Search

The structure of many complex social networks is determined by nodal and dyadic covariates that are endogenous to the tie variables. While exponential-family random graph models (ERGMs) have been very successful in modeling social networks with exogenous covariates, they are often misspecified for networks where some covariates are stochastic. Exponential-family random network models (ERNMs) are an extension of ERGM that retain the desirable properties of ERGM, but allow the joint modeling of tie variables and covariates. We compare ERGM to ERNM to show how conclusions of ERGM modeling are improved by consideration of the ERNM framework. In particular, ERNM simultaneously represents the effects of social influence and social selection processes, while commonly used models do not. • Current social network models ignore the fact that many covariates are endogenous. • ERNMs are flexible and allow the joint modeling of tie variables and covariates. • We show that ERNMs can correctly model the joint effects, while current models fail. [ABSTRACT FROM AUTHOR]

Social network data often contain missing values because of the sensitive nature of the information collected and the dependency among the network actors. As a response, network imputation methods including simple ones constructed from network structural characteristics and more complicated model-based ones have been developed. Although past studies have explored the influence of missing data on social networks and the effectiveness of imputation procedures in many missing data conditions, the current study aims to evaluate a more extensive set of eight network imputation techniques (i.e., null-tie, Reconstruction, Preferential Attachment, Constrained Random Dot Product Graph, Multiple Imputation by Bayesian Exponential Random Graph Models or BERGMs, k-Nearest Neighbors, Random Forest, and Multiple Imputation by Chained Equations) under more practical conditions through comprehensive simulation. A factorial design for missing data conditions is adopted with factors including missing data types, missing data mechanisms, and missing data proportions, which are applied to generated social networks with varying numbers of actors based on 4 different sets of coefficients in ERGMs. Results show that the effectiveness of imputation methods differs by missing data types, missing data mechanisms, the evaluation criteria used, and the complexity of the social networks. More complex methods such as the BERGMs have consistently good performances in recovering missing edges that should have been present. While simpler methods like Reconstruction work better in recovering network statistics when the missing proportion of present edges is low, the BERGMs work better when more present edges are missing. The BERGMs also work well in recovering ERGM coefficients when the networks are complex and the missing data type is actor non-response. In conclusion, researchers analyzing social networks with incomplete data should identify the network structures of interest and the potential missing data types before selecting appropriate imputation methods. [ABSTRACT FROM AUTHOR]

Researchers often use pooled exponential random graph models (ERGM) to analyze samples of networks. However, pooled ERGM—here, understood to include both meta-regression and combined estimation on a stacked adjacency matrix—may be biased if there is heterogeneity in the latent error variance ('scaling') of each lower-level model. This study explores the implications of scaling for pooled ERGM analysis. We illustrate that scaling can produce bias in pooled ERGM coefficients that is more severe than in single-network ERGM and we introduce two methods for reducing this bias. Simulations suggest that scaling bias can be large enough to alter conclusions about pooled ERGM coefficient size, significance, and direction, but can be substantially reduced by estimating the marginal effect within a block diagonal or random effects meta-regression framework. We illustrate each method in an empirical example using Add Health data on 15 in-school friendship networks. Results from the application illustrate that many substantive conclusions vary depending on choice of pooling method and interpretational quantity. • Scaling creates novel problems in pooled exponential random graph models that do not exist in the single network case. • Pooled ERGM coefficients can be the incorrect sign, size, and significance when scaling is present. • Two bias reduction methods are introduced. • Block diagonal estimation and random effects meta-regression of average marginal effects decrease scaling bias. • An empirical example on pooled analysis of 15 AddHealth in-school friendship networks is provided. [ABSTRACT FROM AUTHOR]

Copyright of Kölner Zeitschrift für Soziologie und Sozialpsychologie ( KZfSS) is the property of Springer Nature and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Distinctiveness theory suggests that numeric rarity is correlated with stronger homophily. In this paper, we examine this theory by studying gender homophily in social networks of older adults. We document subjective social networks in multiple long term care settings for older adults over several time points. Homophily for each gender is estimated using exponential random graph models. We find evidence for positive homophily across all networks, and show that it is correlated to the magnitude of the female majority or male minority. Our findings empirically verify distinctiveness theory and could improve interventions to promote tie formation in social networks. • Studied social networks of older adults' long term care settings at various times. • Gender homophily estimated in each network using ERGM to control for other factors. • Homophily correlated with gender distribution, supporting Distinctiveness theory. • No differences found between two types of older adult long term care settings. • Discarded additional alternative explanation of gender homophily. [ABSTRACT FROM AUTHOR]

