Complete Network Data Research Articles

Homophily in Adolescence

This study asks to what extent similarity in status characteristics (gender and ethnicity) is linked with similarity in anti-school behaviour. We argue that the social forces responsible for homophily-based selection are also at work with regard to homophily-based influence. We use data from the Flemish Educational Assessment Study, which collected complete network data from a representative sample of Flemish secondary school adolescents (N = 11,872). Results indicate that similarity on gender and ethnicity is linked with similarity on behavioural characteristics. Furthermore, the association between status homophily and behavioural homophily is stronger for boys than girls. For minorities, status heterophily on ethnicity is associated with behavioural heterophily, which may be an indication that minority students distance themselves from their majority-group friends.

Leveraging social influence to address overweight and obesity using agent-based models: The role of adolescent social networks

The prevalence of adolescent overweight and obesity (hereafter, simply “overweight”) in the US has increased over the past several decades. Individually-targeted prevention and treatment strategies targeting individuals have been disappointing, leading some to propose leveraging social networks to improve interventions. We hypothesized that social network dynamics (social marginalization; homophily on body mass index, BMI) and the strength of peer influence would increase or decrease the proportion of network member (agents) becoming overweight over a simulated year, and that peer influence would operate differently in social networks with greater overweight. We built an agent-based model (ABM) using results from R-SIENA. ABMs allow for the exploration of potential interventions using simulated agents. Initial model specifications were drawn from Wave 1 of the National Longitudinal Study of Adolescent Health (Add Health). We focused on a single saturation school with complete network and BMI data over two waves (n = 624). The model was validated against empirical observations at Wave 2. We focused on overall overweight prevalence after a simulated year. Five experiments were conducted: (1) changing attractiveness of high-BMI agents; (2) changing homophily on BMI; (3) changing the strength of peer influence; (4) shifting the overall BMI distribution; and (5) targeting dietary interventions to highly connected individuals. Increasing peer influence showed a dramatic decrease in the prevalence of overweight; making peer influence negative (i.e., doing the opposite of friends) increased overweight. However, the effect of peer influence varied based on the underlying distribution of BMI; when BMI was increased overall, stronger peer influence increased proportion of overweight. Other interventions, including targeted dieting, had little impact. Peer influence may be a viable target in overweight interventions, but the distribution of body size in the population needs to be taken into account. In low-obesity populations, strengthening peer influence may be a useful strategy.

Balancing act: Government roles in an energy conservation network

Government-led interorganizational alliance networks present a sensible opportunity to overcome many societal challenges through collaborative governance. In particular, few researchers have studied alliance networks in the field of energy conservation in commercial buildings—a sector with unique barriers to greater diffusion of innovative cost-saving strategies. We applied an analytic inductive case-based method and social network analysis to study one particular alliance network: the United States Commercial Building Energy Alliances representing interests from retail, commercial real estate, and healthcare sectors. This alliance network was initiated by the United States Department of Energy, with assistance from several federally funded research laboratories in the United States, to promote the diffusion of knowledge and ultimately encourage greater deployment of energy efficiency and clean energy strategies in commercial buildings. We draw upon interview data from 28 cases of private, non-profit, and governmental organizations and complete network data from the alliance participants. We honed in on eight focal cases of governmental organizations to provide insight on how the four forms of energy and environmental data, information, and knowledge shared within an alliance network address the challenge of a vastly underutilized energy resource, namely conservation. Further, we identify and discuss the public's four roles—Commissioner, Interpreter, Marketer, and User—in providing balance to the diffusion of both private and public goods in a network.

Composition and Structure of a Large Online Social Network in the Netherlands

Limitations in data collection have long been an obstacle in research on friendship networks. Most earlier studies use either a sample of ego-networks, or complete network data on a relatively small group (e.g., a single organization). The rise of online social networking services such as Friendster and Facebook, however, provides researchers with opportunities to study friendship networks on a much larger scale. This study uses complete network data from Hyves, a popular online social networking service in the Netherlands, comprising over eight million members and over 400 million online friendship relations. In the first study of its kind for the Netherlands, I examine the structure of this network in terms of the degree distribution, characteristic path length, clustering, and degree assortativity. Results indicate that this network shares features of other large complex networks, but also deviates in other respects. In addition, a comparison with other online social networks shows that these networks show remarkable similarities.

