Stochastic Actor-oriented Models Research Articles

Social interactions shape individual and collective personality in social spiders.

The behavioural composition of a group and the dynamics of social interactions can both influence how social animals work collectively. For example, individuals exhibiting certain behavioural tendencies may have a disproportionately large impact on the group, and so are referred to as keystone individuals, while interactions between individuals can facilitate information transmission about resources. Despite the potential impact of both behavioural composition and interactions on collective behaviour, the relationship between consistent behaviours (also known as personalities) and social interactions remains poorly understood. Here, we use stochastic actor-oriented models to uncover the interdependencies between boldness and social interactions in the social spider Stegodyphus dumicola We find that boldness has no effect on the likelihood of forming social interactions, but interactions do affect boldness, and lead to an increase in the boldness of the shyer individual. Furthermore, spiders tend to interact with the same individuals as their neighbours. In general, boldness decreases over time, but once an individual's boldness begins to increase, this increase accelerates, suggesting a positive feedback mechanism. These dynamics of interactions and boldness result in skewed boldness distributions of a few bold individuals and many shy individuals, as observed in nature. This group behavioural composition facilitates efficient collective behaviours, such as rapid collective prey attack. Thus, by examining the relationship between behaviour and interactions, we reveal the mechanisms that underlie the emergence of adaptive group composition and collective behaviour.

Elections and the Development of Civil War: An Empirical Study of Sri Lanka

ABSTRACTPresent empirical research on the connection between elections and civil war often treats causality as a one-way effect because it focuses exclusively on either the onset or the consequence of civil war. What is omitted in the literature is the endogenous nature of the election–conflict nexus throughout a civil war. This article dedicates itself to filling that gap. Treating administrative units in conflict as part of a rebellion network, we apply a stochastic actor-oriented model (SAOM) to the case of Sri Lanka. We find that, on the one hand, rebellion is less likely to occur in Tamil regions if national winners in presidential elections enjoy high local approval ratings (selection effect). On the other hand, regions involved in the rebellion network converge in terms of their local support for the national winner of the presidency (influence effect). Overall, our model-based simulation analysis shows that the influence effect has a larger impact on the endogenous relationship than the selection effect.

Depressive Symptomology as a Moderator of Friend Selection and Influence on Substance Use Involvement: Estimates from Grades 6 to 12 in Six Longitudinal School-Based Social Networks.

Although the contributions of friend selection and friend influence to adolescent homophily on substance use behaviors has been of enduring research interest, moderators of these processes have received relatively little research attention. Identification of factors that dampen or amplify selection and influence on substance use behaviors is important for informing prevention efforts. Whereas prior research has examined adolescent drinking, smoking, and marijuana use, the current study examined whether friend selection and friend influence operated on substance use involvement, an indicator of problematic use, and whether depressive symptomology moderated these processes. In addition, it examined whether these relationships varied from grade 6 to 12. The study used a cohort-sequential design in which three cohorts of youth (first surveyed in grades, 6, 7, and 8) in six school-based longitudinal social networks were surveyed up to seven times, yielding N = 6817 adolescents (49% female). Stochastic actor-oriented models were applied to test hypothesized relationships in the six networks, then results were synthesized in a meta-analysis. Depressive symptoms did not moderate selection or influence on substance use involvement at any grade level, but indirectly contributed to diffusion of substance use involvement through school networks via patterns of network ties. Research is needed on contextual factors, particularly in schools, that might account for when, if at all, depressive symptoms condition friend selection and influence on substance use.

Dynamics of Collaborative Networks between Contractors and Subcontractors in the Construction Industry: Evidence from National Quality Award Projects in China

AbstractUsing the method of stochastic actor-oriented models and a longitudinal data set of the National Quality Award Projects in China during 1998–2017, this study empirically investigates how pr...

Peer influence, friend selection, and gender: How network processes shape adolescent smoking, drinking, and delinquency

We examine gender differences in the extent to which the social network processes of peer influence and friend selection explain why adolescents tend to exhibit similar risky behaviors as their friends for three problem behaviors (smoking, drinking, and delinquency). Using dynamic Stochastic Actor-Oriented Models (SAOMs), we analyze five waves of data on a large sample of 13,214 adolescents from 51 friendship networks. While both processes explain patterns of risky activities for girls and boys, the delinquent behavior of girls is more susceptible to influence and girls are especially likely to select friends who have similar smoking behaviors to their own.