Burnout is commonly associated with professions that entail a high rate of close relationships with other individuals or groups. This paper explores the association between burnout and interpersonal relationships using a relational, social network framework. We collected data on advice-seeking relationships among 102 teachers and administrative staff from a secondary school in Melbourne, Australia. Burnout was measured using the Burnout Assessment Tool and we focused on four core subscales: (1) exhaustion; (2) mental distance; (3) emotional impairment; and (4) cognitive impairment. We applied a particular class of statistical model for social networks called Exponential Random Graph Models (ERGMs) to shed new light on how level of burnout relates to formation of advice relations among school staff. Results indicated that high levels of overall burnout were linked to a higher number of advice-seeking ties among school staff. Additionally, teachers who scored high in cognitive impairment (i.e., difficulties in thinking clearly and learn new things at work) tended to seek and to provide advice to a greater number of others. Finally, school staff who scored high in exhaustion (i.e., a severe loss of energy that results in feelings of both physical and mental exhaustion) tended to be sought out less as advisors to others, while those high in mental distance (i.e., psychologically distancing oneself from others) were generally less likely to seek advice from other school staff. We discuss these findings drawing on Conservation of Resource theory. Notably, our results show that burnout is not only an individual-level problem, but that burnout is associated with reduced social connectivity in specific ways that may impact on how other school staff collaborate, culminating in a staff-wide overall impact that affects how schools function.

Exponential‐family random graph models (ERGMs) are widely used in social network analysis when modelling data on the relations between actors. ERGMs are typically interpreted as a snapshot of a network at a given point in time or in a final state. The recently proposed Latent Order Logistic model (LOLOG) directly allows for a latent network formation process. We assess the real‐world performance of these models when applied to typical networks modelled by researchers. Specifically, we model data from an ensemble of articles in the journal Social Networks with published ERGM fits, and compare the ERGM fit to a comparable LOLOG fit. We demonstrate that the LOLOG models are, in general, in qualitative agreement with the ERGM models, and provide at least as good a model fit. In addition, they are typically faster and easier to fit to data, without the tendency for degeneracy that plagues ERGMs. Our results support the general use of LOLOG models in circumstances where ERGMs are considered. [ABSTRACT FROM AUTHOR]

Strong, stable social bonds in primates are characterized by high levels of social affiliation, low levels of aggression, minimal stress, and affiliative reciprocity within the dyad. In relatively well‐studied catarrhine monkeys, these bonds tend to form most frequently between kin, animals close in age, and animals close in rank. This results in patterns of affiliation in which kin, similarly aged animals, and like‐ranked animals tend to affiliate and patterns of aggression and submission where animals tend to aggress more toward nonkin and closely ranked animals, and submit more toward distantly ranked animals. However, literature on how affiliative and agonistic relationships are organized in platyrrhine primate species like brown capuchin monkeys is limited and conflicting. In this study, we used social network analyses to characterize how age, sex, maternal kinship, and dominance rank relate to the patterns of submissive, aggressive, contact, and grooming interactions in a group of captive brown capuchin monkeys. Like catarrhine monkeys, brown capuchin monkeys showed a steep linear dominance hierarchy, tended to affiliate with kin, similarly aged animals, and like‐ranked animals, and tended to aggress more toward nonkin. However, our monkeys showed a pattern of affiliation and grooming down the hierarchy that is inconsistent with grooming up the hierarchy patterns often seen in catarrhine monkey groups, suggesting that brown capuchins do not compete for access to higher ranking social partners. Higher ranking monkeys were most central to the aggression network, and lower ranking monkeys were most central to the submission network. Mid‐ranking monkeys were the most central to the contact network, suggesting that they may play an important role in the affiliative cohesion of the group. These results inform our understanding of brown capuchin social behavior specifically, and of how demographic factors relate to social organization in platyrrhine primates generally. Highlights: Captive brown capuchins show a steep linear dominance hierarchy.Rank differentially predicts who is the most central across networks: High‐rankers for aggression, low‐rankers for submission, & mid‐rankers for contact.They affiliate with kin, similarly aged and like‐ranked animals, & aggress to nonkin. [ABSTRACT FROM AUTHOR]

