|Volume 29, Issue 1, 2009|
|Table of Contents, 2-3|
|The Dutch Soccer Team as a Social Network, 4-14|
, van Kesteren,Frank,
, de Koning,Tim,
Although being very popular all around the globe, soccer has not received much attention from the scientific community. In this paper we will study the Dutch Soccer Team from the perspective of complex networks. In the DST network every node corresponds to a player that has played an official match for the Dutch Soccer Team. A node is connected with another node if both players have appeared in the same match. The aim of this paper is to study the topological properties of the Dutch Soccer Team network. The motivation for studying the DST network is twofold. The first reason is the immense popularity of the DST, in the Netherlands. Through our study we obtain all kind of new statistics about the DST. Secondly, our results could also be used by the coach of the DST, for instance by determining the optimal line-up. Using data available from a public website we have computed the topological metrics for the DST. Furthermore, we have looked at the
evolution of the topological metrics over time and we compared them with those of other real-life networks and of generic network models. We found that the DST is a small world network and that the player with the highest degree also has the lowest clustering coefficient.
|The Measurement of Social Networks: A Comparison of Alter-Centered and Relationship-Centered Survey Designs, 15-25|
Utilizing two surveys administered to a classroom of college students, this study explores differences in social network measures based on survey instrument design. By administering both a relationship-centered survey and an alter-centered survey, we analyze differences in range, mean numbers of relationships, network centralization, and network density.
Nonparametric tests are also used to discern patterns of similarity and difference. We find that measurement differences are often negligible when asking about extremely close relationships like friendship. However, differences often appear when studying “weak tie” types of relationships such as recognition of classmate names or acquaintances.
|Using SAS to Calculate Betweenness Centrality, 26-32|
Betweenness centrality is a useful measure of an actor’s importance in a social network. The SAS PROC IML module presented in this paper facilitates the calculation of betweenness centrality by social scientists by making it possible to run a faster algorithm for betweenness centrality using popular statistical software. The algorithm and module could be extended in order to calculate betweenness centrality for weighted graphs or to calculate other network measures that are based on geodesics, such as closeness centrality, graph centrality, or radiality.
|Different States, Choice, Structure and Aggregation in Simulated Social Networks, 33-42|
The fabric of society lies in the networks of connections and patterns of communication that its members create deliberately or inadvertently. Information, ideas, values and norms are passed across this fabric and members can form aggregates or allegiances that centre on common interests, goals, attitudes and the like. Simple multi-agent models of social networks have provided useful insights into the emergence of global network behavior where agents have limited or binary choices of state. This paper examines the impact that a greater number of choices of state have on the emergence of clusters of agents. It examines the global behavior of static populations of interacting computational agents, connected in fixed networks structures that are faced with multiple choices of state. Results indicate that aggregation around a few states appears to be a universal property, independent of network structure.
|Co-Citation of Prominent Social Network Articles in Sociology Journals: The Evolving Canon, 43-64|
Social network analysis has been a particularly hot area across the social (and some non-social) sciences. How has this growth, in turn, affected the field of social network analysis within sociology, the discipline which has served as the primary home of social network analysis over the last several decades? In order to answer this question, we examined the citation patterns of the social network papers in the two leading general sociology journals, the American Sociological Review and the American Journal of Sociology, from 1990-2005, focusing on the body of literature that was cited by at least two social network papers in a given year. We produced two network snapshots of the social network canon during this period.
These analyses reveal a combination of great change and substantial continuity. There was a substantial increase in interest in social networks in sociology throughout this period, and, in particular, an enormous rise in interest in small world issues, coupled with the abrupt entry of mathematicians and physicists into the sociology social network canon. However, during this entire period Granovetter’s work remained squarely at the center of the canon, with Granovetter (1973) as the most cited piece at both the earlier and later snapshots.
|Volume 29, Issue 2, 2009|
|Letters, Author Instructions, 2|
|Table of Contents, 3|
|Are We Treating Networks Seriously? The Growth of Network Research in Public Administration & Public Policy, 4-17|
The purpose of this research is to explore how the term ‘network’ is used in public administration and public policy. Since O’Toole (1997) first called for scholars of public administration and policy to “[treat] networks seriously,” a growing number of researchers use the term network as if it is a rising fashion trend. A recent article by Berry et al (2004) in Public Administration Review “Three Traditions of Network Research”, illustrates this trend. This article empirically examines the influence of a few prominent scholars on network research over the last decade. Subgroups of network research articles and authors in the citation network are also identified to illustrate the subtopics in network research and to probe what the term network means in these studies. The goal of this research is, in part, to answer Rethemeyer’s (2005) call for an empirical examination of network management. Secondly, this article aims to advance the understanding and use of methodology in the public administration discipline by showcasing the use of citation network analysis.
