Edited by Nick Crossley (University of Manchester: nick.crossley@manchester.ac.uk), Christina Prell (University of Sheffield: c.prell@sheffield.ac.uk), and John Scott (University of Plymouth: john.scott@plymouth.ac.uk)

This special issue on social networks comes out of a National Research Methods Centre/ESRC funded seminar series, led by Nick Crossley, and co-led by John Scott and Christina Prell. The seminar series focused around various themes and issues pertaining to social networks. Part of this series’ aim was to reinvigorate UK interest in a field that had, in the past, been significantly shaped by the contributions of UK scholars, but in more recent times had been eclipsed by the efforts of US and European scholars.

In the past decade, two topics have generated much interest in the idea of social networks and network analysis. These are social capital, popularised by Robert Putnam, and small-worlds, popularised by Duncan Watts and Albert-László Barabási. Social capital highlights local processes and network structures, theorizing the ways in which relations and their patterns link individuals and groups to resources and beneficial outcomes. Small-worlds emphasizes global network structures, describing how large, heterogeneous networks can nonetheless appear small to individual actors, largely as a result of the high clustering and weak, bridging ties that make up these networks’ structure. Although social capital and small-worlds share social networks as a common basis, they emphasize different sides of a spectrum: social capital focuses on the local and small-worlds on the global. In addition, both focus on seemingly different social phenomena: social capital emphasizes access to resources, whereas small-worlds emphasize the tension of actors living in a social world that is simultaneously large and small. In spite of these differences, the literature points towards overlaps in the ways in which network structure is described: both social capital and small-worlds discuss structures of openness and closure, and these structural overlaps provide a means by which to start exploring, on a theoretical level, additional ways in which to bring about a synthesis of the two bodies of literature. In this paper, I situate social capital as an explanatory framework for the emergence of small-worlds. I do this through three phases: first, I discuss how each topic describes and theorizes opennenss and closure. Next, I develop a series of propositions that show how social capital can be linked to small-worlds in a coherent framework. Finally, I offer an empirical illustration of these propositions through the use of p*, one of the models from the larger family of exponential random graph models (ERGMs), which allow analysts to test the probability of certain local structural tendencies in a given network.

Key words:

This paper uses a social network approach to examine gender clustering in a complete network of teenagers and their friends.  It demonstrates the advantages of using increasingly sophisticated social network techniques, including clustering coefficients and their visualization, and social selection models within the ERGM framework, to visualize and explain the process of clustering which takes place in teenagers’ networks.  The paper supports previous findings of gender homophily among teenagers in small cliques of friends, provides evidence of clustering among larger groups of friends that differs by gender and evidence that the process of clustering also differs by gender. Males make more friends and form larger clusters than females.  Differences in clustering is due to differences in selection (males make more friends), triadic closure (more likely for females) and endogenous effects (impacting more on males).   These findings have sociological implications for single-gender and cross-gender influences on teenagers’ behaviour, and for the presumed importance of agency (selection) over structure (endogenous effects) on friendship formation. 

In this paper we discuss the personal network or ‘ego-net’ of Helen Kirkpatrick Watts, a militant suffragette and one of the founder members of the Nottingham Women’s Social and Political Union (WSPU). Our aims are fourfold. Firstly, we want to make a contribution to the growing literature, briefly reviewed, on the significance of social networks in relation to social movements. Secondly, we want to make a contribution to the academic literature on the suffragettes. Thirdly, at a methodological level, we want to contribute to on-going efforts, briefly reviewed, to bring qualitative concerns and issues back into social network analysis, generating a dialogue between quantitative and qualitative approaches. Finally, we conceive of this paper as a pilot for a much larger study of suffragette networks. Specifically, it allows us to experiment with ways of drawing network data from an archive and, relating to our third aim, to dry run an approach to network analysis which integrates quantitative and qualitative approaches.

Symbolic interactionism is generally known as a theory typically linked with a qualitative methodology. Recent developments in quantitative social network analysis, however, can analyze processes theorized within this theoretical tradition. Thick description can be complemented with statistical analyses of network structure and dynamics, expanding the scope and detail of results. This paper argues that social network analysis can bridge the divide between qualitative and quantitative analysis. Results from a long-term project on literary criticism substantiate the argument and illustrate a major development in social network analysis.

Key words: symbolic interactionism ; social network analysis ; actor-based models ; literary criticism

A common goal when analyzing a social network is to try and determine the cohesive subgroup structure of the network. Some techniques result in complex overlapping structures such as cliques or k-plexes whereas others either partition the network or place actors into unique groups, for example factions or components. A standard method implemented in Ucinet uses the overlapping structure to construct a proximity matrix which can be submitted to a clustering routine to find non-overlapping groups. Once the groups have been determined many analysts relate these to observations they have made involving their data. There are few techniques that take the groups as a starting point for additional analysis with the exception of Krackhardt’s work on Simmelian ties. In this paper I examine the roles the actors play in the cohesive structure using the brokerage ideas of Gould and Fernandez.
The method is demonstrated on an original dataset that was collected in 2006 on 100 social services based in a suburb area of Milan, Italy. The dataset is composed of a matrix that contains directed data collected by asking the spokesperson of every social service if he/she knows the other services listed (the list of services having been previously compiled using several sources), together with attribute information. The matrix is analyzed using the clique analysis techniques in the Ucinet software package to find non-overlapping groups. The results are submitted to Ucinet’s brokerage routine and visualized using the Netdraw software package. This analysis offers a new insight into the data, as it is possible to identify what actors play important roles (and which kind of role) in bridging the gaps between the cohesive subgroups.