A recently published paper [Martin (2017) JoSS 18(1):1-21] investigates the structure of an unusual set of social networks, those of the alternate personalities described by a patient undergoing therapy for multiple personality disorder (now known as dissociative identity disorder). The structure of these networks is modeled using the -series, a sequence of nested network distributions of increasing complexity. Martin finds that the first of these networks contains a striking feature of a large "hollow ring"; a cycle with no shortcuts, so that the shortest path between any two nodes in the cycle is along the cycle (in more precise graph theory terms, this is a geodesic cycle). However, the subsequent networks have much smaller largest cycles, smaller than those expected by the models. In this work, I re-analyze these delusional social networks using exponential random graph models (ERGMs) and investigate the distribution of the lengths of geodesic cycles. I also conduct similar investigations for some other social networks, both fictional and empirical, and show that the geodesic cycle length distribution is a macro-level structure that can arise naturally from the micro-level processes modeled by the ERGM. [ABSTRACT FROM AUTHOR]

A recently published paper [Martin (2017) JoSS 18(1):1-21] investigates the structure of an unusual set of social networks, those of the alternate personalities described by a patient undergoing therapy for multiple personality disorder (now known as dissociative identity disorder). The structure of these networks is modeled using the dk-series, a sequence of nested network distributions of increasing complexity. Martin finds that the first of these networks contains a striking feature of a large “hollow ring”; a cycle with no shortcuts, so that the shortest path between any two nodes in the cycle is along the cycle (in more precise graph theory terms, this is a geodesic cycle). However, the subsequent networks have much smaller largest cycles, smaller than those expected by the models. In this work, I re-analyze these delusional social networks using exponential random graph models (ERGMs) and investigate the distribution of the lengths of geodesic cycles. I also conduct similar investigations for some other social networks, both fictional and empirical, and show that the geodesic cycle length distribution is a macro-level structure that can arise naturally from the micro-level processes modeled by the ERGM.

Martin (2020) describes a misinterpretation of exponential random graph (ERGM) parameters in my contribution (Stivala 2020), with the use of this parametric model obscuring, rather than illuminating, the data. He suggests that this is symptomatic of a trend in the social networks community towards a methodological monoculture focussed on the use of ERGMs. In this Reply I try to clarify how this situation arose in this specific case, and address some more general issues Martin raises, including the use of nodal covariates, what we can learn from ERGMs, and methodological monoculturalism in social network research.

Current evidence suggests that nonprofits' use of Twitter is not strongly related to organizational size, unlike other technological developments. However, this evidence is primarily based on studies of large nonprofit organizations. This study uses a random sample of charities, stratified by size, to present evidence that organization size is a significant factor in multiple dimensions of social media use: the percentage of charities owning a Twitter handle, activity on the site, and popularity within the charities' network. Many charities are using Twitter, but larger charities are making more effective use of the platform to connect to other organizations. The very largest charities exhibit an overwhelming popularity effect in the network, whereas small charities are notably less active than their larger counterparts. Besides the substantive findings, we further demonstrate the methodological potential of using exponential random graph modeling to gain a deeper understanding of the characteristics of nonprofits' social media networks. [ABSTRACT FROM AUTHOR]

Exponential random graph models (ERGM) have been widely applied in the social sciences in the past 10 years. However, diagnostics for ERGM have lagged behind their use. Collinearity-type problems can emerge without detection when fitting ERGM, yielding inconsistent model estimates and problematizing inference from parameters. This article provides a method to detect multicollinearity in ERGM. It outlines the problem and provides a method to calculate the variance inflation factor (VIF) from ERGM parameters. It then evaluates the method with a Monte Carlo simulation, fitting 216,000 ERGMs and calculating the VIFs for each model. The distribution of VIFs is analyzed using multilevel regression to determine what network characteristics lend themselves to collinearity-type problems. The relationship between VIFs and unstable standard errors (a standard sign of collinearity) is also examined. The method is shown to effectively detect multicollinearity, and guidelines for interpretation are discussed. [ABSTRACT FROM AUTHOR]

