Search

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). Funding Agencies|University of Groningen

Item does not contain fulltext Although students are part of a group of classmates, they spend the majority of time during lessons with students who are seated next or close to them. Therefore, near-seated peers in elementary school classrooms might play a crucial role in students' academic development. It was hypothesized that near-seated peers influence students' academic engagement and achievement, especially when they are also friends. Participants were 559 fourth-sixth grade students (21 classrooms; 51.9% boys; Mage = 10.65 years, range = 8-12). Longitudinal social network analysis (RSiena) showed that students' academic engagement and achievement got better when friends scored better, and vice versa, regardless of their physical position in the classroom. In contrast, near-seated peers who were not befriended got more diverse scores over time. These results imply that teachers should consider students' friendships and academic engagement and achievement in designing seating arrangements. Moreover, it is recommended to actively monitor ongoing peer influence processes. 11 p.

Background: School bullying is a wide-spread problem with severe consequences for victims, bullies and bystanders. Schools are strongly encouraged to implement both schoolwide, preventive interventions and reactive measures to handle existing bullying situations. In the Dutch implementation of the KiVa anti-bullying programme, pervasive-bullying situations are addressed according to the support group approach. The support group approach is widely used for addressing bullying situations, but little is known about its effectiveness.Purpose: We investigated the effectiveness of the support group approach in reducing victimisation, increasing defending and improving the victim’s well-being over the course of a school year, over and beyond of the effects of the universal KiVa intervention.Programme description: The support group approach is a non-punitive, problem-solving strategy to address pervasive-bullying situations. In this intervention, trained teachers form a support group that consists of 6–8 children, including the bullies and their assistants, defenders or friends of the victim and prosocial classmates. The purpose of the support group is to create mutual concern for the well-being of the victim and to trigger the bullies’ willingness to alter their behaviour.Sample: We used data from 66 Dutch elementary schools that participated in the KiVa intervention study. Data were collected in October 2012 and 2013, and May 2013 and 2014. The sample used in the analyses consisted of 38 victims for whom a support group intervention was organised (44.7% boy, Mage = 9.24; SDage = 1.20).Design and method: To get insight into the effects above and beyond those of the KiVa programme itself, victims with a support group (N = 38) were matched to similar victims without a support group (N = 571). Statistical analyses were undertaken to examine whether the changes in victimisation, defending and well-being at school differed between the two groups.Result: Victims reported positive effects of the support group approach in reducing victimisation in the short term, but this decrease in victimisation was not lasting over the course of a school year. The intervention also did not improve the victims’ well-being at school in the longer term. Victims with a support group, however, were found to have more defenders at the end of the school year than victims without a support group.Conclusion: The effectiveness of the support group approach in tackling bullying situations appears to fade over time. The findings of this study illustrate that for future evaluations of anti-bullying interventions it is essential to investigate longer term effects.

In this study, the associations between peer effects and academic functioning in middle adolescence (N = 342; 14-15 years old; 48% male) were investigated longitudinally. Similarity in achievement (grade point averages) and unexplained absences (truancy) was explained by both peer selection and peer influence, net of acceptance, and connectedness. Friendships were formed and maintained when adolescents had low levels of achievement or high levels of truancy. Friends influenced one another to increase rather than decrease in achievement and truancy. Moreover, friends' popularity moderated peer influences in truancy in reciprocal friendships but not in unilateral friendships, whereas friends' acceptance moderated peer influences in achievement in both unilateral and reciprocal friendships. The findings illustrate the dynamic interplay between peer effects and academic functioning.

This study addresses the interacting effects of classroom cohesion and hierarchy on the relationships between victimization and aggression with peer acceptance and rejection. Classroom cohesion and hierarchy were constructed from friendship nominations. Multilevel analysis conducted in a sample of seventh- and eighth-grade students from the Sociescuela program in Spain ( N = 6,600) showed that in cohesive and hierarchical classrooms, a higher level of victimization was found; peer rejection was more strongly associated with victimization. In contrast to previous research, for males, aggression was more strongly associated with peer acceptance in less hierarchized classrooms.

