Actor-based Model Research Articles

Research note: The consequences of different methods for handling missing network data in stochastic actor based models

Although stochastic actor-based models (e.g., as implemented in the SIENA software program) are growing in popularity as a technique for estimating longitudinal network data, a relatively understudied issue is the consequence of missing network data for longitudinal analysis. We explore this issue in our research note by utilizing data from four schools in an existing dataset (the AddHealth dataset) over three time points, assessing the substantive consequences of using four different strategies for addressing missing network data. The results indicate that whereas some measures in such models are estimated relatively robustly regardless of the strategy chosen for addressing missing network data, some of the substantive conclusions will differ based on the missing data strategy chosen. These results have important implications for this burgeoning applied research area, implying that researchers should more carefully consider how they address missing data when estimating such models.

Co-Evolution of Social Networks and Student Performance

Peer-effects describe the influence of social environment on individual academic achievements. Social environment is usually considered as a randomly formed group. However, it forms and develops as a result of an individual conscious dynamic choice. Students may initially choose friends among peers with similar level of academic performance, and the influence of close friends on each other’s achievements may reveal itself over time. Using stochastic actor-based models, we demarcated the boundaries between social selection and social influence evaluated through academic performance. Having analyzed the dynamics of friendship ties and academic achievements throughout the first year at university, we discovered that students were not guided by the level of academic performance when choosing friends among peers but academic achievements of the latter affected their own performance over time. This could be explained by social segregation of students based on their academic outcomes.

State Distribution Policy for Distributed Model Checking of Actor Models

Model checking temporal properties is often reduced to finding accepting cycles in Buchi automata. A key ingredient for an effective distributed model ¨ checking technique is a distribution policy that does not split the potential accepting cycles of the corresponding automaton among several nodes. In this paper, we introduce a distribution policy to reduce the number of split cycles. This policy is based on the call dependency graph, obtained from the message passing skeleton of the model. We prove theoretical results about the correspondence between the cycles of call dependency graph and the cycles of the concrete state space and provide empirical data obtained from applying our distribution policy in state space generation and reachability analysis. We take Rebeca, an imperative interpretation of actors, as our modeling language and implement the introduced policy in its distributed state space generator. Our technique can be applied to other message-driven actor-based models where concurrent objects or services are units of concurrency.

Predicting Social Influence Based on Dynamic Network Structures

This study examines how network structure and dynamics interplay with the effect of social influence to facilitate diffusion. The context we consider is the diffusion of a new smartphone from a major wireless carrier in two medium-sized cities in China. The study is carried out in three stages: (1) New phone adopters (seeds) are selected within our sample period, and a two-layer snowball sampling of their mobile contacts is used to extract individual networks. (2) Given longitudinal networks and temporal adoption behaviors, the stochastic actor-based dynamic network model is used to determine the homophily effect on network formation and the social influence effect on adoptions. (3) The measures of the network structure are linked to the estimated effects of social influence based on meta-analysis. As a result, the two most significant network measures related to social influence are diversity of connection and time variation of edge numbers, after controlling for network size and density. The simulation further reveals that a certain amount of initial adopters is needed to trigger positive social influence during the diffusion process. Our findings provide a new perspective on buzz marketing campaigns by emphasizing the appropriate selection of social networks prior to the selection of individuals.

Early adolescent friendships and academic adjustment: examining selection and influence processes with longitudinal social network analysis.

This study investigated early adolescent friendship selection and social influence with regard to academic motivation (self-efficacy and intrinsic value), engagement (effortful and disruptive behavior), and achievement (GPA calculated from report card grades) among 6th graders (N = 587, 50% girls at Wave 1; N = 576, 52% girls at Wave 2) followed from fall to spring within 1 academic year. A stochastic actor-based model of social network analysis was used to overcome methodological limitations of prior research on friends, peer groups, and academic adjustment. Evidence that early adolescents sought out friends who were similar to themselves (selection) was found in regard to academic self-efficacy, and a similar trend was found for achievement. Evidence that friends became more similar to their friends over time (influence) was found for all aspects of academic adjustment except academic self-efficacy. Collectively, results indicate that selection effects were not as pervasive as influence effects in explaining similarity among friends in academic adjustment. (PsycINFO Database Record (c) 2014 APA, all rights reserved).

A Dynamic Model of Adolescent Friendship Networks, Parental Influences, and Smoking.