How does international financial aid affect the cooperative behavior of local non-governmental organizations (NGOs)? Can NGOs, while turning global, preserve peer connections with local actors and be engaged in local issues? The civil society literature contains competing perspectives on and reports of how international financial aid may restructure local civic networks. Some scholars argue that international support comes at the expense of local integration as inclusion in global networks takes local NGOs out of the local context, while others find evidence that organizations do not have to face "a forced choice", and may well be integrated both globally and locally. Drawing on this scholarship, we examine two hypotheses on how transnational funding influences cooperation patterns among NGOs. The hierarchy argument states that public entities tend to cooperate with internationally funded NGOs as external contact signals quality and trustworthiness. The segregation argument, on the contrary, suggests financial homophily according to which organizations are more likely to choose peers similar in sources of funding. To test these hypotheses, we apply Exponential Random Graph Models to the data on cooperation among 221 Kazakhstani NGOs. Results support the segregation hypothesis implying that NGOs are likely to have a bias towards similarly funded peers. [ABSTRACT FROM AUTHOR]

Environmental problems often span a set of challenges that each may engage different policy actors across different policy domains. These challenges, or policy issues, nonetheless exhibit interdependencies that may constrain the ability of actors to work together towards joint solutions.Still, we have limited knowledge about whether and how policy issue interdependencies actually shape how actors collaborate.Using data derived from two venues for collaborative water governance in the Norrström basin, Sweden, we investigate whether and how policy issues and policy issue interdependencies influence actors' selection of collaborative partners. We test two alternative sets of propositions; one set assumes that partner selection is driven by actors' engagement in policy issues and their interdependencies, while the other set emphasises social positions and actor attributes.Our results show that in one venue, actors' choices of collaborative partner were associated with factors from both sets, but not with policy issue interdependencies specifically. In the other venue, only actor and relational attributes shaped social tie formation. These results suggest that how actors interact does not necessarily align with the policy issues and the policy issue interdependencies defined by the environmental problem they are to address.Our results provide an important step towards arriving at evidence‐based recommendations for more effective collaborative efforts in addressing complex environmental problems that no actor can address alone. A free Plain Language Summary can be found within the Supporting Information of this article. A free Plain Language Summary can be found within the Supporting Information of this article. [ABSTRACT FROM AUTHOR]

• Compares state of the art missing network data treatment methods. • Evaluation on descriptive statistics, link reconstruction, and model parameters. • Tested on simulated networks varying in structure and size. • Multiple imputation outperforms simpler methods. • Multiple imputation with Bayesian ERGMs overall performed best. This paper compares several missing data treatment methods for missing network data on a diverse set of simulated networks under several missing data mechanisms. We focus the comparison on three different outcomes: descriptive statistics, link reconstruction, and model parameters. The results indicate that the often used methods (analysis of available cases and null-tie imputation) lead to considerable bias on descriptive statistics with moderate or large proportions of missing data. Multiple imputation using sophisticated imputation models based on exponential random graph models (ERGMs) lead to acceptable biases in descriptive statistics and model parameters even under large amounts of missing data. For link reconstruction multiple imputation by simple ERGM performed well on small data sets, while missing data was more accurately imputed in larger data sets with multiple imputation by complex Bayesian ERGMs (BERGMs). [ABSTRACT FROM AUTHOR]

• Students' self-identified and perceived ethnicity may differ. • Perceived, rather than self-identified, ethnicity plays a role in victimization. • Both Roma and non-Roma students tend to bully peers they perceive as Roma. In this study, we investigate the associations between self-reported and victim-reported bullying and two dimensions of ethnicity (self-identification and ethnic perceptions) among non-Roma majority and Roma minority Hungarian secondary school students. Results of the meta-analysis of exponential random graph models for 12 classes (347 students, 4 schools) show that both self-declared Roma and non-Roma students are more likely to report that they bully peers they perceive as Roma compared to peers they perceive as non-Roma. This is after controlling for gender, socio-economic status, and structural characteristics of the bullying networks. Similar associations have not been found, however, analysing victims' reports. [ABSTRACT FROM AUTHOR]