This study examined to what extent adolescents’ and their friends’ risk behaviors (i.e., delinquency and alcohol use) hinder or promote their academic achievement (grade point average [GPA]), and vice versa. Longitudinal data were used (N = 1,219 seventh- to ninth-grade adolescents; Mage = 13.69). Results showed that risk behaviors negatively affected adolescents’ GPA, whereas GPA protected against engaging in risk behaviors. Moreover, adolescents tended to select friends who have similar behaviors and friends’ behaviors became more similar over time (same-behavior selection and influence). Furthermore, although same-behavior effects seemed to dominate, evidence was found for some cross-behavior selection effects and a tendency in seventh grade for cross-behavior influence effects. Concluding, it is important to investigate the interplay between different behaviors with longitudinal social network analysis. Funding agencies: Dutch Scientific Organization (NWO) Program Council for Fundamental Scientific Education Research (PROO) [411-12-027]

The aim of this study was to unravel the interrelatedness of friendship and help, and to examine the characteristics of friendship and help networks. The effects of mutual versus one‐sided help relations on friendship initiation and maintenance, and vice versa, were examined. Friendship and help networks were analyzed (N = 953 students; 41 classrooms; Mage = 12.7). The results illustrate that friendship and help networks show some similarities, but only partly overlap and have distinct characteristics. Longitudinal multiplex social network analyses showed that mutual help was important for the maintenance of friendship, but not for the initiation of friendship. Further, particularly mutual friendships provided a context in which help took place. Implications of these findings are discussed.

Peer cliques form an important context for the social development of adolescents. Although clique members are often similar in social status, also within cliques, status differences exist. How differences in social status between clique members are related to behaviors of its individual members is rather unknown. This study examined to what extent the relationship of individual social status (i.e., perceived popularity) with aggression and prosocial behavior depends on the level of internal clique hierarchy. The sample consists of 2674 adolescents (49.8 % boys), with a mean age of 14.02. We focused specifically on physical and relational aggression, and practical and emotional support, because these behaviors have shown to be of great importance for social relationships and social standing among adolescents. The internal status hierarchy of cliques was based on the variation in individual social status between clique members (i.e., clique hierarchization) and the structure of status scores within a clique (pyramid shape, inverted pyramid, or equal distribution of social status scores) (i.e., clique status structure). The results showed that differences in aggressive and prosocial behaviors were particularly moderated by clique status structure: aggression was stronger related to individual social status in (girls') cliques where the clique status structure reflected an inverted pyramid with relatively more high status adolescents within the clique than low status peers, and prosocial behavior showed a significant relationship with individual social status, again predominantly in inverted pyramid structured (boys' and girls') cliques. Furthermore, these effects differed by types of gender cliques: the associations were found in same gender but not mixed-gender cliques. The findings stress the importance of taking into account internal clique characteristics when studying adolescent social status in relationship to aggression and prosociality.

A common assumption about Freshmen Learning Communities (FLCs) is that academic relationships contribute to students’ success. This study investigates how students in learning communities connect with fellow students for friendship and academic support. Longitudinal social network data across the first year, collected from 95 Dutch students in eight FLCs, measure both social and academic relational choices within and beyond the FLCs. Using stochastic actor-based models, the study tests two competing hypotheses. The alignment hypothesis states that students connect with their similar-achieving friends for both academic and social support, leading to an alignment of both types of networks over time. In contrast, the duality hypothesis states dissimilarity between academic support networks and friendship networks: students should connect with better-achieving fellow students for academic support and to more similar peers for friendship. The data support the alignment hypothesis but not the duality hypothesis; in addition, they show evidence of achievement segregation in FLCs: the higher the students’ achievement level, the more they connect with other students for both academic support and friendship, relating in particular to peers with a similarly high achievement level. The results suggest that lower-achieving students are excluded from the support provided by higher-achieving students and instead ask similar lower achievers for support. They thus cannot benefit optimally from the academic integration FLC offer. The article concludes with recommendations of how to support students in an FLC so that they can reach optimal achievement potential.

While several models for analysing longitudinal network data have been proposed, their main differences, especially regarding the treatment of time, have not been discussed extensively in the literature. However, differences in treatment of time strongly impact the conclusions that can be drawn from data. In this article we compare auto-regressive network models using the example of TERGMs – a temporal extensions of ERGMs – and process-based models using SAOMs as an example. We conclude that the TERGM has, in contrast to the ERGM, no consistent interpretation on tie-level probabilities, as well as no consistent interpretation on processes of network change. Further, parameters in the TERGM are strongly dependent on the interval length between two time-points. Neither limitation is true for process-based network models such as the SAOM. Finally, both compared models perform poorly in out-of-sample prediction compared to trivial predictive models.

