., M. Spreen, and R. Zwaagstra. 1995. The use of multilevel modeling for analysing personal networks: Networks of cocaine users in an urban area. Journal of Quantitative Anthropology 5.2: 85–105.

                                                                                  Save Citation »Export Citation »E-mail Citation »

                                                                                  Explains how multilevel methods can be used to analyze egonets when the dependent variable is a function of the relations contained in such egonets. It illustrates the models in an empirical example of egonets of cocaine users.

                                                                                  Find this resource:

                                                                                  • Wellman, B., and S. Wortley. 1990. Different strokes from different folks. American Journal of Sociology 96.3: 558–588.

                                                                                    DOI: 10.1086/229572Save Citation »Export Citation »E-mail Citation »

                                                                                    One of the earliest and classic study of egonets in East York, which employs quantitative and qualitative data to evaluate six potential explanations of why different types of ties provide different kinds of supportive resources.

                                                                                    Find this resource:

                                                                                    Statistical Methods

                                                                                    Standard statistical methods can only be applied when the sampled units themselves are independent networks. This is rarely the case and it follows that specialist methods are needed to rigorously analyze social networks. This is a very active area of research and there are few books devoted to the topic with the exception of the edited volume Lusher, et al. 2013. In essence there are three classes of methods developed. The oldest is to use standard statistical techniques but to use permutation tests to obtain significance levels as suggested by Krackhardt 1987. The second approach is to develop log-linear type models specifically for networks, which are known as exponential random graph models. Finally Snijders 2001 has proposed a model that focuses on how an actor controls their outgoing ties over time with network panel data. These are known as actor oriented models.

                                                                                    • Butts, C. T. 2008. A relational event framework for social action. Sociological Methodology 38.1: 155–200.

                                                                                      Save Citation »Export Citation »E-mail Citation »

                                                                                      A relational event is one that is generated by an actor and directed to a receiver. This paper proposes a general statistical framework to analyze data in this form.

                                                                                      Find this resource:

                                                                                      • Dekker, D., D. Krackhardt, and T. A. B. Snijders. 2007. Sensitivity of MRQAP tests to collinearity and autocorrelation conditions. Pschometrika 72:563–581.

                                                                                        DOI: 10.1007/s11336-007-9016-1Save Citation »Export Citation »E-mail Citation »

                                                                                        In this paper a technique is introduced which corrects for collinearity and autocorrelation among the dependent network variables.

                                                                                        Find this resource:

                                                                                        • Frank, O., and D. Strauss. 1986. Markov graphs. Journal of the American Statistical Association 81:832–842.

                                                                                          DOI: 10.1080/01621459.1986.10478342Save Citation »Export Citation »E-mail Citation »

                                                                                          A statistical article that first introduces the basic ideas behind exponential random graph models.

                                                                                          Find this resource:

                                                                                          • Krackhardt, D. 1987. Predicting with networks—Nonparametric multiple-regression analysis of dyadic data. Social Networks 10:359–381.

                                                                                            DOI: 10.1016/0378-8733(88)90004-4Save Citation »Export Citation »E-mail Citation »

                                                                                            This article introduces the idea of using permutation tests in a multiple regression framework where both the dependent and independent variables are networks defined on the same sets of actors.

                                                                                            Find this resource:

                                                                                            • Lusher, D., J. Koskinen, and G. Robins, eds. 2013. Exponential random graph models for social networks: Theory, methods and applications. Cambridge, UK: Cambridge Univ. Press.

                                                                                              Save Citation »Export Citation »E-mail Citation »

                                                                                              An edited book that gives a detailed description of how to apply exponential random graph models in a variety of situations.

                                                                                              Find this resource:

                                                                                              • Robins, G., P. Pattison, Y. Kalish, and D. Lusher. 2007. An introduction to exponential random graph (p*) models for social networks. Social Networks 29:173–191.

                                                                                                Save Citation »Export Citation »E-mail Citation »

                                                                                                A clear and easy to follow introduction to exponential random graph models aimed at the practitioner.

                                                                                                Find this resource:

                                                                                                • Snijders, T. A. B. 2001. The statistical evaluation of social network dynamics. In Sociological methodology 2001. Edited by M. E. Sobel and M. P. Becker, 361–395. Boston and London: Basil Blackwell.