Peer and parental influences are critical socializing forces shaping adolescent development, including the co-evolving processes of friendship tie choice and adolescent smoking. This study examines aspects of adolescent friendship networks and dimensions of parental influences shaping friendship tie choice and smoking, including parental support, parental monitoring, and the parental home smoking environment using a Stochastic Actor-Based model. With data from three waves of the National Longitudinal Study of Adolescent Health of youth in grades 7 through 12, including the In-School Survey, the first wave of the In-Home survey occurring 6months later, and the second wave of the In-Home survey, occurring one year later, this study utilizes two samples based on the social network data collected in the longitudinal saturated sample of sixteen schools. One consists of twelve small schools (n=1,284, 50.93% female), and the other of one large school (n=976, 48.46% female). The findings indicated that reciprocity, choosing a friend of a friend as a friend, and smoking similarity increased friendship tie choice behavior, as did parental support. Parental monitoring interacted with choosing friends who smoke in affecting friendship tie choice, as at higher levels of parental monitoring, youth chose fewer friends that smoked. A parental home smoking context conducive to smoking decreased the number of friends adolescents chose. Peer influence and a parental home smoking environment conducive to smoking increased smoking, while parental monitoring decreased it in the large school. Overall, peer and parental factors affected the coevolution of friendship tie choice and smoking, directly and multiplicatively.

Timed Rebeca schedulability and deadlock freedom analysis using bounded floating time transition system

Timed Rebeca is an extension of the actor-based modeling language Rebeca that supports timing features. Rebeca is purely actor-based with no shared variables and asynchronous message passing with no explicit receive. Both computation time and network delays can be modeled in Timed Rebeca. In this paper, we propose a new approach for checking schedulability and deadlock freedom of Timed Rebeca models. The key features of Timed Rebeca, asynchrony of actors and absence of shared variables, and the focus on events instead of states in the selected properties, led us to a significant reduction in the state space. In the proposed method, there is no unique time value for each state, instead of that we store the local time of each actor separately. We prove the bisimilarity of the generated transition system, called floating time transition system, and the state space generated from the original semantics of Timed Rebeca. In addition, to avoid state space explosion because of time progress, we define a type of equivalency among states called shift equivalency. The shift equivalence relation between states can be used for Timed Rebeca as the timing features are based on relative values. We developed a tool, and the experimental results show that our approach mitigates the state space explosion problem of the former methods and allows model checking of larger systems.

The interplay of friendship networks and social networking sites: longitudinal analysis of selection and influence effects on adolescent smoking and alcohol use.

We examined the coevolution of adolescent friendships and peer influences with respect to their risk behaviors and social networking site use. Investigators of the Social Network Study collected longitudinal data during fall 2010 and spring 2011 from 10th-grade students in 5 Southern California high schools (n = 1434). We used meta-analyses of stochastic actor-based models to estimate changes in friendship ties and risk behaviors and the effects of Facebook and MySpace use. Significant shifts in adolescent smoking and drinking occurred despite little change in overall prevalence rates. Students with higher levels of alcohol use were more likely to send and receive friendship nominations and become friends with other drinkers. They were also more likely to increase alcohol use if their friends drank more. Adolescents selected friends with similar Facebook and MySpace use habits. Exposure to friends' risky online pictures increased smoking behaviors but had no significant effects on alcohol use. Our findings support a greater focus on friendship selection mechanisms in school-based alcohol use interventions. Social media platforms may help identify at-risk adolescent groups and foster positive norms about risk behaviors.

Victims, bullies, and their defenders: A longitudinal study of the coevolution of positive and negative networks

The complex interplay between bullying/victimization and defending was examined using a longitudinal social network approach (stochastic actor-based models). The (co)evolution of these relations within three elementary schools (Grades 2-5 at Time 1, ages 8-11, N = 354 children) was investigated across three time points within a year. Most bullies and defenders were in the same grade as the victims, although a substantial number of bullies and defenders were in other grades (most often one grade higher). Defenders were usually of the same gender as the victims, whereas most bullies were boys, with boys bullying both boys and girls. In line with goal-framing theory, multiplex network analyses provided evidence for the social support hypothesis (victims with the same bullies defended each other over time) as well as the retaliation hypothesis (defenders run the risk of becoming victimized by the bullies of the victims they defend). In addition, the analysis revealed that bullies with the same victims defended each other over time and that defenders of bullies initiated harassment of those bullies' victims. This study can be seen as a starting point in unraveling the relationship dynamics among bullying, victimization, and defending networks in schools.