The frequency and severity of natural hazards are predicted to increase with climate change. Collaboration among actors across scales and organizational boundaries is essential to address this escalation. Pre-existing social networks are generally considered a catalyst enabling actors to more quickly address collective action problems. However, empirically derived knowledge about if, how, and why pre-established social networks facilitate effective collaborations in addressing natural hazards is scarce. We use survey data from crisis responders of large-scale wildfires in Sweden and Canada to investigate factors that shape actors' (i) ability and willingness to form new social ties with other actors and (ii) propensity to "activate" pre-existing social ties. Our results show that many new social ties were established in both events, but also that pre-existing ties comprised a considerable proportion (54–82%) of all ties in use. Using exponential random graph models for temporal networks, we demonstrate that two actors that are working with or have previously worked with a common third actor are more likely to activate pre-existing social ties. Further, new social ties tend to be formed around a few central actors, whereas the opposite seems to apply for the activation of pre-existing ties. The extent to which actors consider others' organizational affiliation, formal role, previous experience, and level of professionalization differs between the cases. We suggest these tie formation and activation differences can be attributed to diverging organizational contexts varying in their reliance upon self-organizing versus command-and-control approaches. [ABSTRACT FROM AUTHOR]

Since few states are able to produce all of their own military hardware, a majority of countries' military systems rely on weapon imports. The structure of the international defense technology exchange network remains an important puzzle to understand, along with the factors that drive its evolution. Drawing on a political economy model of arms supply, we propose a new network-oriented explanation for the worldwide transactions of major conventional weapons in the period after World War II. Using temporal exponential random graph models, our dynamic approach illustrates how network dependencies and the relative weighting of economic versus security considerations vary over time. One of our major results is to demonstrate how security considerations started regaining importance over economic ones after 2001. Additionally, our model exhibits strong out-of-sample predictive performance, with network dependencies contributing to model improvement especially after the Cold War. [ABSTRACT FROM AUTHOR]

Two approaches for the statistical analysis of social network generation are widely used; the tie-oriented exponential random graph model (ERGM) and the stochastic actor-oriented model (SAOM) or Siena model. While the choice for either model by empirical researchers often seems arbitrary, there are important differences between these models that current literature tends to miss. First, the ERGM is defined on the graph level, while the SAOM is defined on the transition level. This allows the SAOM to model asymmetric or one-sided tie transition dependence. Second, network statistics in the ERGM are defined globally but are nested in actors in the SAOM. Consequently, dependence assumptions in the SAOM are generally stronger than in the ERGM. Resulting from both, meso- and macro-level properties of networks that can be represented by either model differ substantively and analyzing the same network employing ERGMs and SAOMs can lead to distinct results. Guidelines for theoretically founded model choice are suggested. [ABSTRACT FROM AUTHOR]

Causal inference is a fast-growing multidisciplinary field that has drawn extensive interests from statistical sciences and health and social sciences. In this article, we gather comprehensive information on publications and citations in causal inference and provide a review of the field from the perspective of citation network analysis. We provide descriptive analyses by showing the most cited publications, the most prolific and the most cited authors, and structural properties of the citation network. Then, we examine the citation network through exponential random graph models (ERGMs). We show that both technical aspects of the publications (e.g., publication length, time and quality) and social processes such as homophily (the tendency to cite publications in the same field or with shared authors), cumulative advantage, and transitivity (the tendency to cite references’ references), matter for citations. We also provide specific analysis of citations among the top authors in the field and present a ranking and clustering of the authors. Overall, our article reveals new insights into the landscape of the field of causal inference and may serve as a case study for analyzing citation networks in a multidisciplinary field and for fitting ERGMs on big networks. Supplementary materials for this article are available online. [ABSTRACT FROM AUTHOR]