                                                                                                  Save Citation »Export Citation »E-mail Citation »

                                                                                                  The foundational paper in which the actor oriented models for network panel data are introduced.

                                                                                                  Find this resource:

                                                                                                  • Snijders, T. A. B., C. E. G. Steglich, and G. G. van de Bunt. 2010. Introduction to actor-based model for network dynamics. Social Networks 32:44–60.

                                                                                                    Save Citation »Export Citation »E-mail Citation »

                                                                                                    A clear and easy to follow introduction to actor-oriented models and the software package SIENA for such models aimed at the practitioner.

                                                                                                    Find this resource:

                                                                                                    Mixed Methods

                                                                                                    Studies are considered “mixed method” if they satisfy three conditions: they have to involve both quantitative and qualitative data; they have to employ both quantitative and qualitative analytical strategies; and data or analytical strategies have to be integrated at least in one stage of research, being data collection, data analysis, or data interpretation. Bellotti 2014 and Dominguez and Hollstein 2014 contain comprehensive accounts on how to conduct mixed methods research in social network analysis, also providing empirical examples. Edwards 2010 offers a useful overview of various approaches to mixed methods in social network analysis, while Crossley 2010 and Fuhse 2009 reflect upon the theoretical foundations of social networks that call for mixed methodologies. Bernardi, et al. 2007, and Curtis, et al. 1995 offer examples of mixed methods studies, respectively, in the fields of social influence in family choices and risk behaviors in drug consumption. Whyte 1943, Street Corner Society, represents one of the first and classic studies of social networks where qualitative methods were used to map an urban community.

                                                                                                    • Bellotti, E. 2014. Qualitative networks: Mixed methods in sociological research. London: Routledge.

                                                                                                      Save Citation »Export Citation »E-mail Citation »

                                                                                                      The book summarizes three case studies where social network analysis is combined with various qualitative methods. The first case looks at the role of friendship for single people mixing egonets with in-depth interviews; the second case investigates the social space and life of a street gang using ethnographic methods; and the third case analyzes research collaboration in philosophy incorporating statistical analysis of network properties with content analysis of research projects.

                                                                                                      Find this resource:

                                                                                                      • Bernardi, L., S. Keim, and H. von der Lippe. 2007. Social influences on fertility: A comparative mixed methods study in eastern and western Germany. Journal of Mixed Methods Research 1.1: 23–47.

                                                                                                        DOI: 10.1177/2345678906292238Save Citation »Export Citation »E-mail Citation »

                                                                                                        The article uses a mixed methods design to investigate the effects of social influence on family formation in a sample of eastern and western German young adults at an early stage of their family formation. In-depth interviewing, network charts, and network grids are used to map individual personal relationships and their influence on family formation decisions. In addition, an analysis of friendship dyads is provided.

                                                                                                        Find this resource:

                                                                                                        • Crossley, N. 2010. The social world of the network: Qualitative aspects of network analysis. Sociologica 1.

                                                                                                          Save Citation »Export Citation »E-mail Citation »

                                                                                                          The article reflects upon the value of mixing methods, and in particular of integrating qualitative and quantitative approaches, in social network analysis. It argues that each has different strengths and weaknesses but that these are broadly complimentary, thus supporting the argument for their combined use.

                                                                                                          Find this resource:

                                                                                                          • Curtis, R., S. Friedman, A. Neaigus, B. Jose, M. Goldstein, and G. Ildefonso. 1995. Street level markets: Network structure and HIV risk. Social Networks 17:229–249.

                                                                                                            DOI: 10.1016/0378-8733(95)00264-OSave Citation »Export Citation »E-mail Citation »

                                                                                                            Ethnographic and formal social network methods are used to define and situate three categories of injecting drug users—a core, an inner periphery, and an outer periphery—in a street-level drug market scene. Different locations in these network structures are shown to be associated with different levels of AIDS risk behaviors and of HIV infection rates.

                                                                                                            Find this resource:

                                                                                                            • Dominguez, S., and B. Hollstein, eds. 2014. Mixed methods social networks research: Design and applications. Cambridge, UK: Cambridge Univ. Press.