This paper compares several imputation methods for missing data in network analysis on a diverse set of simulated networks under several missing data mechanisms. Previous work has highlighted the biases in descriptive statistics of networks introduced by missing data. The results of the current study indicate that the default methods (analysis of available cases and null-tie imputation) do not perform well with moderate or large amounts of missing data. The results further indicate that multiple imputation using sophisticated imputation models based on exponential random graph models (ERGMs) lead to acceptable biases even under large amounts of missing data.

This chapter proposes a statistical method for investigating network structure together with relevant actor attributes as joint dependent variables in a longitudinal framework, assuming that data have been collected according to a panel design. It discusses the specification of the stochastic model for dynamics of networks and behavior and then proceeds to parameter estimation and model selection. The chapter analyses network-behavioral coevolution and parameter estimation. It addresses Goodness-of-fit issues and model selection. The chapter presents a statistical model for the simultaneous, mutually dependent, dynamics of a relation on a given set of social actors, and the behavior of these actors as represented by one or more ordinal categorical variables. The process of network-behavioral coevolution is regarded as an emergent group-level result of the network actors’ individual decisions. Whereas the rate functions model the timing of the different actors’ different types of decisions, the objective functions model which changes are made.

This study concerns peer selection and influence dynamics in early adolescents' friendships regarding academic achievement. Using longitudinal social network analysis (RSiena), both selection and influence processes were investigated for students' average grades and their cluster-specific grades (i.e., language, exact, and social cluster). Data were derived from the SNARE (Social Network Analysis of Risk behavior in Early adolescence) study, using 6 waves (N = 601; Mage = 12.66, 48.9% boys at first wave). Results showed developmental differences between the first and second year of secondary school (seventh and eighth grade). Whereas selection processes were found in the first year on students' cluster-specific grades, influence processes were found in the second year, on both students' average and cluster-specific grades. These results suggest that students initially tend to select friends on the basis of similar cluster-based grades (first year), showing that similarity in achievement is attractive for friendships. Especially for low-achieving students, similar-achieving students were highly attractive as friends, whereas they were mostly avoided by high-achieving students. Influence processes on academic achievement take place later on (second year), when students know each other better, indicating that students' grades become more similar over time in response to their connectedness. Concluding, this study shows the importance of developmental differences and specific school subjects for understanding peer selection and influence processes in adolescents' academic achievement. (PsycINFO Database Record

Interfirm networks improve performance outcomes and higher performance is contagious between cooperating firms.

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 microspecifications of actor-based models.

The present study examined whether parenting and parental smoking can prevent children from selecting smoking friends during adolescence. 254 Adolescents of one Belgian secondary school participated. Self-administered questionnaires were distributed among 2nd-4th graders (mean ages = 14.2-16.2 years) during spring 2006. Follow-up was conducted 12 months later. Data was analyzed conducting longitudinal social network analyses. Results showed adolescents perceiving high parental psychological control had a significant higher tendency to select smoking friends. Perceived behavioral control and perceived parental support did not affect the selection of smoking friends. Furthermore, maternal smoking behavior affected the selection of smoking friends, although no effect of paternal smoking behavior on the selection of smoking friends was found. Adolescent smoking prevention efforts should focus on the influence of parents through their smoking behavior and their psychological control to decrease adolescents' tendency to select smoking friends resulting in fewer opportunities for negative peer influences to occur. (C) 2013 The Foundation for Professionals in Services for Adolescents. Published by Elsevier Ltd. All rights reserved.

