Stochastic Actor-based Models Research Articles

Collaboration and social capital in meta-organisations: bonding or bridging?

ABSTRACT How do firms collaborate to achieve their strategic goals within inter-organisational settings? Inter-firm collaboration is a double-edge sword – it helps firms to obtain external resources, but it also increases risks of knowledge spillovers with competitors. Normally, spillover risks are reduced through formal contracting. However, meta-organisations, like standard-setting organisations, lack formal contracts, meaning that firms often rely on social capital to offset transaction costs and opportunism. This paper applies theories of social capital to explain inter-firm collaboration within the meta-organisational environment of 3GPP (Third Generational Partnership Project) – the primary global mobile telecommunications standard-setting organisation. It utilises a stochastic actor-based model and a comprehensive collaboration dataset from the 3G to 4G eras to examine the mechanism of collaboration. We find that firms are more likely to choose partner firms based on the ‘bonding’ view of social capital, rather than the ‘bridging’ view. The paper advances research on collaboration mechanism within meta-organisation, while also contributing to social capital theory in emphasising the importance of the bonding view of social capital under conditions when no formal contracts are available.

The Roles of Supply Networks and Board Interlocks in Firms’ Technological Entry and Exit: Evidence from the Chinese Automotive Industry

ABSTRACTIn this research, we explore how supply networks and board interlocks – as distinct, yet parallel interorganizational networks – jointly influence firms’ entry into new technology domains and exit from old technology domains. Drawing from the perspectives of social networks and organizational learning we highlight the relevance of the interdependency between these networks for a firm's technological entry and exit decisions. We argue that a firm that maintains a large number of supplier ties is more likely to enter new technology domains and exit from old technology domains instead. We further find empirical evidence that the degree centrality of a firm in its board interlock network strengthens these effects. Our theoretical arguments are supported through stochastic actor-based modeling analysis for the longitudinal and multilevel networks of 86 firms active in the Chinese automotive during 2011–2015. These findings inform the literature on interorganizational network dynamics as we insert relational pluralism to examine the complexities of organizational relationships as antecedents to a firms’ technological entry and exit. Finally, we imagine the implications of our analysis for management as they shed light on how multiple interorganizational relationships affect firms’ decisions on new technology entry and old technology exit.

A friend of a friend? Informal authority, social capital, and networks in telecommunications standard-setting organizations

Meta-organizations are assemblages of autonomous formal organizations with a central goal, but which lack typical instruments of formal authority, like contracts and equity. In this organizational context, how do firms establish informal authority to access other firms' resources and thereby achieve their private strategic goals? Using a novel application of resource dependence theory, this paper applies stochastic actor-based model to examine how firms' informal authority is established and evolves to achieve their strategic goals within the Third Generational Partnership Project (3GPP), the primary global mobile telecommunications standard-setting organization. By using a comprehensive dataset of interfirm collaboration over the 3G and 4G generations, our analysis shows that firms tend to establish mutual dependencies by selecting collaborators who possess social capital (high reputations), but avoid collaborators with political capital (high status) within the interorganizational collaboration network. This research provides new insights into how meta-organizations are internally organized through informal authority, while also extending resource dependence theory to the network-level. It also advances our understanding of the formation of mutual dependencies from an actor-based perspective in an interorganizational collaboration network.

Dynamic weight status changes and peer lifestyles in early adolescence: A social network analysis on a longitudinal cohort of Taiwanese youth

BackgroundFriendship networks are relevant to adolescent weight status and lifestyle, but little is known about their relationship in an East Asian setting. Using a longitudinal cohort of Taiwanese adolescents, we studied how weight status and lifestyles might influence friendship. Simultaneously considering the influence of friendship networks, we further examined the effects of peers’ lifestyles on the body weight status of adolescents. MethodsData were retrieved from the first three waves of the Taiwan Youth Project (N = 2409; 51 % boys). Participants were asked to nominate their friends at school and to report their lifestyle factors (e.g. sleep habits, internet use, and athletic performance). Weight status was proxied by body mass index calculated from self-reported weight and height. Stochastic actor-based modeling was used for the friendship network analyses. ResultsConsidering network structure, adolescents selected friends based on similarities in weight status (β = 0.147 [standard error = 0.044], p < 0.001) and lifestyle factors, such as athletic performance (β = 0.181 [0.034], p < 0.001), internet use (β = 0.156 [0.039], p < 0.001), and social jetlag (β = 0.055 [ ± 0.022], p < 0.05). In the behavior dynamics model, we did not observe any peer influence on changes in adolescents’ weight status. ConclusionThese findings suggest a pervasive homophilic friendship selection explained by weight and lifestyle-based similarities, despite no evidence of peer influence on weight status. Stakeholders and care providers of adolescent health should be aware of the social network processes that underlie weight problems.