                                                                                                              Save Citation »Export Citation »E-mail Citation »

                                                                                                              The edited volume demonstrates the potential of mixed-methods designs for the research of social networks and the utilization of social networks for other research. It provides methodological concepts to guide mixed-methods network studies with precise research designs and methods to investigate social networks of various sorts. Each chapter describes the research design used and discusses the strengths of the methods for that particular field and for specific outcomes.

                                                                                                              Find this resource:

                                                                                                              • Edwards, G. 2010. Mixed-method approaches to social network analysis. ESRC National Centre for Research Methods Review Paper.

                                                                                                                Save Citation »Export Citation »E-mail Citation »

                                                                                                                An excellent review paper that considers the arguments for adopting a mixed-method approach to network analysis, firstly as they arise out of the existing research literature, and secondly as they have been highlighted in explicit theoretical debates about combining quantitative and qualitative data and analysis. By unpacking the different ways in which researchers have combined quantitative and qualitative methods in network projects it also seeks to provide some guidance for others on “how to” mix methods in social network analysis.

                                                                                                                Find this resource:

                                                                                                                • Fuhse, J. 2009. The meaning structure of social networks. Sociological Theory 27:51–73.

                                                                                                                  DOI: 10.1111/j.1467-9558.2009.00338.xSave Citation »Export Citation »E-mail Citation »

                                                                                                                  A theoretical article that proposes to view networks as sociocultural structures, as in configurations of social relationships interwoven with meaning. Social relationships as the basic building blocks of networks are conceived of as dynamic structures of reciprocal (but not necessarily symmetric) expectations between alter and ego. Through their transactions, alter and ego construct an idiosyncratic “relationship culture” comprising symbols, narratives, and relational identities. The conclusion offers an agenda for research on this meaning structure of social networks.

                                                                                                                  Find this resource:

                                                                                                                  • Whyte, W. F. 1943. Street corner society. Chicago: Chicago Univ. Press.

                                                                                                                    Save Citation »Export Citation »E-mail Citation »

                                                                                                                    A classic ethnographic study based on a three-and-a-half-year participant observation of a slum district in Boston, mostly inhabited by first- and second-generation immigrants from Italy. The book describes various groups and communities within the district: the first part contains detailed accounts of how local gangs were formed and organized; the second part describes the relations of social structure, politics, and racketeering in that district.

                                                                                                                    Find this resource:

                                                                                                                    Network Theories

                                                                                                                    In addition to its methodological contribution, social network analysis has made two important theoretical contributions to sociological debate in recent years. The first centres upon the idea of social capital; the second upon the idea of small worlds. These two ideas overlap to some extent and they have each been subject to exciting debate both amongst and beyond network specialists.

                                                                                                                    Social Capital

                                                                                                                    There is little consensus in sociology as to what constitutes “theory” and this has inevitably rubbed off, to some extent, upon social network analysis. Here, however, we define “network theory” as comprising rationally founded empirical testable claims about predominant network patterns, their generative mechanisms, and their causal effects. For purposes of brevity, moreover, we limit ourselves only to the most widely discussed. Much early work on networks was focused upon cohesion and its effects. James Coleman’s pioneering work on social capital, for example, argues that density and closure within a network foster the conditions under which trust and norms of cooperation, reciprocity, and the like will emerge (Coleman 1988). As early as 1973, however, Mark Granovetter’s celebrated work “The Strength of Weak Ties” was suggesting that cohesive groupings, at least for certain purposes, including information gathering, involve a great deal of redundancy and that social actors often get more useful information (in this case regarding job opportunities) from alters who fall outside of their (dense) circles of close friends (Granovetter 1973). A very similar idea, also celebrated, was developed in Burt 1992 on “structural holes.” A structural hole is a gap in a network between unconnected parties. Burt, positing a competing definition of social capital to his teacher, Coleman, argued that individuals who can fill such gaps benefit individually from doing so and also generate benefits for the parties whom they connect (for which they will be rewarded and thus further benefit). A later work, Burt 2005, suggested that both cohesion (or “closure”) and the bridging of structural holes (“brokerage”) were forms of social capital and that an optimum situation would involve both. Both Coleman and Burt understand social capital as an effect (or potential effect) of particular configurations of ties. An alternative account, still rooted in the social network analysis tradition but less focused upon patterns of ties, is that of Nan Lin. Lin 2002 conceptualizes social capital in terms of the indirect access to other resources that an individual enjoys in virtue of his or her network ties: for example, an individual who has influential friends may enjoy access to and/or use of their friends’ influence.