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., Sociological Methods & Research, 49 (4), ISSN:0049-1241, ISSN:1552-8294

We investigated adolescent prosocial relations by examining social networks based on the question "Who helps you (e.g., with homework, with repairing a flat [bicycle] tire, or when you are feeling down?)." The effects of individual characteristics (academic achievement, symptoms of depressive mood, and peer status) on receiving help and giving help were examined, and we investigated the contribution of (dis)similarity between adolescents to the development of prosocial relations. Gender, structural network characteristics, and friendship relations were taken into account. Data were derived from the Social Network Analysis of Risk behavior in Early adolescence (SNARE) study, and contained information on students in 40 secondary school classes across 3 waves (N = 840, M age = 13.4, 49.7% boys). Results from longitudinal social network analyses (RSiena) revealed tendencies toward reciprocation of help and exchange of help within helping groups. Furthermore, boys were less often mentioned as helpers, particularly by girls. Depressed adolescents were less often mentioned as helpers, especially by low-depressed peers. Moreover, lower academic achievers indicated that they received help from their higher achieving peers. Rejected adolescents received help more often, but they less often helped low-rejected peers. Last, low- and high-popular adolescents less often helped each other, and also high-popular adolescents less often helped each other. These findings show that (dis)similarity in these characteristics is an important driving factor underlying the emergence and development of prosocial relations in the peer context, and that prosocial behavior should be defined in terms of benefitting particular others. (PsycINFO Database Record

Current understanding of the emergence of communication networks is hindered by at least two limitations: (a) studies tend to incorporate only 1 theoretical mechanism, and (b) studies tend to be cross-sectional. We address these concerns by analyzing the influence of 11 factors on the evolution of a communication network over 2 years. Our model is grounded in the multitheory, multilevel framework, and our analysis uses recently developed actor-oriented techniques for statistically modeling network emergence. Results suggest accurate and complete understanding of network evolution depends on studies simultaneously incorporating multiple theories that offer both complementary and contrasting explanations of the phenomenon.

Objective: Similarity in alcohol consumption among adolescent friends could be caused by the influence of friends or by the selection of friends who consume similar levels of alcohol. This article aims to disentangle influence and selection processes while specifically examining changes over time in these processes and possible differences between reciprocal and nonreciprocal friendships. Method: The design was longitudinal with four observations (Time 1-Time 4 [T1-T4]). Data consisted of a longitudinal sample of 1,204 Finnish adolescents in I 0 junior high schools. The main measurements were adolescents' friendship networks and alcohol consumption. For three successive periods, T1-T2, T2-T3, and T3-T4, actor-based models for the co-evolution of networks and behavior were analyzed (M(age): T1 = 13.6 years, T2 = 14.6 years, T3 = 15.6 years, T4 = 16.1 years). Results: Selection, as well as influence processes, played an important role in adolescent alcohol consumption. Influence was found during the first period (T1-T2), whereas support for selection was found during the last two periods (T2 T3 and 13 T4). The strength of influence and selection processes did not differ for reciprocal and nonreciprocal friendships. Conclusions: The impact of selection and influence processes changed over time such that influence was only present during early adolescence, whereas selection was present during mid-adolescence. During early adolescence, youngsters would benefit from learning to resist social influence. Alcohol-consumption prevention programs targeting mid-adolescence should consider peer selection processes. These findings stress the importance of considering changes over time in future practice and research. (J Stud. Alcohol Drugs, 73, 99-110,2012)

Studies of peer effects in educational settings confront two main problems. The first is the presence of endogenous sorting which confounds the effects of social influence and social selection on individual attainment. The second is how to account for the local network dependencies through which peer effects influence individual behavior. We empirically address these problems using longitudinal data on academic performance, friendship, and advice seeking relations among students in a full-time graduate academic program. We specify stochastic agent-based models that permit estimation of the interdependent contribution of social selection and social influence to individual performance. We report evidence of peer effects. Students tend to assimilate the average performance of their friends and of their advisors. At the same time, students attaining similar levels of academic performance are more likely to develop friendship and advice ties. Together, these results imply that processes of social influence and social selection are sub-components of a more general a co-evolutionary process linking network structure and individual behavior. We discuss possible points of contact between our findings and current research in the economics and sociology of education. (C) 2011 Elsevier Inc. All rights reserved.

Structuration theory (ST) and network analysis are promising approaches for studying the emergence of communication networks. We offer a model that integrates the conceptual richness of structuration with the precision of relevant concepts and mechanisms offered from communication network research. We leverage methodological advancements (i.e., stochastic actor-oriented models) to test hypotheses deduced from ST using longitudinal communication network data collected over a two-year period. Results indicate that while structural rules external to a social network play a significant role, internal structural rules that emerge from the aggregate of individual actions during previous time periods also predict current structures, and that the reification influence of the latter is greater than that of the external factors.