|The Structure of Undergraduate Association Networks: A Quantitative Ethnography, 18-31|
The challenge of collecting complete associational networks has restricted network studies to small datasets. To deal with larger processes, two general procedures have been developed: the use of indicators such as citation structures or the diffusion of innovations to model human interactions, and limiting the sample of associates' names. A body of theoretical and empirical work has identified several problems with these methods. We examine a unique solution to these problems—measuring online social networks of college students. In this paper we present an original network dataset of undergraduate Facebook users and demonstrate the feasibility and acceptability of this form of measurement. We conclude with a preliminary exploration of Network Homophily and Multiplexity on Facebook.
|Productivity and Performance in Academic Networks: Applications of Liaison Communication to Simmelian Ties, Structural Holes, and Degree Centrality, 32-44|
In 1968, Donald Schwartz completed what is now seen as the first network analysis performed in the field of communication (Rogers, 1994). The results found in this paper confirm the significance of Schwartz’ (1968) original research and extend his research to findings associated with the performance and productivity of academic researchers. The ability to retest the data collected by Schwarz in 1968 is a testament to his methods and processes, while new processes, such as a unique measure of Simmelian ties, were developed and utilized in this study. The similarity of the perceived and demographic data across three dimensions, Simmelian tie, structural holes, and degree centrality, not only support the original research but also provide insights into the effects of structure and position on performance and perception of academic networks. Findings related to categorical demographic data, rank and gender, offer a view into the nature of academic organizational networks and help tell their story. Structural holes (constraints) were found to decrease as tenure increased in an educational context, contrary to Burt's (1992b) findings and in support of Susskind et al.'s (1998) findings. This finding is explained as a combination of the level of seniority of the respondents and general organizational structure. The current research highlights the ability of network analysis to reveal organizational structure via communication linkages.
|Identifying Organizational Influentials: Methods and Application using Social Network Data, 45-61|
Uncovering the most influential individuals in an organization may be of great use for researchers and practitioners. As central hubs in the organization, these individuals can be key co-creators or co-adapters for the diffusion of organizational reform. In this paper we examine the question “Who are the most influential individuals in an organization?” Using social network data, we assess organizational members’ levels of influence in four different advice-seeking networks, as well as in one “friendship” network through a measure of peer-endorsement. We investigate four methods for the classification of individuals as “influentials”. These methods are compared and contrasted according to their performance in handling problems of differing network sizes, densities, non-response rates, researcher decisions and parsimony. A nonparametric random permutation method is shown to be a consistent and objective process for the identification of influential individuals in a sample of High school staff members in nine schools.
|Node Discovery Problem for a Social Network, 62-76|
A node discovery problem is defined as a problem in discovering a covert node within a social network. The covert node is a person who is not directly observable. The person transmits influence to neighbors and affects the resulting collaborative activities (e.g. meetings) within a social network, but does not appear in any information reported by the intelligence. Throughout this study, the information comes from data that record the participants of collaborative activities. Discovery of the covert node refers to the retrieval of the data and the corresponding collaborative activities that result from the influence of the covert node. The nodes that appear commonly in the retrieved data are likely to neighbor the covert node. Two methods are presented for detecting covert nodes within a social network. A novel statistical inference method is discussed and compared with a conventional heuristic method (data crystallization). The statistical inference method employs the maximal likelihood estimation and outlier detection techniques. The performance of the methods is demonstrated with test datasets that are generated from computationally synthesized networks and from a real organization.
|A Note on Creating Networks from Social Network Data, 77-84|
We are interested in the variance of social networks when using supporting evidence to define “friend” relationships, particularly in online social networks. Of related interest, relevant to this study, is the impact of various network sampling methods as well as the ability to capture the true structure of a network given incomplete or inaccurate data. Using empirical social network data, we explore the effect of requiring friendship relationships to be supported with communications between members, where friendship between members must also be corroborated with friendships to a third member. Ultimately, we hope to identify substantial relationships between members that may be capable of influencing behavior. A hypothetical scenario of measuring the vitality of an online community allows us to assess the effect on pertinent network metrics. Results indicate some amount of stability in certain measurements, but enough variance is present to suggest that in empirical networks, the presence of key nodes and edges reduces the robustness previously measured in random networks. Most significantly, the study demonstrates a possible way to identify the robustness of relationships within networks, as well as identify high-level groupings of communities as stable under different relationships, by increasing the amount of 'evidence' required to create ties between members, irrespective of the strength of the ties.
|Back Cover, 85|