                                                                                                                    • Burt, R. 1992. Structural holes. Cambridge, MA: Harvard Univ. Press.

                                                                                                                      Save Citation »Export Citation »E-mail Citation »

                                                                                                                      The key statement of Burt’s “structural hole” theory. It involves theoretical argument and definition; an extensive empirical testing of its main ideas (supporting those ideas); and definitions of a number of egonet measures which Burt introduces in effort to capture structural holes and their effects.

                                                                                                                      Find this resource:

                                                                                                                      • Burt, R. 2005. Brokerage and closure. Oxford: Oxford Univ. Press.

                                                                                                                        Save Citation »Export Citation »E-mail Citation »

                                                                                                                        A later statement by Burt in which he considers the relative value of both his own approach to social capital and that of James Coleman—concluding that both have value and recommending their integration.

                                                                                                                        Find this resource:

                                                                                                                        • Coleman, J. 1988. Social capital in the creation of human capital. American Journal of Sociology 94:95–120.

                                                                                                                          DOI: 10.1086/228943Save Citation »Export Citation »E-mail Citation »

                                                                                                                          A clear and concise statement of Coleman’s definition and understanding of social capital which has been widely cited.

                                                                                                                          Find this resource:

                                                                                                                          • Granovetter, M. 1973. The strength of weak ties. American Journal of Sociology 78:1360–1380.

                                                                                                                            DOI: 10.1086/225469Save Citation »Export Citation »E-mail Citation »

                                                                                                                            The original paper by Granovetter to which all subsequent papers on this issue refer. Clear, concise, and still both insightful and rich today.

                                                                                                                            Find this resource:

                                                                                                                            • Lin, N. 2002. Social capital. Cambridge, UK: Cambridge Univ. Press.

                                                                                                                              Save Citation »Export Citation »E-mail Citation »

                                                                                                                              A further network-related approach to social capital which focuses upon the ways in which ties provide indirect access to resources which an actor does not directly hold.

                                                                                                                              Find this resource:

                                                                                                                              Small Worlds

                                                                                                                              Granovetter 1973 has also been important in so-called “small world” theory. The small world idea has a long history but its point of origin in social science is usually identified in the work of Stanley Milgram and his coworkers in the 1960s (especially Milgram 2004, first published in 1967). They developed a new methodology for surveying path lengths in a national population. Much of their original work was intended to explore the impact of social division upon social structure. They wanted to know how easily (or not) ties crossed racial, gender and other status lines. The headline for which their work is much better known, however, is the claim that any two individuals taken at random from within the US population are, on average, at only six degrees of separation from one another. This work has been taken up, more recently, by physicists working in the area of complexity science, particularly Watts 1999, Watts 2004, and Barabási 2003. Working with systems/networks comprising many millions of nodes they were puzzled as to the high level of coordination they observed. They assumed that path lengths would be very long in such networks and that this would both slow down the speed at which goods/information traveled and lead to degeneration of those goods/information, threatening coordination. Milgram, however, appeared to show that big networks can be characterized by short (average) path lengths. Replicating his work the physicists found a similar pattern in their own networks. But they then wanted to know how this counterintuitive finding was possible. A classic paper in mathematics had suggested that short average path lengths are commonly found in random graphs (networks with randomly assigned ties) but social networks are not random and ties tend to cluster, an issue discussed in a more sociological vein by Pool and Kochen 1978–1979. Watts 2004 suggested that the work of Granovetter 1973 provides an answer to this problem. Granovetter suggests that strong ties are transitive and therefore clustered but that most people also have weak ties which are not transitive and which “behave” in a random fashion. This, Watts was able to show (through simulations), is sufficient to create the small world effect. An alternative to Watts’s theory was suggested in Barabási 2003. Barabási posits that the small world effect is created by huge hubs to which a large proportion of nodes within the network are attached. As many nodes attach to these large hubs the hubs tend to link large numbers of other nodes to one another, shortening average path lengths. The huge inequalities in degree in such networks, between hubs and other nodes, is said to manifest as a scale-free distribution and this work is the source of the great interest in scale-free networks in much contemporary work on networks. Many of the key early papers applying the small world idea in natural science are reproduced in Newman, et al. 2006. Crossley 2008 suggests some ways in which these ideas might be borrowed back for sociology and reasons for doing so.