Social contagion of academic behavior: Comparing social networks of close friends and admired peers.

Peer relations become significant socializing agents for diverse behaviors during adolescence. This study investigated relationship selection and social influence of early adolescents’ close friends and admired peers with regard to academic behavioral engagement. A stochastic actor-based model of social network analysis was used to examine classroom social networks across 2 waves (Mage = 11.46; N = 542) based on peer nominations. Adolescents were asked to nominate their “close friends they hang around with and talk to the most” and peers that they “admire, respect, and want to be like” Results indicated that adolescents who were similar in academic engagement more often became friends. Also, close friends’ and admired peers’ academic engagement contributed to adolescents’ own academic engagement over time. The results suggest that both close friends and admired peers are important channels for social contagion of academic behavior and that examining social relations beyond friends are important for advancing our understanding of peer social influence during adolescence.

Preliminary insights into associations between C-reactive protein and social network dynamics

Recent social psychoneuroimmunology models suggest bidirectional associations between social experiences and the immune system. This work posits that social relationships and networks may influence the functioning of the immune system, but we know little about the role that the immune system plays in how social networks are created and maintained. We examine how salivary C-reactive protein (CRP), as an inflammatory protein, is associated with making new and keeping existing friendship and conflicted relationships among young adult members of a social group. Participants (n = 37; 67.6% female; M age = 18.18 years, 56.81% white/non-Hispanic) provided nominations of friends and individuals with whom they have conflict at wave 1 and two months later at wave 2. At wave 1, in a group setting, participants donated saliva, later assayed for CRP. Stochastic actor-based models revealed that CRP levels were negatively associated with keeping existing friends and positively associated with developing new friendships. We also found that CRP levels were negatively associated with creating new conflicted relationships and predicted an increased likelihood that group members continue conflicted relationships with the focal individual. These preliminary results support the premises of recent social psychoneuroimmunology models by suggesting that inflammation can also serve as a signal to seek new supportive relationships such as friendships and avoid creating new relationships characterized by threat and/or conflict. Findings provide new insights into the theorized function of the immune system for social approach and withdrawal patterns through which our social connections are constructed.

Peer Network Processes in Adolescents' Health Lifestyles.

Combining theories of health lifestyles-interrelated health behaviors arising from group-based identities-with those of network and behavior change, we investigated network characteristics of health lifestyles and the role of influence and selection processes underlying these characteristics. We examined these questions in two high schools using longitudinal, complete friendship network data from the National Longitudinal Study of Adolescent to Adult Health. Latent class analyses characterized each school's predominant health lifestyles using several health behavior domains. School-specific stochastic actor-based models evaluated the bidirectional relationship between friendship networks and health lifestyles. Predominant lifestyles remained stable within schools over time, even as individuals transitioned between lifestyles. Friends displayed greater similarity in health lifestyles than nonfriend dyads. Similarities resulted primarily from teens' selection of friends with similar lifestyles but also from teens influencing their peers' lifestyles. This study demonstrates the salience of health lifestyles for adolescent development and friendship networks.

High school students\u2019 friendship network, physical activity and residential locations \u2013 a stochastic actor based model

Evidence shows that adolescents do not do enough physical activity (PA), which could contribute to childhood overweight and obesity. Studies have shown that both the built environment and social networks could influence adolescents’ PA, but more studies are needed to investigate their combined influence using longitudinal data. We used a stochastic actor-based model analyzing two waves of Add Health data to test if (1) home location has a significant influence on high school student’s friendships, and (2) the neighborhood built environment has a significant influence on high school student’s PA while controlling for friendship networks. The results indicate that students’ PA level emulated peers’ PA levels and students who lived closer together, increased the likelihood of forming friendships. However, the built environment variables that described adolescents’ residential neighborhoods did not show a significant influence on students’ PA dynamics. This study contributes to our understanding of the joint impacts of social networks and home location on adolescents’ friend networks and PA dynamics in urban settings.

The Co-Evolution of Network Structure and PrEP Adoption among a Large Cohort of PrEP Peer Leaders: Implications for Intervention Evaluation and Community Capacity-Building.