The evolution of innovation networks: the case of a publicly funded German automotive network

In this paper we analyze the evolution of a publicly funded interfirm R&D network in the German automotive industry between 1998 and 2007. A stochastic actor-based model allows us to estimate the role of various drivers of the evolutionary change process. We test hypotheses derived in an innovation and evolutionary economics framework and show that structural positions of firms as well as actor covariates and dyadic covariates are influential determinants of network evolution. Our results indicate that knowledge-related effects such as absorptive capacity, technological distance and knowledge-base modularity are significant determinants of network evolution. Moreover, transitivity is found to be a relevant endogenous structural effect. Also, firms with an experience in cooperation tend to be engaged more intensive in cooperation both as initiators and as targets. The same holds for geographically neighboring firms.

Does It Matter What Friends Think, Say, or Do? The Role of Friends’ Smoking Attitudes and Behavior for Dutch Adolescents’ Smoking Behavior

Using stochastic actor-based models for longitudinal network analysis, this study examines the role of friends’ smoking attitudes and behavior for Dutch adolescents’ smoking behavior in four secondary schools (N = 875). The data were collected in two waves in two small suburban towns under second graders in 2008 to 2009 by means of a standardized questionnaire. Stochastic actor-based models for longitudinal network analysis can control for friendship selection while examining the effect of friends’ attitudes and smoking behavior on the smoking behavior of a student. The findings suggest that friends tend to select each other on similar smoking behavior. Influence of friends’ smoking behavior seems to play no role. In one school, an effect of friends’ attitudes towards smoking on the smoking behavior is found. The implications for future research are to consider attitudes when examining the influence of friendship network on smoking behavior. The main limitation of this study lies in the limited sample, which makes generalizations to the general population difficult.

Dynamic Social Networks in Recovery Homes

Acute treatment aftercare in the form of sober living environments-i.e., recovery houses-provide an inexpensive and effective medium-term treatment alternative for many with substance use disorders. Limited evidence suggests that house-situated social relationships and associated social support are critical determinants of how successful these residential experiences are for their members, but little is known about the mechanisms underlying these relationships. This study explored the feasibility of using dynamic social network modeling to understand house-situated longitudinal associations among individual Alcoholics Anonymous related recovery behaviors, length of residence, dyadic interpersonal trust, and dyadic confidant relationship formation processes. Trust and confidant relationships were measured 3 months apart in U.S. urban-area recovery houses, all of which were part of a network of substance use recovery homes. A stochastic actor-based model was successfully estimated from this data set. Results suggest that confidant relationships are predicted by trust, while trust is affected by recovery behaviors and length of residence. Conceptualizing recovery houses as a set of independent, evolving social networks that can be modeled jointly appears to be a promising direction for research.

Exploring the joint dynamics of intercity internet and corporate networks: a stochastic actor‐based modelling approach

AbstractIn this article, I show one possible solution to synthesis dynamics in multiple intercity networks. I adopt a stochastic actor‐based modelling approach to explore the co‐evolution of an intercity corporate network of 57 globalized advanced producer service firms across 93 cities, and another intercity internet network between these 93 cities, for the period 2004–2010. Stochastic actor‐based models (SABMs) help to connect interactions among cities and firms on the local scale with empirically observed networks on the global scale. My analysis accounts for the co‐evolution/interdependence among multiple world city networks (WCNs) and associated network changes in individual WCNs with exogenous city‐related covariates and endogenous local network structures.

Social Networks and Smoking

Adolescent smoking and friendship networks are related in many ways that can amplify smoking prevalence. Understanding and developing interventions within such a complex system requires new analytic approaches. We draw on recent advances in dynamic network modeling to develop a technique that explores the implications of various intervention strategies targeted toward micro-level processes. Our approach begins by estimating a stochastic actor-based model using data from one school in the National Longitudinal Study of Adolescent Health. The model provides estimates of several factors predicting friendship ties and smoking behavior. We then use estimated model parameters to simulate the coevolution of friendship and smoking behavior under potential intervention scenarios. Namely, we manipulate the strength of peer influence on smoking and the popularity of smokers relative to nonsmokers. We measure how these manipulations affect smoking prevalence, smoking initiation, and smoking cessation. Results indicate that both peer influence and smoking-based popularity affect smoking behavior and that their joint effects are nonlinear. This study demonstrates how a simulation-based approach can be used to explore alternative scenarios that may be achievable through intervention efforts and offers new hypotheses about the association between friendship and smoking.