Selection and Influence: A Meta-Analysis of the Association Between Peer and Personal Offending

Whether people are affected by the criminal behavior of peers (the “influence” perspective) or simply prefer to associate with others who are similar in their offending (the “selection” perspective) is a long-standing criminological debate. The relatively recent development of stochastic actor-oriented models (SAOMs—also called SIENA models) for longitudinal social network data has allowed for the examination of selection and influence effects in more comprehensive ways than was previously possible. This article reports the results of a systematic review and meta-analysis of studies that use SAOMs to test for peer selection and influence effects. A systematic review and 3-level random effects meta-analysis of studies that have used SAOMs to test selection and influence dynamics for offending behavior. There is support for both influence (mean log odds ratio = 1.23, p < 0.01, 21 effects, pooled n = 21,193) and selection dynamics (mean log odds ratio = 0.31, p < 0.01, 28 effects, pooled n = 21,269). Type of behavior, country, and the year of the first wave of data collection are found to moderate the influence effect; no significant moderation effects are found for peer selection on offending. People are both influenced by the offending of their peers and select into friendships based on similarity in offending.

Statistical Power in Longitudinal Network Studies

Longitudinal social network studies can easily suffer from insufficient statistical power. Studies that simultaneously investigate change of network ties and change of nodal attributes (selection and influence studies) are particularly at risk because the number of nodal observations is typically much lower than the number of observed tie variables. This article presents a simulation-based procedure to evaluate statistical power of longitudinal social network studies in which stochastic actor-oriented models are to be applied. Two detailed case studies illustrate how statistical power is strongly affected by network size, number of data collection waves, effect sizes, missing data, and participant turnover. These issues should thus be explored in the design phase of longitudinal social network studies.

Sinister connections: How to analyse organised crime with social network analysis?

Networks have recently become ubiquitous in many scientific fields. In criminology, social network analysis (SNA) provides a potent tool for analysis of organized crime. This paper introduces basic network terms and measures as well as advanced models and reviews their application in criminological research. The centrality measures – degree and betweenness – are introduced as means to describe relative importance of actors in the network. The centrality measures are useful also in determining strategically positioned actors within the network or providing efficient targets for disruption of criminal networks. The cohesion measures, namely density, centralization, and average geodesic distance are described and their relevance is related to the idea of efficiency-security trade-off. As the last of the basic measures, the attention is paid to subgroup identification algorithms such as cliques, k-plexes, and factions. Subgroups are essential in the discussion on the cell-structure in criminal networks. The following part of the paper is a brief overview of more sophisticated network models. Models allow for theory testing, distinguishing systematic processes from randomness, and simplification of complex network structures. Quadratic assignment procedure, blockmodels, exponential random graph models, and stochastic actor-oriented models are covered. Some important research examples include similarities in co-offending, core-periphery structures, closure and brokerage, and network evolution. Subsequently, the paper reflects the three biggest challenges for application of SNA to criminal settings – data availability, proper formulation of theories and adequate methods application. In conclusion, readers are referred to books and journals combining SNA and criminology as well as to software suitable to carry out SNA.

Illicit Network Dynamics: The Formation and Evolution of a Drug Trafficking Network

The project aims to: (1) investigate structural and functional changes in an Australian drug trafficking network across time to determine ways in which such networks form and evolve. To meet this aim, the project will answer the following research questions: (1) What social structural changes occur in drug trafficking networks across time? (2) How are these structural changes related to roles/tasks performed by network members? (3) What social processes can account for change over time in drug trafficking networks? The relational data on the network was divided into four two years periods. Actors were allocated to specific roles. We applied a stochastic actor-oriented model to explain the dynamics of the network across time. Using RSiena, we estimated a number of models with the key objectives of investigating: (1) the effect of roles only; (2) the endogenous effect of degree-based popularity (Matthew effect); (3) the endogenous effect of balancing connectivity with exposure (preference for indirect rather than direct connections); (4) how degree-based popularity is moderated by tendencies towards reach and exposure. Preferential attachment is completely moderated by a preference for having indirect ties, meaning that centralization is a result of actors preferring indirect connections to many others and not because of a preference for connecting to popular actors. Locally, actors seek cohesive relationships through triadic closure. Actors do not seek to create an efficient network that is highly centralized at the expense of security. Rather, actors strive to optimize security through triadic closure, building trust, and protecting themselves and actors in close proximity through the use of brokers that offer access to the rest of the network.