Composition and Structure of a Large Online Social Network in the Netherlands

Limitations in data collection have long been an obstacle in research on friendship networks. Most earlier studies use either a sample of ego-networks, or complete network data on a relatively small group (e.g., a single organization). The rise of online social networking services such as Friendster and Facebook, however, provides researchers with opportunities to study friendship networks on a much larger scale. This study uses complete network data from Hyves, a popular online social networking service in the Netherlands, comprising over eight million members and over 400 million online friendship relations. In the first study of its kind for the Netherlands, I examine the structure of this network in terms of the degree distribution, characteristic path length, clustering, and degree assortativity. Results indicate that this network shares features of other large complex networks, but also deviates in other respects. In addition, a comparison with other online social networks shows that these networks show remarkable similarities.

Bayesian Meta-Analysis of Social Network Data via Conditional Uniform Graph Quantiles

Many basic questions in the social network literature center on the distribution of aggregate structural properties within and across populations of networks. Such questions are of increasing relevance given the growing availability of network data suitable for meta-analytic studies, as well as the rise of study designs that involve the collection of data on multiple networks drawn from a larger population. Despite this, little work has been done on model-based inference for the properties of graph populations, or on methods for comparing such populations. Here, we attempt to rectify this gap by introducing a family of techniques that combines an existing approach to the identification of structural biases in network data (the use of conditional uniform graph quantiles) with strategies drawn from nonparametric Bayesian analysis. Conditional uniform graph quantiles are the quantiles of an observed structural property in the reference distribution produced by evaluating that property over all graphs with certain fixed characteristics (e.g., size or density). These quantiles have long been used to measure the extent to which a property of interest on a single network deviates from what would be expected given that network's other characteristics. The methods introduced here employ such quantile information to allow for principled inference regarding the distribution of structural biases within (and comparison across) populations of networks, given data sampled at the network level. The data requirements of these methods are minimal, thus making them well-suited to meta-analytic applications for which complete network data (as opposed to summary statistics) are often unavailable. The structural biases inferred using these methods can be expressed in terms of posterior predictives for familiar and easily communicated quantities, such as p-values. In addition to the methods themselves, we present algorithms for posterior simulation from this model class, illustrating their use with applications to the analysis of social structure within urban communes and radio communications among emergency personnel. We also discuss how this approach may applied to quantiles arising from other reference distributions, such as those obtained using general exponential-family random graph models.

Mechanisms of Peer Influence among Adolescents: Cohesion versus Structural Equivalence

Peer influence is regarded as one of the strongest determinants of adolescents' behavior. Yet the concept of peer influence has been studied almost exclusively in terms of dyadic friendship relations. Based on actors' similar positions in the friendship network, the authors test an alternative mechanism of influence in this article, drawing on insights from network-inspired theories. This study focuses on both sexual activity and school deviant behavior. The authors used data from the Flemish Educational Assessment Study, which collected complete network data from a representative sample of Flemish secondary school adolescents ( N = 11,872), clustered in 160 networks. Results indicate that best friends influence both adolescents' sexual activity and their school deviancy. However, the authors found no evidence of influence between actors who share a similar position in the network.

Social Connectedness in Health, Morbidity and Mortality, and Health Care - The Contributions, Limits and Further Potential of Health and Retirement Study