Social networks have an important effect on health, and social network analysis has become essential for understanding human behavior and vulnerability. Using exponential random graph models (ERGM), this study explores the associations between mental health and network structure (or more specifically, mental health homophily) and the association between poor mental health and social isolation. Two classes of Romanian adolescents aged 12-14 years participated in the study (n = 26 in each class). We assessed school network, sociodemographic covariates, and mental health using the Strengths and Difficulties Questionnaire (SDQ). ERGM was first used to test the presence of sex and mental health homophily and then to test whether mental health was a predictor of social isolation. The results showed homophily patterns regarding sex and mental health. Moreover, participants with a higher SDQ score had a lower probability of a tie. Overall, this study showed how social networks are structured with different forms of homophily and that adolescents with poor mental health are more likely to be social isolates. Thus, prevention and interventions should focus on these vulnerable adolescents. Methodological advances like ERGM constitute a promising avenue for further research. [ABSTRACT FROM AUTHOR]

The local theatrical offer is the result of all the theatre companies which perform shows in each other's venues. Theatre hospitality is an inherently relational phenomenon, and besides big national and international tours, it is an important part of the local cultural landscape. Aiming at contributing to the literature on network analysis applied to the inquiry on culture, the research adopts the network perspective to test hypotheses on companies' relational behaviors and mechanisms of network formation in a local context in Italy. The analyses show that companies which get more public funding tend to host more; there is a homophily effect based on audience levels; companies tend to reciprocate hospitality relations and form clusters of close collaborations. [ABSTRACT FROM AUTHOR]

Copyright of Annee Sociologique is the property of Presses Universitaires de France and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Network analysis has become an increasingly prevalent research tool across a vast range of scientific fields. Here, we focus on the particular issue of comparing network statistics, i.e. graph-level measures of network structural features, across multiple networks that differ in size. Although “normalized” versions of some network statistics exist, we demonstrate via simulation why direct comparison is often inappropriate. We consider normalizing network statistics relative to a simple fully parameterized reference distribution and demonstrate via simulation how this is an improvement over direct comparison, but still sometimes problematic. We propose a new adjustment method based on a reference distribution constructed as a mixture model of random graphs which reflect the dependence structure exhibited in the observed networks. We show that using simple Bernoulli models as mixture components in this reference distribution can provide adjusted network statistics that are relatively comparable across different network sizes but still describe interesting features of networks, and that this can be accomplished at relatively low computational expense. Finally, we apply this methodology to a collection of ecological networks derived from the Los Angeles Family and Neighborhood Survey activity location data. [ABSTRACT FROM AUTHOR]

Organizations have been experimenting with intraorganizational crowdsourcing ( IOC), yet the mechanisms of IOC production remain an underresearched topic. Drawing on a 2-mode ERGM, we examine structural mechanisms and individual-level factors that shape the network structure of idea generation and selection yielded by an IOC idea challenge in a global IT corporation. Results show a Matthew effect leading to 1) highly centralized employee participation around a few 'superactive' employees who engage with many ideas and 2) highly centralized idea popularity with a few ideas attracting most employee attention. We find support for shared affiliations among employee-idea clusters in the first half of the participation, which is, however, less likely in the second half. We also find support for geographic homophily. [ABSTRACT FROM AUTHOR]

In this study we investigate the interplay between knowledge workers’ formal project team memberships and their informal interactions from a multilevel network perspective. Conceptualizing knowledge workers’ affiliation with project teams as a membership network and their interactions as an advice network, we discuss how shared project team memberships as well as multiple memberships influence patterns of informal exchange in knowledge-intensive organizations. To empirically determine the impact of formal organization on informal exchange we apply exponential random graph models for multilevel networks to relational data collected on 434 R&D employees working on 218 project teams in a high-tech firm in Germany. Our results show that employees sharing project memberships create advice ties to each other but do not exchange advice reciprocally. In addition, we find a negative relationship between having a high number of project memberships and informally seeking or providing advice. [ABSTRACT FROM AUTHOR]