This study applies multilevel social network analytic techniques to examine processes of homophilic selection and social influence related to alcohol use among friends in early adolescence. Participants included 3,041 Dutch youth (M age =12 years, 49% female) from 120 classrooms in 14 schools. Three waves with 3-month intervals of friendship nomination data and self-reports of drinking behavior were collected. Results revealed that within classrooms, friendship nominations tended to be reciprocated and dyadic friendships tended to be embedded within cohesive subgroups (e.g., cliques). Students tended to nominate friends who were the same sex, from a similar ethnic background, and who they previously knew from primary school. Selection processes turned out to play a more significant role than social influence processes in predicting similarity between early adolescent friends' alcohol use. Although friendship dynamics and individual drinking trajectories substantially differed between classrooms, the effects of homophilic selection and social influence did not.

A recurrent problem in the analysis of behavioral dynamics, given a simultaneously evolving social network, is the difficulty of separating the effects of partner selection from the effects of social influence. Because misattribution of selection effects to social influence, or vice versa, suggests wrong conclusions about the social mechanisms underlying the observed dynamics, special diligence in data analysis is advisable. While a dependable and validmethod would benefit several research areas, according to the best of our knowledge, it has been lacking in the extant literature. In this paper, we present a recently developed family of statistical models that enables researchers to separate the two effects in a statistically adequate manner. To illustrate our method, we investigate the roles of homophile selection and peer influence mechanisms in the joint dynamics of friendship formation and substance use among adolescents. Making use of a three-wave panel measured in the years 1995–1997 at a school in Scotland, we are able to assess the strength of selection and influence mechanisms and quantify the relative contributions of homophile selection, assimilation to peers, and control mechanisms to observed similarity of substance use among friends.

Aims The main goal of this study was to examine differences between adolescent male and female friendship networks regarding smoking-based selection and influence processes using newly developed social network analysis methods that allow the current state of continuously changing friendship networks to act as a dynamic constraint for changes in smoking behaviour, while allowing current smoking behaviour to be simultaneously a dynamic constraint for changes in friendship networks. Design Longitudinal design with four measurements. Setting Nine junior high schools in Finland. Participants A total of 1163 adolescents (mean age = 13.6 years) who participated in the control group of the ESFA (European Smoking prevention Framework Approach) study, including 605 males and 558 females. Measurements Smoking behaviour of adolescents, parents, siblings and friendship ties. Findings Smoking-based selection of friends was found in male as well as female networks. However, support for influence among friends was found only in female networks. Furthermore, females and males were both influenced by parental smoking behaviour. Conclusions In Finnish adolescents, both male and female smokers tend to select other smokers as friends but it appears that only females are influenced to smoke by their peer group. This suggests that prevention campaigns targeting resisting peer pressure may be more effective in adolescent girls than boys.

Positive association of relevant characteristics is a widespread pattern among adolescent friends. A positive association may be caused by the selection of similar others as friends and by the deselection of dissimilar ones, but also by influence processes where friends adjust their behavior to each other. Social control theory argues that adolescents select each other as friends based on delinquency. Differential association theory, on the other hand, argues that adolescent friends influence each other's delinquency levels. We employ new statistical methods for assessing the empirical evidence for either process while controlling for the other process. These methods are based on ‘actor‐oriented’ stochastic simulation models. We analyze longitudinal data on friendship networks and delinquent behavior collected in four waves of 544 students in 21 first‐grade classrooms of Dutch secondary schools. Results indicate that adolescents select others as friends who have a similar level of delinquency compared with their own level. Estimates of the social influence parameters are not significant. The results are consistent with social control theory but provide no support for differential association theory.

The mutual influence of smoking behavior and friendships in adolescence is studied. It is attempted to disentangle influence and selection processes in reciprocal and non-reciprocal friendships. An actor-based model is described for the co-evolution of friendship networks and smoking behavior. This model considers alternative selection and influence mechanisms, and models continuous-time changes in network and behavior. The data consists of a longitudinal sample of 1326 Finnish adolescents in 11 high schools. Findings suggest that selection as well as influence Processes play air important role in adolescent smoking behavior. Selection had a relatively stronger role than influence, in particular when selecting non-reciprocal friends. The strength of both influence and selection processes decreased over time. (C) 2009 Elsevier B.V. All rights reserved.