Background: Peer leader interventions are effective strategies for promoting prevention behaviors in communities at risk for HIV, yet little is known about their effects on the social and behavioral dynamics of peer leaders themselves. Methods: Using data from PrEP Chicago, an RCT PrEP for prevention intervention for young Black MSM (YBMSM), we apply stochastic actor-based models to longitudinally model the impact of study participation on the online friendship and PrEP adoption dynamics among a network of peer leaders (n = 174) and a network of control group counterparts (n = 166). Results: Peer leaders assigned to the same leadership training workshop were more likely to form new Facebook friendships with one another, whereas control participants assigned to the same attention control workshop were no more or less likely to form new friendships. Further, peer leaders with greater PrEP intentions and those living with HIV were more active in forming new friendships with other peer leaders, effects not found in the control network. PrEP adoption was not influenced by network dynamics in either group. Conclusions: The implications of these findings are discussed through the lens of community-capacity building and the role that peer leader interventions and the networks they engage can impact public health.

Peer influence in adolescent drinking behavior: A meta-analysis of stochastic actor-based modeling studies.

ObjectivesTo measure the effects of peer influence and peer selection on drinking behavior in adolescence through a rigorous statistical approach designed to unravel these interrelated processes.MethodsWe conducted systematic searches of electronic databases, thesis collections and conference proceedings to identify studies that used longitudinal network design and stochastic actor-oriented modeling to analyze drinking behavior in adolescents. Parameter estimates collected from individual studies were analyzed using multilevel random-effects models.ResultsWe identified 26 articles eligible for meta-analysis. Meta-analyses for different specifications of the peer influence effect were conducted separately. The peer influence effect was positive for every specification: for average similarity (avSim) mean log odds ratio was 1.27 with 95% confidence interval [0.04; 2.49]; for total similarity (totSim) 0.46 (95% CI = [0.44; 0.48]), and for average alter (avAlt) 0.70 (95% CI = [-0.01; 1.41]). The peer selection effect (simX) was also positive: 0.46 (95% CI = [0.28; 0.63]). Conversion log odds ratio values to Cohen’s d gives estimates from 0.25 to 0.70, which is considered as medium to large effect.ConclusionsAdvances in methodology for social network analysis have made it possible to accurately estimate peer influence effects free from peer selection effects. More research is necessary to clarify the roles of age, gender, and individual susceptibility on the changing behavior of adolescents under the influence of their peers. Understanding the effects of peer influence should inform practitioners and policy makers to design and deliver more effective prevention programs.

The role of prosocial attitudes and academic achievement in peer networks in higher education

After the transition to university, students need to build a new peer network, which helps them to adapt to university life. This study investigated to what extent students’ prosocial attitudes and academic achievement facilitate the embeddedness in friendship and help-seeking networks, while taking structural network characteristics into account. Participants were 95 first-year bachelor’s degree students and were part of learning communities consisting of 12 students at a university in the Netherlands. Measures included student-reports of prosocial attitudes, peer nominations of friendship and help-seeking networks, and officially registered grades (GPA). Longitudinal social network analysis, stochastic actor-based modeling with the package RSiena, revealed that both students’ own prosocial attitudes and achievement played a role in their friendship formation, whereas only students’ own achievement made the formation of their help-seeking relationships more likely. When students were friends, it was more likely that they approached each other for help and vice versa. Similarity in achievement level contributed to relationship formation in friendship and help-seeking networks. Overall, the results underscore the importance of both student’ prosocial attitudes and achievement for their social adjustment (i.e., making friends) and only achievement for their academic adjustment (i.e., seeking help) during the first year of university within the context of small-scale teaching.

Moderators of Friend Selection and Influence in Relation to Adolescent Alcohol Use.

Friendships form an important context in which adolescents initiate and establish alcohol use patterns, but not all adolescents may be equally affected by this context. Therefore, this study tests whether parenting practices (i.e., parental discipline, parental knowledge, unsupervised time with peers) and individual beliefs (i.e., alcohol descriptive norms, positive social expectations, moral approval of alcohol use) moderate friend selection and influence around alcohol use. Stochastic actor-based models were used to analyze longitudinal social network and survey data from 12,335 adolescents (aged 11 to 17, 51.3% female) who were participating in the PROSPER project. A separate model was estimated for each moderating variable. Adolescents who reported consistent parental discipline, less unsupervised time with peers, higher descriptive alcohol use norms, and less positive social expectations about alcohol use were less likely to select alcohol-using friends. Those who reported consistent parental discipline, better parental knowledge, lower descriptive alcohol use norms, and less positive social expectations were more influenced by their friends' level of alcohol use. Thus, adolescents with these characteristics whose friends frequently use alcohol are at greater risk whereas those whose friends do not use alcohol are at lower risk of using alcohol. The findings show that, although selection and influence processes are connected, they may function in different ways for different groups of adolescents. For some adolescents, it is particularly important to prevent them from selecting alcohol-using friends, because they are more susceptible to influence from such friends. These peer network dynamics might explain how proximal outcomes targeted by many prevention programs (i.e., parenting practices and individual beliefs) translate into changes in alcohol use.