Economic sociology has established the interdependencies between economic and social structures using the notion of embeddedness of the former in the latter. However research usually studies inter-organizational commercial networks and inter-individual informal networks separately. In this article we use a multilevel framework to analyze jointly economic networks between firms and informal networks between their members in order to reframe this embeddedness hypothesis. Based on a network study of a trade fair for television programmes in Eastern Europe we show that while each level has its own specific processes they are partly nested. Beyond this result, we observe that these levels of agency emerge in different contexts and in different temporalities. To conclude, we show that in order to understand performance on a market one needs to look at this dual positioning of individuals and organizations. [ABSTRACT FROM AUTHOR]

As part of the shift within second language acquisition (SLA) research toward complex systems thinking, researchers have called for investigations of social network structure. One strand of social network analysis yet to receive attention in SLA is network statistical models, whereby networks are explained in terms of smaller substructures of social ties. This article offers an overview and instructive demonstration of exponential random graph models as applied to a cohort of English for Academic Purposes students ( N = 75) and their willingness to communicate in the second language (L2 WTC). Results indicate that cross-cultural ties build off of intracultural ties through triadic closure (meeting through a mutual contact). Furthermore, situational subtypes of L2 WTC are associated with various network tie patterns. [ABSTRACT FROM AUTHOR]

This paper investigates several different aspects of inter-ethnic relationships. It focuses on friendships and negative ties between secondary school students from different ethnic backgrounds, introducing and measuring two different aspects of ethnicity: self-declared ethnicity, and ethnicity based on peer perception. These are first applied separately and then together on a sample of secondary school students in Hungary consisting of two ethnic groups: Roma and non-Roma Hungarian ( N = 420). Friendships and negative ties are modelled using cross-sectional exponential random graph models for sixteen classrooms separately, and then individual models are summarized using meta-analysis. Based on the social identity approach, we predict that inter-ethnic friendships are less likely, and negative ties are more likely, than those within ethnic groups; and that majority students reject their minority peers more than the other way around. Moreover, minority students are expected to exclude those whom they perceive as minorities, but who, at the same time, identify with the majority group, since these classmates might seem to them as “traitors” of their “original” ethnic group. Results mostly confirm our hypotheses, emphasizing the role of perceived ethnicity: majority students tend to dislike peers whom they perceive as minorities, regardless of these peers’ self-declared ethnicity; on the other hand, minority students are likely to send friendship nominations towards their perceived minority classmates if these also declare themselves as minorities, but, as predicted, negative nominations if these declare themselves as majorities. This supports our general idea that different ethnicity aspects might influence friendships and negative ties in different ways, and inconsistencies in someone's ethnic categorization might play an important role in social rejection. [ABSTRACT FROM AUTHOR]

A growing number of studies indicate that aspects of psychology and cognition influence network structure, but much remains to be learned about how network information is stored and retrieved from memory. Are networks recalled as dyads, as triads, or more generally as sub-groups? We employ an experimental design coupled with exponential random graph models to address this issue. We find that respondents flexibly encode social information as triads or groups, depending on the network, but not as dyads. This supports prior research showing that networks are stored using “compression heuristics”, but also provides evidence of cognitive flexibility in the process of encoding relational information. [ABSTRACT FROM AUTHOR]

Long-term changes in visibility patterns between urban settlements are considered an important factor for understanding Iron Age II settlement locations in Southern Spain. From some settlements, the surrounding landscape and other settlements could be visually controlled, and some settlements are argued to be intervisible to allow for communication through visual signals. However, the study of how these visibility patterns changed in the subsequent Roman period in this region is largely ignored. In this paper, we argue that visibility might still have structured interactions between communities in Roman times and should not be dismissed out of hand merely because more and other data sources are available as compared to the Iron Age. However, the way in which it affected human behaviour might have been different in Roman times as compared to the Iron Age. We argue that simulating archaeologists' hypotheses about the emergence of inter-settlement visibility is a promising way of understanding such differences. To do this, we use exponential random graph modelling (ERGM), a statistical network simulation modelling technique that allows us to simulate hypotheses about the emergence and long-term change of visibility networks. We combine this approach with an exploratory analysis of the observed visibility networks between identified urban settlements, which will reveal similarities and differences in the changing patterns of visibility networks through time. The results of the ERGMs are then compared with the changes in the observed network structure. We conclude that our knowledge of the changes from the Iron Age II to the Roman settlement pattern suggests only gradual changes in the role of visibility in structuring inter-settlement interactions, possibly followed by a disintegration of the visibility network after the Roman Early Imperial period. [ABSTRACT FROM AUTHOR]