                                                                                                                              • Barabási, A. -L. 2003. Linked. New York: Plume.

                                                                                                                                Save Citation »Export Citation »E-mail Citation »

                                                                                                                                Another “pop science” book on small worlds but this time focusing upon the Barabási-Albert model. Like Six Degrees (Watts 2004) the book is sufficiently nuanced and sophisticated to serve as a useful introduction to the interest of and ideas regarding small worlds within the physics community.

                                                                                                                                Find this resource:

                                                                                                                                • Crossley, N. 2008. Small world networks, complex systems and sociology. Sociology 42.2: 261–277.

                                                                                                                                  DOI: 10.1177/0038038507087353Save Citation »Export Citation »E-mail Citation »

                                                                                                                                  A reflection upon the sociological significance of the small world debate and upon some of the areas of sociological interest where it is relevant.

                                                                                                                                  Find this resource:

                                                                                                                                  • Granovetter, M. 1973. The strength of weak ties. American Journal of Sociology 78:1360–1380.

                                                                                                                                    DOI: 10.1086/225469Save Citation »Export Citation »E-mail Citation »

                                                                                                                                    The original paper by Granovetter which proved a key inspiration to Watts and many others who followed him. Clear, concise, and still both insightful and rich today.

                                                                                                                                    Find this resource:

                                                                                                                                    • Milgram, S. 2004. The small world problem. In Empirical approaches to sociology. Edited by G. Carter, 111–118. Boston: Pearson.

                                                                                                                                      Save Citation »Export Citation »E-mail Citation »

                                                                                                                                      One of several papers in which Milgram explains his small world experiment and its “six degrees of separation” conclusion. This is the most general of the papers and therefore the best firsthand introduction to Milgram. Originally published in 1967.

                                                                                                                                      Find this resource:

                                                                                                                                      • Newman, M., L. Barabási, and D. Watts. 2006. The structure and dynamics of networks. Princeton, NJ: Princeton Univ. Press.

                                                                                                                                        Save Citation »Export Citation »E-mail Citation »

                                                                                                                                        An important collection of classic and contemporary papers on the small world debate, albeit biased toward the interests of the physicists.

                                                                                                                                        Find this resource:

                                                                                                                                        • Pool, I., and M. Kochen. 1978–1979. Contacts and influence. Social Networks 1:5–51.

                                                                                                                                          DOI: 10.1016/0378-8733(78)90011-4Save Citation »Export Citation »E-mail Citation »

                                                                                                                                          A further early intervention in the small world debate, on the social science side. Not very widely cited but quite wide ranging, insightful, and clear. Useful for putting the work of Milgram into more context before turning to the physicists.

                                                                                                                                          Find this resource:

                                                                                                                                          • Schnettler, S. 2009. A structured overview of 50 years of small-world research. Social Networks 31.3: 165–178.

                                                                                                                                            DOI: 10.1016/j.socnet.2008.12.004Save Citation »Export Citation »E-mail Citation »

                                                                                                                                            See also S. Schnettler, A small world on feet of clay? Social Networks 31.3: 179–189. Two connected papers which offer a very comprehensive overview of the small world debate from the perspective of social network analysis, and which offer reason for us to be cautious in our appropriation of these ideas.

                                                                                                                                            Find this resource:

                                                                                                                                            • Watts, D. 1999. Small worlds. Princeton, NJ: Princeton Univ. Press.

                                                                                                                                              Save Citation »Export Citation »E-mail Citation »

                                                                                                                                              A book-length statement of Watts’s position on the small world debate and its relevance to issues in complexity theory. Quite heavy-going for the non-mathematician but interesting because, as a book, it draws out and discusses issues in more depth than one finds in the articles that Watts also wrote on this topic during this period. For an easier read see Watts 2004.

                                                                                                                                              Find this resource:

                                                                                                                                              • Watts, D. 2004. Six degrees. London: Vintage.

                                                                                                                                                Save Citation »Export Citation »E-mail Citation »

                                                                                                                                                A “pop science” take on the small world idea and its relevance for complexity theory (and, indeed, social science) but one which maintains sufficient nuance and sophistication to be of interest to social scientists. Develops many of the ideas much further than Watts’s Small Worlds (Watts 1999) and much further than any of his contemporaneous papers.