The influence of friends on adolescents’ behavior problems at school: The role of ego, alter and dyadic characteristics

This paper deals with the influence of friends in class on adolescents’ problematic school behavior (i.e. inattention in class and not doing homework). We examine whether this influence is moderated by ego (i.e. the adolescent's indegree), alter (i.e. friends’ indegree) and dyadic characteristics (i.e. friendship reciprocity). Influence processes are analyzed with a stochastic actor-based model (SIENA), while controlling for friendship selection. Using a 4-wave panel dataset, we find that friends influence adolescents’ problematic school behavior. Adolescents with a higher indegree are less likely to be influenced. The influence of friends is not significantly moderated by friends’ indegree and friendship reciprocity.

Representing Micro-Macro Linkages by Actor-Based Dynamic Network Models.

Stochastic actor-based models for network dynamics have the primary aim of statistical inference about processes of network change, but may be regarded as a kind of agent-based models. Similar to many other agent-based models, they are based on local rules for actor behavior. Different from many other agent-based models, by including elements of generalized linear statistical models they aim to be realistic detailed representations of network dynamics in empirical data sets. Statistical parallels to micro-macro considerations can be found in the estimation of parameters determining local actor behavior from empirical data, and the assessment of goodness of fit from the correspondence with network-level descriptives. This article studies several network-level consequences of dynamic actor-based models applied to represent cross-sectional network data. Two examples illustrate how network-level characteristics can be obtained as emergent features implied by micro-specifications of actor-based models.

Dynamic Social Networks and Physical Aggression: The Moderating Role of Gender and Social Status Among Peers

We examined three interrelated questions: (1) Who selects physically aggressive friends?; (2) Are physically aggressive adolescents influential?; and (3) Who is susceptible to influence from these friends? Using stochastic actor-based modeling, we tested our hypotheses using a sample of 480 adolescents (ages 11-13) who were followed across four assessments (fall and spring of 6th and 7th grade). After controlling for other factors that drive network and behavioral dynamics, we found that physically aggressive adolescents were attractive as friends, physically aggressive adolescents and girls were more likely to select physically aggressive friends, and peer-rejected adolescents were less likely to select physically aggressive friends. There was an overall peer influence effect, but gender and social status were not significant moderators of influence.

Peers and the Emergence of Alcohol Use: Influence and Selection Processes in Adolescent Friendship Networks

This study addresses not only influence and selection of friends as sources of similarity in alcohol use, but also peer processes leading drinkers to be chosen as friends more often than non-drinkers, which increases the number of adolescents subject to their influence. Analyses apply a stochastic actor-based model to friendship networks assessed five times from 6th through 9th grades for 50 grade cohort networks in Iowa and Pennsylvania, which include 13,214 individuals. Results show definite influence and selection for similarity in alcohol use, as well as reciprocal influences between drinking and frequently being chosen as a friend. These findings suggest that adolescents view alcohol use as an attractive, high status activity and that friendships expose adolescents to opportunities for drinking.

Onset to First Alcohol Use in Early Adolescence: A Network Diffusion Model

A novel version of Snijders's stochastic actor-based modeling (SABM) framework is applied to model the diffusion of first alcohol use through middle school-wide longitudinal networks of early adolescents, aged approximately 11-14 years. Models couple a standard SABM for friendship network evolution with a proportional hazard model for first alcohol use. Meta-analysis of individual models for 12 schools found significant effects for friendship selection based on the same alcohol use status, and for an increased rate of onset to first use based on exposure to already-onset peers. Onset rate was greater at higher grades and among participants who spent more unsupervised time with friends. Neither selection nor exposure effects interacted with grade, adult supervision, or gender.

Direct and Indirect Peer Socialization of Adolescent Nonsuicidal Self-Injury.

This study examined direct and indirect forms of peer socialization of non-suicidal self-injury (NSSI) in adolescent friendship networks. Data were collected among 348 adolescents (55% females; Mage = 15.02 years; SD = 0.53) at 4 assessment waves. Stochastic actor-based models revealed no evidence for direct socialization of NSSI: adolescents whose friends reported higher NSSI did not increase their NSSI over time. However, indirect forms of socialization were found. After controlling for direct socialization and selection effects, friends' depressive symptoms predicted changes in male and female adolescents' NSSI and friends' impulsivity predicted changes in male adolescents' NSSI. Findings highlight the importance of extending peer influence research beyond the classical "modeling" paradigm by providing evidence that peers may indirectly socialize adolescents' NSSI.