Innovation in the bioeconomy – dynamics of biorefinery innovation networks

ABSTRACTThe bioeconomy has become a central concept in many strategies for future economic development, emphasising an increasing need for collaboration across industries and sectors for innovation. This paper unpacks aspects of collaboration in the bioeconomy by looking at the development of innovation networks for biorefinery technologies from 2004 to 2014 based on innovation project data from Swedish public funding agencies using a stochastic actor-oriented model for network analysis. The analysis shows that although the network grew significantly during the time period, indicating an increasing interest in biorefinery technology innovation, inter-sectoral collaboration is not favoured over intra-sectoral collaboration. As is known from previous work on social networks trust-building is a key driver for collaboration, as actors tend to form collaborations with previous partners or indirectly connected partners, creating clustered networks.

Social influence on observed race

This article introduces a novel theoretical approach for understanding racial fluidity, emphasizing the social embeddedness of racial classifications. We propose that social ties affect racial perceptions through within-group micromechanisms, resulting in discrepancies between racial self-identifications and race as classified by others. We demonstrate this empirically on data from 12 Hungarian high school classes with one minority group (the Roma) using stochastic actor-oriented models for the analysis of social network panel data. We find strong evidence for social influence: individuals tend to accept their peers' judgement about another student’s racial category; opinions of friends have a larger effect than those of nonfriends. Perceived social position also matters: those well-accepted among majority-race peers are likely to be classified as majority students themselves. We argue that similar analyses in other social contexts shall lead to a better understanding of race and interracial processes.

Endogenous dynamics of innovation networks in the German automotive industry: analysing structural network evolution using a stochastic actor-oriented approach

The generation of innovation is well known to be a social process depending on mutual interactions, aiming at accessing and exchanging knowledge in order to generate novel goods and services. Accordingly, interest in interfirm innovation networks has increased sharply over the last decade. Preceding research indicates that the structural dynamics of networks is driven both by endogenous and exogenous forces. In particular, we focus on the role of the endogenous determinants of the network evolution of interfirm networks - a category of often underestimated forces. We employ a longitudinal dataset that comprises German automotive firms' performance between 2002 and 2006 and apply a stochastic actor-oriented model (SAOM) designed to analyse both the endogenous and exogenous determinants of network change. Our results show that endogenous determinants - approximated by measures for local and global clustering - exhibit greater explanatory power than exogenous firm characteristics such as age, size, and R&D activity.

The Emergence and Stability of Groups in Social Networks

An important puzzle in social network research is to explain how macro-level structures emerge from micro-level network processes. Explaining the emergence and stability of groups in social networks is particularly difficult for two reasons. First, because groups are characterized both by high connectedness within (group cohesion) and lack of connectedness between them (group boundaries). Second, because a large number of theoretical micro-level network processes contribute to their emergence. We argue that traditional social network theories that are concerned with the evolution of positive relations (forces of attraction) are not sufficient to explain the emergence of groups because they lack mechanisms explaining the emergence of group boundaries. Models that additionally account for the evolution of negative ties (forces of repulsion) may be better to explain the emergence and stability of groups in social networks. We build a theoretical model and illustrate it by fitting stochastic actor-oriented models (SAOMs) to empirical data of co-evolving networks of friendship and dislike among 479 secondary-school students. The SAOMs include a number of newly developed effects expressing the co-evolution between positive and negative ties. We then simulate artificial networks from the estimated models to explore the micro-macro link. We find that a model that considers forces of attraction and repulsion simultaneously is better at explaining groups in social networks. In the long run, however, the empirically informed simulations generate networks that are too stylized to be realistic, raising further questions about model degeneracy and time heterogeneity of group processes.

Endogenous dynamics of innovation networks in the German automotive industry: analysing structural network evolution using a stochastic actor-oriented approach

The generation of innovation is well known to be a social process depending on mutual interactions, aiming at accessing and exchanging knowledge in order to generate novel goods and services. Accordingly, interest in interfirm innovation networks has increased sharply over the last decade. Preceding research indicates that the structural dynamics of networks is driven both by endogenous and exogenous forces. In particular, we focus on the role of the endogenous determinants of the network evolution of interfirm networks - a category of often underestimated forces. We employ a longitudinal dataset that comprises German automotive firms' performance between 2002 and 2006 and apply a stochastic actor-oriented model (SAOM) designed to analyse both the endogenous and exogenous determinants of network change. Our results show that endogenous determinants - approximated by measures for local and global clustering - exhibit greater explanatory power than exogenous firm characteristics such as age, size, and R&D activity.