Highlights: [•] Reanalysis of first social network data from 1880 to 1881 (pupils friendship relations). [•] Historic network reveals structural parameters like reciprocity, triadic closure. [•] Class ranking (‘Klassenrangordnung’) (school performance) has influence on popularity of pupils. [ABSTRACT FROM AUTHOR]

Abstract: The imposition of international economic sanctions is a strategic and often multilateral phenomenon of states attempting to coerce each other into altering their behavior by means of economic pain. The interlocking connections of states issuing sanctions and being sanctioned creates a network of interdependent relations and, we argue, the structure of dependencies endogenous to the network is a major determinant of the network's formation and persistence. We consider endogenous structures, both theoretically and empirically, with three foci: the tendency to sanction frequently, the tendency to be sanctioned frequently, and, most of all, reciprocity. The empirical support we find for each of these processes adds a new dimension to our existing knowledge of the sanctioning process, casts doubt upon some previous findings, and opens important avenues for future research. [Copyright &y& Elsevier]

The analysis of network data is an area that is rapidly growing, both within and outside of the discipline of statistics. This review provides a concise summary of methods and models used in the statistical analysis of network data, including the Erdős-Renyi model, the exponential family class of network models, and recently developed latent variable models. Many of the methods and models are illustrated by application to the well-known Zachary karate dataset. Software routines available for implementing methods are emphasized throughout. The aim of this paper is to provide a review with enough detail about many common classes of network models to whet the appetite and to point the way to further reading. © 2012 Wiley Periodicals, Inc. Statistical Analysis and Data Mining, 2012 [ABSTRACT FROM AUTHOR]

Abstract: Costly signaling has been proposed as a possible mechanism to explain food sharing in foraging populations. This sharing-as-signaling hypothesis predicts an association between sharing and status. Using exponential random graph modeling, this prediction is tested on a social network of between-household food-sharing relationships in the fishing and sea-hunting village of Lamalera, Indonesia. Previous analyses (Nolin, 2010. Food-sharing networks in Lamalera, Indonesia: reciprocity, kinship, and distance. Human Nature 21, 243–268) have shown that most sharing in Lamalera is consistent with reciprocal altruism. The question addressed here is whether any additional variation may be explained as sharing-as-signaling by high-status households. The results show that high-status households both give and receive more than other households, a pattern more consistent with reciprocal altruism than costly signaling. However, once the propensity to reciprocate and household productivity are controlled, households of men holding leadership positions show greater odds of unreciprocated giving when compared to households of nonleaders. This pattern of excessive giving by leaders is consistent with the sharing-as-signaling hypothesis. Wealthy households show the opposite pattern, giving less and receiving more than other households. These households may reciprocate in a currency other than food, or their wealth may attract favor-seeking behavior from others. Overall, status covariates explain little variation in the sharing network as a whole, and much of the sharing observed by high-status households is best explained by the same factors that explain sharing by other households. This pattern suggests that multiple mechanisms may operate simultaneously to promote sharing in Lamalera and that signaling may motivate some sharing by some individuals even within sharing regimes primarily maintained by other mechanisms. [Copyright &y& Elsevier]

Copyright of Mathématiques & Sciences Humaines is the property of Editions du CNRS and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Abstract: Gossip is informal talking about colleagues. Taking a social network perspective, we argue that group boundaries and social status in the informal workplace network determine who the objects of positive and negative gossip are. Gossip networks were collected among 36 employees in a public child care organization, and analyzed using exponential random graph modeling (ERGM). As hypothesized, both positive and negative gossip focuses on colleagues from the own gossiper''s work group. Negative gossip is relatively targeted, with the objects being specific individuals, particularly those low in informal status. Positive gossip, in contrast, is spread more evenly throughout the network. [Copyright &y& Elsevier]