Abstract This part of the mid-term review of the Health and Retirement Study (HRS) provides an overall assessment of the utility of HRS data for research targeting the nature and influence of social connectedness. As one of the major dimensions of the social aspects of psychosocial influences, social connectedness is among the most complicated in terms of definition, conceptualization, and measurement. However, the century-long body of theory and findings couple with a recent resurgence of research on the critical impact of these ties for health, illness, and health care to call for an examination of the richness in and limitations of current HRS data.This assessment is comprised of three broad steps: 1) an overview of the nature of social connectedness, and of the dimensions and methodological approaches that can and have been used in studying health, health care, and aging; 2) the range, strengths and limitations of the HRS data on each approach; and 3) suggestions for potential directions to increase the utility of data collected and further research contributions from the HRS. While no tabular listing of items relevant to social connectedness is presented, the sets of items that tap this notion are referenced throughout.Overall, the HRS represents one of, if not the most impressive data sets regarding the ability to examine the influence of social connectedness on health, illness and health care. Given different theoretical and methodological traditions of social connectedness (e.g., the local or ego-centered perspective; social support perspective; social capital perspective; Pescosolido 2006a), the HRS either currently offers a way to tap into various views of social connectedness or holds the potential to do so. Specifically, the HRS includes four kinds of social connectedness data: socio-demographic proxies that represent a tie (e.g., marital status) with detailed data on the nature of the bond; social support batteries which offer respondent perceptions of the overall positive and negative aspects of sets of relationships; eco-centric tie data, which provide a list of names or roles that can provide support (i.e. latent ties); and networks of event response in which respondents list individuals who were called upon (e.g., activated ties) under certain conditions. Given the individually-based and national scope of the HRS, the collection of full or complete network data is not feasible at present.Four strategies could improve the collection and use of social connectedness data in the HRS. First, data collection sections that are explicit or implicit ego-centric name generators or activated ties lists could be expanded and refined to provide more complete data. Under the “looping” structure of the HRS, both the ego-centric and event response batteries can serve as a foundation for expanded network batteries. Second, given the increasing role of social media in contemporary American lives, the HRS section on the use of technology should be reviewed and expanded to tap into virtual ties. Third, locator data designed to improve follow-up of the HRS samples can form the basis of a network roster and for analyses of the dynamics of ties and its influence on health and health care. Fourth, while it is not possible to “go back” and recapture data about social connectedness, a sub study which targets the named “social convoy” over a person’s life (defined only as time in the HRS) would provide invaluable data that could not be collected from any other existing study. That is, while subject to a variety of criticism (e.g., telescoping effects), the ability to collect data on extent of turnover and the reasons for shifts in social connectedness would allow an analysis of the impact of social network dynamics in later life, potentially reveal key turning points in social network support, and offer targeted points of interventions for fostering the social connectedness that has, to date, been shown (in the HRS and other studies) to be so essential to health and well-being.

A spatio-temporal data model for road network in data center based on incremental updating in vehicle navigation system

The technique of incremental updating, which can better guarantee the real-time situation of navigational map, is the developing orientation of navigational road network updating. The data center of vehicle navigation system is in charge of storing incremental data, and the spatio-temporal data model for storing incremental data does affect the efficiency of the response of the data center to the requirements of incremental data from the vehicle terminal. According to the analysis on the shortcomings of several typical spatio-temporal data models used in the data center and based on the base map with overlay model, the reverse map with overlay model (RMOM) was put forward for the data center to make rapid response to incremental data request. RMOM supports the data center to store not only the current complete road network data, but also the overlays of incremental data from the time when each road network changed to the current moment. Moreover, the storage mechanism and index structure of the incremental data were designed, and the implementation algorithm of RMOM was developed. Taking navigational road network in Guangzhou City as an example, the simulation test was conducted to validate the efficiency of RMOM. Results show that the navigation database in the data center can response to the requirements of incremental data by only one query with RMOM, and costs less time. Compared with the base map with overlay model, the data center does not need to temporarily overlay incremental data with RMOM, so time-consuming of response is significantly reduced. RMOM greatly improves the efficiency of response and provides strong support for the real-time situation of navigational road network.

“Kracking” the Missing Data Problem: Applying Krackhardt's Cognitive Social Structures to School-Based Social Networks

Social network analysis can enrich school-based research on children's peer relationships. Unfortunately, accurate network analysis requires near-complete data on all students and is underutilized in school-based research because of low rates of parental consent. This article advocates Krackhardt's cognitive social structures (CSS) as a solution to the problem of missing data in social network research. It compares CSS to other common strategies for dealing with missing data and shows how CSS can be used to yield more complete and precise network data. The article also uses an example of a sixth-grade classroom network to demonstrate the application of CSS and potential analyses.