The Co-evolution of online social networks and syphilis incidence among young black men who have sex with men.

Network contexts have been implicated in the facilitation of syphilis transmission, however little is known about the relationship between online social networks, a proxy for physical networks, and rates of syphilis infection and how they co-evolve over time. To these ends, this study explored the interdependent relationship between Facebook friendship network dynamics and rates of syphilis incidence among young Black men who have sex with men (YBMSM). Specifically, we investigated the extent to which syphilis status, as a latent profile of network members, serves as a selection mechanism of Facebook tie formation and the degree to which Facebook friendships influenced dynamics of syphilis seroconversion. Sexual health, behavioral, and social network data were collected from a cohort of YBMSM (N=286, Mean age=22.8) at two time points between 2013 and 2015 in Chicago, IL, USA. The interdependencies between Facebook friendship networks and syphilis infection were assessed using stochastic actor-based models for social networks and behavior. Results showed that YBMSM tended to form Facebook friendships with other YBMSM who had similar syphilis and HIV status profiles, and the hazard of contracting syphilis was likewise influenced by their Facebook friendships, albeit subtly, by being connected to infectious network members. Meanwhile, intrinsic characteristics like condomless sex, HIV status, and online partner-seeking were not associated with dynamics in syphilis incidence. These findings show that there is a mutually dependent relationship between the online network dynamics of YBMSM and rates of syphilis incidence. Findings also direct attention to network-based strategies for intervention.

A Longitudinal Social Network Perspective on Adolescents' Exposure to Violent Video Games and Aggression.

Adolescents play video games as a social leisure activity, yet it is unclear whether peer influences play a role in spreading violent video game exposure (VVE) and aggression. It has been suggested that adolescents' aggression increases because of their friend's exposure to violent video games. This study tests this suggestion by using longitudinal social network analyses to investigate selection and socialization of aggression owing to VVE. A total of 796 adolescents from 34 different classrooms were followed from grade 7 to grade 8 (Mage = 12.60 years, 51 percent male adolescents). Exposure to violent video games, physical aggression, and within-classroom friendships were assessed at both time points. Data were analyzed by means of stochastic actor-based modeling in RSiena to estimate the effects of VVE and aggressive behavior on changes in friendships (selection), and the effects of friendships on changes in participants' VVE and aggressive behavior (socialization). Results showed homotypic selection effects, that is, adolescents became friends with peers who were similar in aggression and similar in violent video game exposure. Furthermore, there was a homotypic socialization effect, as friends became more similar in aggression over time. Friends did not become more similar in VVE over time. Violent games played by friends did not increase adolescents' own aggressive behavior. This suggests that concerns about peer influences on violent video games are unwarranted. Future studies on socialization processes of VVE should focus on influences from closest friends and investigate behavior during actual play.

Isolation but Diffusion? A Structural Account of Depression Clustering among Adolescents

Depression can cause people to withdraw from friendships or be avoided by others, protecting others from exposure to that depression. Yet, researchers observe depression contagion, particularly among adolescents. We address this empirical puzzle by examining the role of gender in structuring friendship networks and the implications for isolation and the spreading of depression. Using stochastic actor-based models of friendships among 421 adolescents from mixed-gender, all-girls, and all-boys classrooms in six Taiwanese high schools, we find that networks with only girls are characterized by high reciprocity and low transitivity. This, in turn, facilitates the withdrawal of depressed girls from interactions. In contrast, networks with all boys create more opportunities for depression to spread through interconnected pathways. Our computational experiment further demonstrates that local preferences governing friendship choice influence levels of network connectivity. This, coupled with depression withdrawal and peer influence, shapes depression prevalence at the network level. These findings refine our understanding of the mechanisms through which friendships expose boys and girls unequally to health risks of depression.

The role of peer groups in adolescents’ educational expectations: a stochastic actor-based model

ABSTRACT Adolescent educational expectations predict educational attainment, which leads to labour-market success. Given peers’ similarity in educational expectations, it is methodologically challenging to differentiate whether adolescents select friends who are similar in their educational aspirations or whether adolescents influence each other’s educational expectations. To overcome this methodological challenge, the analysis utilizes stochastic actor-based modelling. This study uses survey data from the U.S. National Longitudinal Study of Adolescent Health (Add Health). The results show that selection effects may be more prominent than influence effects in the relationship between educational expectation and adolescent friendships. Educational interventions to promote friendship formation with peers with higher educational aspirations may not result in greater educational expectations. Family bonding is associated with greater educational aspirations. Greater support of the families through work–life balance public policies and tangible strategies for improving family bonding opportunities in schools and in local communities may be needed to increase educational aspirations among adolescents.