                                                                                                                                                Find this resource:

                                                                                                                                                Large, Complex, and Multimode Data

                                                                                                                                                The type of data we have dictates what methods can be applied. Multimode data is data in which there are different types of nodes and nodes of the same type do not interact with each other. Large and complex data are more difficult to define but in essence it is when the data by virtue of its size restricts what methods can be used. Most work on multimode data has been on two-mode data traditionally analyzed by converting to two one-mode datasets as first proposed by Breiger 1974. Borgatti and Everett 1997 suggested methods for directly analyzing two-mode data and until recently this has been the approach taken by most analysts. In a recent paper, Everett and Borgatti 2013 suggested a method which they call dual projection which is similar to the original Breiger approach. Most of the social network research undertaken by social scientists has been on relatively simple small datasets. The advent of big data and the interest in networks from the physics community has pushed social network research in new directions. Christakis and Fowler 2009 examined the spread of smoking, obesity, and happiness through what they called a large social network. For social scientists Pajek is the tool for large-scale network analysis and the recent book by some of the authors of Pajek gives a comprehensive insight into techniques for analyzing large networks. There are a number of sources of large networks and scientific collaboration networks have often been analyzed and Newman 2001 is an excellent example. Given large networks we can examine different properties, and the idea of scale-free networks proposed by Barabási and Albert 1999 is an example of this although most social networks are not of this form.

                                                                                                                                                • Barabási, A. L., and R. Albert. 1999. Emergence of scaling in random networks. Science 286.5439: 509–512.

                                                                                                                                                  DOI: 10.1126/science.286.5439.509Save Citation »Export Citation »E-mail Citation »

                                                                                                                                                  A seminal paper exploring the idea of scale-free networks.

                                                                                                                                                  Find this resource:

                                                                                                                                                  • Batagelj, V., P. Doreian, A. Ferligoj, and N. Kejzar. 2014. Understanding large temporal networks and spatial networks: Exploration, pattern searching, visualization and network evolution. Chichester, UK: Wiley.

                                                                                                                                                    DOI: 10.1002/9781118915370Save Citation »Export Citation »E-mail Citation »

                                                                                                                                                    A book describing in detail how to analyze large complex data drawn from a number of different sources from citation networks to the transfer of football players in the English Premier League.

                                                                                                                                                    Find this resource:

                                                                                                                                                    • Borgatti, S. P., and M. G. Everett. 1997. Network analysis of 2-mode data. Social Networks 19:243–269.

                                                                                                                                                      DOI: 10.1016/S0378-8733(96)00301-2Save Citation »Export Citation »E-mail Citation »

                                                                                                                                                      This paper describes how to analyze two-mode data as if it is one mode by applying corrections and adjustments to the standard one-mode concepts.

                                                                                                                                                      Find this resource:

                                                                                                                                                      • Breiger, R. L. 1974. The duality of persons and groups. Social Forces 53:181–190.

                                                                                                                                                        Save Citation »Export Citation »E-mail Citation »

                                                                                                                                                        A classic paper that introduces the connection between the projected person-by-person matrix and the group-by-group matrix.

                                                                                                                                                        Find this resource:

                                                                                                                                                        • Christakis, N. A., and J. H. Fowler. 2009. Connected: The surprising power of our social networks and how they shape our lives. New York: Little, Brown.

                                                                                                                                                          Save Citation »Export Citation »E-mail Citation »

                                                                                                                                                          A simple introduction for the lay reader that draws on the authors published academic studies mostly taken from what they term large social networks.

                                                                                                                                                          Find this resource:

                                                                                                                                                          • Clauset, A., M. E. Newman, and C. Moore. 2004. Finding community structure in very large networks. Physical Review E 70.6: 066111.

                                                                                                                                                            DOI: 10.1103/PhysRevE.70.066111Save Citation »Export Citation »E-mail Citation »

                                                                                                                                                            A physics paper that introduces a very fast algorithm for finding communities (cohesive subgroups) in large networks. It is demonstrated on a network of 400,000 nodes and over two-million edges.

                                                                                                                                                            Find this resource:

                                                                                                                                                            • Everett, M. G., and S. P. Borgatti. 2013. The dual projection approach for 2-mode networks. Social Networks 35:204–210.