CLog: Low Cost Gigabit Full Packet Logging

Creating high quality network trace files is a difficult task to accomplish on a limited budget. High network speeds may overburden an individual system running packet logging software such as tcpdump, resulting in trace files with missing information and making analysis difficult or incomplete. High end specialized systems may perform the job well, but may be out of reach due to financial constraints. To that end, we developed the Cheap Logger (CLog) system which utilizes inexpensive COTS hardware to create high quality, complete network trace files. A scalable distributed storage system enables the CLog system to expand and continue to create high quality, complete network data trace files even at extremely high data rates.

RELATIONS IN DESIGN THINKING: A CASE STUDY OF A SOCIAL NETWORK.

This paper describes an inductive study of the structure of relations supporting design thinking work. Complete social network data of a design thinking team reveals three categories of relations b...

The Multilevel p2 Model

The p 2 model is a random effects model with covariates for the analysis of binary directed social network data coming from a single observation of a social network. Here, a multilevel variant of the p 2 model is proposed for the case of multiple observations of social networks, for example, in a sample of schools. The multilevel p 2 model defines an identical p 2 model for each independent observation of the social network, where parameters are allowed to vary across the multiple networks. The multilevel p 2 model is estimated with a Bayesian Markov Chain Monte Carlo (MCMC) algorithm that was implemented in free software for the statistical analysis of complete social network data, called StOCNET. The new model is illustrated with a study on the received practical support by Dutch high school pupils of different ethnic backgrounds.

Developmental networks and professional identity: a longitudinal study

Purpose – This paper seeks to examine the relationship between individuals' developmental mentoring networks and a subjective career outcome, clarity of professional identity. How developmental network characteristics are related to professional identity over time is explored. Design/methodology/approach – This is a three‐wave, longitudinal survey study, covering a five‐year span (1996‐2001). The participants (n=136), full‐time MBA students at the inception of the study, provided complete developmental network data on each survey. The relationships between clarity of professional identity and three different measures of developmental network density were explored: early‐career density; general density; and density dynamics (e.g. the change in density over time). Findings – Developmental network density, which reflects the professional identity exploration process, is negatively related to clarity of professional identity. Research limitations/implications – The study is limited by the use of graduating MBA students from a single, top‐20 business school as participants. Practical implications – The findings suggest that people might be able to improve their careers through changing their developmental networks, particularly during their early‐career years. Originality/value – This paper provides novel insights to the mentoring, identity, and careers literatures. Given the previously uncharted territory of understanding the dynamics of developmental networks and its relationship to career outcomes, this study opens avenues for future research, while also answering questions about developmental networks and the ways they function over time.

6. Exponential Family Models for Sampled and Census Network Data

Much progress has been made on the development of statistical methods for network analysis in the past ten years, building on the general class of exponential family random graph (ERG) network models first introduced by Holland and Leinhardt (1981). Recent examples include models for Markov graphs, “p*” models, and actor-oriented models. For empirical application, these ERG models take a logistic form, and require the equivalent of a network census: data on all dyads within the network. In a largely separate stream of research, conditional log-linear (CLL) models have been adapted for analyzing locally sampled (“egocentric”) network data. While the general relation between log-linear and logistic models is well known and has been exploited in the case of a priori blockmodels for networks, the relation for the CLL models is different due to the treatment of absent ties. For fully saturated tie independence models, CLL and ERG are equivalent and related via Bayes' rule. For other tie independence models, the two do not yield equivalent predicted values, but we show that in practice the differences are unlikely to be large. The alternate conditioning in the two models sheds light on the relationship between local and complete network data, and the role that models can play in bridging the gap between them.

Schools as environments for social learning—shaping mechanisms? Comparisons of smaller and larger rural schools in Norway

This article analyses and compares the learning environment in smaller and bigger rural schools by focusing on the arenas of both formal and informal learning; the lessons and the recesses between. Relational patterns are both analysed using complete network data from 19 schools in four different municipalities in four Norwegian counties and by comparisons based on data from 80 pupil interviews. The analyses are replicated and followed up by comparison of schools in two additional municipalities. The smaller and bigger rural schools offer radically different learning environments. The aspects of individual and collective agency in learning, self‐centrism, intimization, social exclusion, independence and control as part of everyday life in schools are analysed and discussed. Mechanisms shaping the differences in the learning environments are identified. Needs for further research are formulated.