Frequent and/or Durable? The Predictive Impact of Initial Face-to-Face Contacts on the Formation and Evolution of Students’ Developmental Peer Network Relationships

Despite the relevance of relationships to others acting towards advancing a person’s career for individual career development, there is little research on how developmental relationships emerge and evolve over time. This two-year longitudinal study analyses the predictive impact of frequency and duration of initial face-to-face social interaction among newcomer students measured via RFID technology. Results from stochastic actor-oriented models suggest that the more frequent and durable initial contacts were at the first encounter, the more likely actors will create a developmental relationship, supporting proximity and mere-exposure-effect theory. Moreover, they show long-term effect tendencies of initial contacts’ frequency on the network’s evolution.

Coevolution of Networking and User Innovation in an Online Community

Online user communities are increasingly important organizational structures and they can be the sources of competitive advantages and engines of collaborative innovation. Yet, only little is known about the dynamics of tie creation between community members in such communities and of how user innovation develops over time. This paper asks how social networking in online user communities evolve over time and how user innovation spreads through such communities. It also asks how user innovators behave in the communities compared with regular users. The analysis comprises three snapshots of a social network between 165 core community members over a two-year period, including their ties to peripheral community members outside the core. Both dynamic tie formation and adoption of user innovation behavior are analyzed, applying a dynamic stochastic actor-oriented modeling framework for social network analytics. This study finds that specific patterns of dynamic social networking precedes occurrences of user innovation in the community, and that user innovation in turn affects network evolution. These findings align with the theoretically understood but rarely empirically shown conception that social networks and innovation are endogenous to each-other, whereas previous research tends to infer the one while keeping the other constant.

The Dynamic Relationship between Unhealthy Weight Control and Adolescent Friendships: A Social Network Approach.

Although adolescence marks a vulnerable stage for peer influence on health behavior, little is known about the longitudinal and dynamic relationship between adolescent friendship and weight control. The current study aims to explain these dynamic processes among a sample of 1156 American adolescents in grades 9-11 (48.6% girls, 23.4% European American, 25.2% African American) from the National Longitudinal Study of Adolescent Health. Stochastic actor-oriented models were fit to examine changes in friendship networks and unhealthy weight control across two waves. The findings support a bidirectional relationship where weight control predicts future friendship seeking and friendship seeking predicts future weight control. The findings also indicate that adolescents prefer friends with similar weight control patterns. Taken together, the results of the current study indicate that adolescent friendships play an integral role in the development of unhealthy weight control and thus can be used to identify adolescents at risk and serve as targets within preventive interventions.

Creation and persistence of ties in cluster knowledge networks

Knowledge networks are important to understand learning in industry clusters but surprisingly little is known about what drives the formation, persistence and dissolution of ties. Applying stochastic actor-oriented models on longitudinal relational data from a mature cluster in a medium-tech industry, we show that triadic closure and geographical proximity increase the probability of tie creation but does not influence tie persistence. Cognitive proximity is positively correlated to tie persistence but firms create ties to cognitively proximate firms only if they are loosely connected through common third partners. We propose a micro-perspective to understand how endogenous network effects, cognitive proximity of actors and their interplay influence the evolutionary process of network formation in clusters.

Co-evolution of social networks and continuous actor attributes

Social networks and the attributes of the actors in these networks are not static; they may develop interdependently over time. The stochastic actor-oriented model allows for statistical inference on the mechanisms driving this co-evolution process. In earlier versions of this model, dynamic actor attributes are assumed to be measured on an ordinal categorical scale. We present an extension of the stochastic actor-oriented model that does away with this restriction using a stochastic differential equation to model the evolution of continuous actor attributes. We estimate the parameters by a procedure based on the method of moments. The proposed method is applied to study the dynamics of a friendship network among the students at an Australian high school. In particular, we model the relationship between friendship and obesity, focusing on body mass index as a continuous co-evolving attribute.

Young immigrants' host country identification and their friendships with natives: Does relative group size matter?

Recent network research indicates that native youth prefer to befriend immigrants with stronger rather than weaker host country identification. Surprisingly, however, no respective preference of high-identifying immigrants for native friends has been found, and there is little evidence that friends influence immigrants' identification. Seeking to make sense of these unexpected findings, my aims are twofold: First, I reproduce an earlier study using three waves of newly collected network panel data. Second, going beyond a robustness test with better data, I suggest that relative group size within school accounts for earlier findings. I hypothesize that immigrants' host country identification only affects their own friendship choices in schools with high shares of immigrants, because only in those schools they can be picky about befriending natives. Stochastic actor-oriented models support this notion, pointing to an interplay of preferences and opportunities in shaping the relation between host country identification and interethnic friendships.