Interplay between social support tie formations and subjective mental health conditions in a community currency system in Japanese disaster-affected communities: The ambivalent effects of social capital

A variety of community currency schemes have been implemented in Japanese disaster areas as a community-based tool to facilitate the revival of disaster-affected communities after the 2011 Great East Japan Earthquake. One of the key variables in the recovery of these disaster-affected communities is the restoration of disaster survivors' subjective mental health conditions, yet little is known about how subjective mental health coevolves with the supportive connections emerging in the postdisaster stages. Using stochastic actor-based modelling, we show that social capital, referring to postdisaster emergent social support networks and broader engagement, not only helped 101 disaster victims overcome the mental impacts of the disaster but also empowered them to provide social support to others. However, a dilemma emerged regarding the greater engagement of residents with poor psychological statuses in supportive networks during postdisaster recovery periods, which led to differential and detrimental impacts on initially mentally healthier residents’ conditions through the contagion mechanism.

Modeling the Evolutionary Mechanism of China’s Domestic Air Transport Network

With the rapid expansion of China’s domestic air transport network (CATN), it is fundamental to model which factors and mechanisms impact this development. This paper investigates how the combined endogenous and exogenous factors influencing the evolution of CATN based on longitudinal data by utilizing a more all-encompassing methodology of stochastic actor based-modeling (SABM). Endogenous variables include a density effect, a betweenness effect, a transitivity closure effect, and a ‘number of distances-two’ effect. Exogenous variables incorporate airport hierarchy, a distance effect, presence or absence of low-cost carriers (LCCs) and high-speed rail (HSR). The systematic classification of Chinese airports into more than the typical two or three tiers allows the impacts of the four endogenous covariates to be revealed. Overall, the CATN has tended to evolve into a more compacted and non-concentrated network structure through the creation of non-stop routes and closed triads. The integrated inclusion of low-cost carrier and high-speed rail effects highlights the importance of market presence to the initiation of new routes at initial stages, cultivating potential demand and increasing accessibility. In addition, the construction of HSR to one primary airport within a multi-airport system can raise “shadow effects” for other airports. Our findings provide policy suggestions for airport operators in terms of developing accurate positions in the hierarchy and strengthening transfer ability.

Causal inference in collaboration networks using propensity score methods

Using panel data of school-class networks of 11–13-year-old students, this study investigates effects of schoolwork collaboration-networks on grades and school-related well-being. It suggests propensity score weighting-regression as a method of causal inference for data collected in social contexts, and in studies analyzing node-attributes as outcomes of interest. It will argued that this alternative approach is useful when stochastic actor-based models (SAOMs) show convergence problems in sparse networks. Three methods of causal analysis dealing with the problems of endogeneity bias and interference between observations will be discussed in this study: first, SAOMs for the co-evolution of networks and behavior/attitudes will be estimated, but this results in a systematic loss of data. Second, propensity score matching compares treated cases with untreated nearest neighbors. However, the stable-unit-treatment-value assumption (SUTVA) requires that the analysis controls for network embeddedness in the final analysis. This is possible by using propensity score weighting-regression, which is a flexible approach to capture treatment diffusion via multiplex networks.

Model Adequacy Checking/Goodness-of-fit Testing for Behavior in Joint Dynamic Network/Behavior Models, with an Extension to Two-mode Networks

The recent popularity of models that capture the dynamic coevolution of both network structure and behavior has driven the need for summary indices to assess the adequacy of these models to reproduce dynamic properties of scientific or practical importance. Whereas there are several existing indices for assessing the ability of the model to reproduce network structure over time, to date there are few indices for assessing the ability of the model to reproduce individuals’ behavior patterns. Drawing on the widely used strategy of assessing model adequacy by comparing index values summarizing features of the observed data to the distribution of those index values on simulated data from the fitted model, we propose four goals that a researcher could reasonably expect of a joint structure/behavior model regarding how well it captures behavior and describe indices for assessing each of these. These reasonably simple and easily implemented indices can be used for assessing model adequacy with any dynamic network models jointly working with networks and behavior, including the stochastic actor-based models implemented within software packages such as RSien version 1.2-24. We demonstrate the use of our indices with an empirical example to show how they can be employed in practical settings, with an additional extension to modeling affiliation dynamics in two-mode networks. Key scripts are provided in the Supplemental Document (which can be found at http://smr.sagepub.com/supplemental/ ).