Asking social network questions

Research findings indicate that different types of social relations have an important influence on the performance of employees in organisations. This paper focuses on a comparison of different methods for acquiring information on advice, cooperation, friendship, adversarial and superficial networks in knowledge organisations. We investigate the applicability of three distinct measurement methods to acquire different kinds of complete network data by means of the recognition method. Data were collected in a small governmental organisation consisting of knowledge workers. First, employees were presented a short description of a specific situation in which social relations with their colleagues might play a significant role. They had to indicate if (or how often) this specific situation occurred with each of the colleagues. Second, respondents were asked to indicate whether a specific relational concept (in this case 'advice' or 'friendship') applied to each of their relations with their colleagues. Third, we provided respondents with four semantic differentials (e.g. distrust-trust) on which they needed to position their relation with the other employees. Whether these different measurement instruments capture distinct aspects of the relation between employees, or whether they measure the same underlying concepts, is one of the major concerns of this paper. The aim of this paper is twofold. First of all, we want to know to what extent these different measurement instruments overlap. Second, we would like to find out to what degree these different methods as a whole give us conceptually different and complementary information. To the extent that items are correlated within one method and between methods we need to investigate which of these different instruments is best suited for our content related purposes. The criteria used for selecting the most appropriate method are minimal item non-response – i.e. from the viewpoint of measuring complete networks – and maximum relational diversity with a minimum of questions.

Simulation study of Internet based inter control center data exchange for complete network modeling

The model used for network security and transmission capability analyses requires the use of a large area network and has to be brought up to real-time conditions. To build a complete network model, supervisory control and data acquisition (SCADA) data have to be made available to many external bodies. In this paper, an Internet interface to SCADA systems is proposed for inter control center data exchange. A discrete event technique is utilized to simulate the interarea data exchange. The performance modeling of data exchange on the Internet is established, and a quantitative analysis of the data exchange delay is presented. Simulation and experimental results have indicated that the Internet provides a convenient way for intercontrol center data exchange and, by using state estimation (SE), accurate complete network models can be obtained for practical security and available transfer capability analyses.

Simulation Study of Internet-Based Inter-Control-Center Data Exchange for Complete Network Modeling

The model used for real-time network security and transmission capability analyses requires the use of a large area network and has to be brought up to real-time conditions. To build a complete network model, SCADA data must be made available to many external bodies. An Internet interface to SCADA systems is proposed for intercontrol-center data exchange. A discrete even technique is utilized to simulate the interarea data exchange. The performance modeling of data exchange on the Internet is established, and a quantitative analysis of the data exchange delay is presented. Simulation and experimental results have indicated that the Intemet provides a convenient way for inter-control-center data exchange, and, by using state estimation (SE), accurate complete network models can be obtained for practical security and made available for transfer capability analyses.

Egocentric and sociocentric measures of network centrality

Egocentric centrality measures (for data on a node’s first-order zone) parallel to Freeman’s [Social Networks 1 (1979) 215] centrality measures for complete (sociocentric) network data are considered. Degree-based centrality is in principle identical for egocentric and sociocentric network data. A closeness measure is uninformative for egocentric data, since all geodesic distances from ego to other nodes in the first-order zone are 1 by definition. The extent to which egocentric and sociocentric versions of Freeman’s betweenness centrality measure correspond is explored empirically. Across seventeen diverse networks, that correspondence is found to be relatively close—though variations in egocentric network composition do lead to some notable differences in egocentric and sociocentric betweennness. The findings suggest that research design has a relatively modest impact on assessing the relative betweenness of nodes, and that a betweenness measure based on egocentric network data could be a reliable substitute for Freeman’s betweenness measure when it is not practical to collect complete network data. However, differences in the research methods used in sociocentric and egocentric studies could lead to additional differences in the respective betweenness centrality measures.