                                                                                                                                                              DOI: 10.1016/j.socnet.2012.05.004Save Citation »Export Citation »E-mail Citation »

                                                                                                                                                              This paper argues that the data loss attributed to standard one-mode projections can be negated by recombining the results of the analysis of the two one-mode projections. This is illustrated for core-periphery, structural equivalence, and centrality.

                                                                                                                                                              Find this resource:

                                                                                                                                                              • Newman, M. E. 2001. Scientific collaboration networks. II. Shortest paths, weighted networks, and centrality. Physical Review E 64.1: 016132.

                                                                                                                                                                DOI: 10.1103/PhysRevE.64.016132Save Citation »Export Citation »E-mail Citation »

                                                                                                                                                                This paper reveals how the physics community goes about analyzing a large complex network of scientific collaboration drawn from a number of sources. The way the network is constructed and the analysis that is done are highly informative.

                                                                                                                                                                Find this resource:

                                                                                                                                                                Visualization

                                                                                                                                                                The visualization of social networks has always been seen as one of the subject’s most attractive features. Freeman 2000 gives an historic account of the development of the subject. McGrath, et al. 1997 nicely demonstrates the limitations of human perception and so emphasizes the importance of rigorous methods. The software package Pajek and UCINET/Netdraw are the most used packages for network analysis by nonspecialist social scientists and have a variety of graphical outputs and the two books connected with the software give many examples of how to do practical visualization. Krempel 2011 gives more detail looking at issues such as color and shading all from a social science perspective. There are numerous more sophisticated tools for visualizing networks, with more appearing all the time, and the subject clearly spills into both mathematics and computer science. Understanding the principles behind the visualization (without deep knowledge of the detailed algorithms) is essential when doing any analysis.

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

                                                                                                                                                                  Save Citation »Export Citation »E-mail Citation »

                                                                                                                                                                  A book linked to the software package UCINET/Netdraw with a chapter devoted to visualization, but many of the chapters contain details of different visualization methods. Not a manual, but there is an online document that shows how to do all the examples with the software.

                                                                                                                                                                  Find this resource:

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

                                                                                                                                                                    Save Citation »Export Citation »E-mail Citation »

                                                                                                                                                                    An accessible introduction to the software Pajek which has a number of sophisticated visualization tools. The book provides detailed descriptions of the commands required to produce the graphics.

                                                                                                                                                                    Find this resource:

                                                                                                                                                                    • Freeman, L. C. 2000. Visualizing social networks. Journal of Social Structure 1.

                                                                                                                                                                      Save Citation »Export Citation »E-mail Citation »

                                                                                                                                                                      An online article that traces the development of network visualization from the earliest hand-drawn images through to modern developments.

                                                                                                                                                                      Find this resource:

                                                                                                                                                                      • Jünger, M., and P. Mutzel, eds. 2004. Graph drawing software. Berlin: Springer.

                                                                                                                                                                        Save Citation »Export Citation »E-mail Citation »

                                                                                                                                                                        A book with fourteen contributed chapters describing state of the art software for visualization from general toolkits to software designed for specific applications.

                                                                                                                                                                        Find this resource:

                                                                                                                                                                        • Krempel, L. 2011. Network visualization. In The Sage handbook of social network analysis. Edited by J. Scott and P. Carrington, 558–577. London: SAGE.

                                                                                                                                                                          Save Citation »Export Citation »E-mail Citation »

                                                                                                                                                                          A comprehensive review by a social scientists of the factors and issues in producing good visualizations. Color versions of the figures are available online.

                                                                                                                                                                          Find this resource:

                                                                                                                                                                          • McGrath, C., J. Blythe, and D. Krackhardt. 1997. The effect of spatial arrangement on judgments and errors in interpreting graphs. Social Networks 19.3: 223–242.

                                                                                                                                                                            DOI: 10.1016/S0378-8733(96)00299-7Save Citation »Export Citation »E-mail Citation »

                                                                                                                                                                            This paper reports on an experiment in which different layouts of the same network are shown to subjects and they are asked to draw conclusions.

                                                                                                                                                                            Find this resource:

                                                                                                                                                                            • Tamassia, R., ed. 2013. Handbook of graph drawing and visualization. Boca Raton, FL: CRC.

                                                                                                                                                                              Save Citation »Export Citation »E-mail Citation »

                                                                                                                                                                              A collection of twenty-six contributed chapters giving state of the art information on graph drawing. The book is written more for computer scientists and mathematicians but covers important aspects of social networks.

                                                                                                                                                                              Find this resource:

                                                                                                                                                                              Cohesive Subgroups/Community Detection

                                                                                                                                                                              The identification of areas of the network in which connections between nodes is relatively dense was introduced by Luce and Perry 1949. The area became known as cohesive subgraph analysis. An important contribution was made by Mokken 1979 and at this time a number of cohesive subgraph concepts were developed. These are discussed in the classic book on network analysis, Wasserman and Faust 1994. In analyzing real data many of these concepts are of limited use and the relevant chapters in the books by Borgatti, et al. 2013 and De Nooy, et al. 2005 give practical guidance on dealing with real data. In 2002 the physics community became more interested in social networks and independently introduced their own terminology, community detection, for cohesive subgroups. Some of their work reproduced work already known to the social network community but they also introduced new ideas; for example, the Girvan-Newman algorithm. Fortunato 2010 provides a comprehensive review of the contributions of the physics community in his report.

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

                                                                                                                                                                                Save Citation »Export Citation »E-mail Citation »

                                                                                                                                                                                A good introduction to networks with a chapter devoted to cohesive subgroups which contains a step-by-step guide on how to find subgroups. Chapters on multivariate techniques and two-mode data describe additional methods applicable to different types of data.

                                                                                                                                                                                Find this resource:

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

                                                                                                                                                                                  Save Citation »Export Citation »E-mail Citation »

                                                                                                                                                                                  An accessible introduction to the software Pajek which has a chapter on cohesive subgraphs but also contains other clustering techniques as well as methods for signed graphs.

                                                                                                                                                                                  Find this resource:

                                                                                                                                                                                  • Fortunato, S. 2010. Community detection in graphs. Physics Reports 486.3: 75–174.

                                                                                                                                                                                    DOI: 10.1016/j.physrep.2009.11.002Save Citation »Export Citation »E-mail Citation »

                                                                                                                                                                                    A comprehensive review of community detection and cohesive subgroups. This article focuses more on the contributions of the physics community but does draw in papers from the social sciences. It is fairly technical and requires a certain level of mathematical sophistication.

                                                                                                                                                                                    Find this resource:

                                                                                                                                                                                    • Girvan, M., and M. E. Newman. 2002. Community structure in social and biological networks. Proceedings of the National Academy of Sciences 99.12: 7821–7826.

                                                                                                                                                                                      DOI: 10.1073/pnas.122653799Save Citation »Export Citation »E-mail Citation »

                                                                                                                                                                                      The first of many papers by the physics community. This paper introduces a new technique for finding cohesive subgroups along with a measure (modularity) on how good a particular partition is.

                                                                                                                                                                                      Find this resource:

                                                                                                                                                                                      • Luce, R. D., and A. D. Perry. 1949. A method of matrix analysis of group structure. Psychometrika 14.2: 95–116.

                                                                                                                                                                                        DOI: 10.1007/BF02289146Save Citation »Export Citation »E-mail Citation »

                                                                                                                                                                                        Seminal paper introducing the concept of a clique and describing a matrix technique and hence algorithm for their detection.

                                                                                                                                                                                        Find this resource:

                                                                                                                                                                                        • Mokken, R. J. 1979. Cliques, clubs and clans. Quality & Quantity 13.2: 161–173.

                                                                                                                                                                                          DOI: 10.1007/BF00139635Save Citation »Export Citation »E-mail Citation »

                                                                                                                                                                                          A paper looking at how the ideas of cliques could be extended in a way that they still had internal cohesiveness.

                                                                                                                                                                                          Find this resource:

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

                                                                                                                                                                                            DOI: 10.1017/CBO9780511815478Save Citation »Export Citation »E-mail Citation »

                                                                                                                                                                                            The standard reference text on social network analysis, slightly dated but contains definitions and explanations of the majority of cohesive subgroup concepts up to that time.

                                                                                                                                                                                            Find this resource:

                                                                                                                                                                                            back to top

                                                                                                                                                                                            Article

                                                                                                                                                                                            Up

                                                                                                                                                                                            Down