This research considers how the connectedness of venture capital (VC) firms changes with the job mobility of managers between those firms. A continuous-time Markov chain model is developed to test whether managers are able to 'drag' prior inter-firm relational ties with them as they move between employing firms. The findings support the hypothesis that increasing the stock of managers is material to creating inter-firm ties. However, managers are not found to drag their prior links to new network positions. This result is consistent with an institutional view of VC syndication, suggesting that inter-firm relations are other than managerial in nature. European Management Review (2007) 4, 161-171. doi: 10.1057/palgrave.emr.1500083

We give a nontechnical introduction into recently developed methods for analyzing the coevolution of social networks and behavior(s) of the network actors. This coevolution is crucial for a variety of research topics that currently receive a lot of attention, such as the role of peer groups in adolescent development. A family of dynamic actor-driven models for the coevolution process is sketched, and it is shown how to use the SIENA software for estimating these models. We illustrate the method by analyzing the coevolution of friendship networks, taste in music, and alcohol consumption of teenagers.

The association between experienced victimization and students' psychological and social adjustment depends on the intensity of victimization. We examined how frequency and multiplicity of victimization, and the number of bullies involved, account for differences in students' psychosocial well-being and social standing in the classroom. Multilevel analyses were conducted on the control group of an intervention study among students in grades 3-6 of Dutch elementary schools (N = 2859 students from 124 classes and 33 schools; ages 8-12; 49.6% boys). It was found that victims of frequent and multiple victimization, and victims who were victimized by several bullies, had higher levels of psychosocial adjustment problems than victims of less frequent and non-multiple victimization, and victims with only one bully. Moreover, these more severe victims turned out to be least accepted and most rejected among their classmates. The findings illustrate that it can be fruitful to use several measures of victimization so that (differences in) adjustment problems can be better understood. Moreover, the results suggest that it is important to find out who is victimized, in what ways, and by whom. Anti-bullying interventions should provide resources to do this.

We study the role of teacher-perceived friendship on teachers' assessments of their students' learning efforts. Based on earlier studies, we expect that teacher-perceived friendship differs from students' actual, self-reported friendship, and can account for teachers' judgement of students' learning efforts, above and beyond effects of students' self-reported study effort and friendship. Data were collected as part of the Networks and Actor Attributes in Early Adolescence study, which traces the development of individual behaviour, norms, and classroom social structure in 126 school classes (3332 students, 14 schools) during the first year at secondary school in the Netherlands. Students' individual study effort was assessed based on students' self-reports of four effort-related behaviours and as an overall judgement by the class teacher. Students' networks were assessed with the classroom as boundary, on the one hand as students' self-reported friendships, on the other hand as friendships perceived by the class teachers. Results were obtained with multivariate stochastic actor-based modelling of the feedback processes between study effort, study norms, and friendship networks.

Researchers have become increasingly interested in disentangling selection and influence processes. This literature review provides context for the special issue on network-behavior dynamics. It brings together important conceptual, methodological, and empirical contributions focusing on longitudinal social network modeling. First, an overview of mechanisms underlying selection and influence is given. After a description of the shortcomings of previous studies in this area, the stochastic actor-based model is sketched; this is used in this special issue to examine network-behavior dynamics. The preconditions for such analyses are discussed, as are common model specification issues. Next, recent empirical advances in research on adolescence are discussed, focusing on new insights into moderating effects, initiation of behaviors, time heterogeneity, mediation effects, and negative ties.

While in Euclidean, equiaffine or centroaffine differential geometry there exists a unique, distinguished normalization of a regular hypersurface immersion x: Mn → An+1, in the geometry of the general affine transformation group, there only exists a distinguished class of such normalizations, the class of relative normalizations. Thus, the appropriate invariants for speaking about affine hypersurfaces are invariants of the induced classes, e.g. the conformai class of induced metrics and the projective class of induced conormal connections. The aim of this paper is to study such invariants. As an application, we reformulate the fundamental theorem of affine differential geometry.

The main goal of this study was to examine differences between adolescent male and female friendship networks regarding smoking-based selection and influence processes using newly developed social network analysis methods that allow the current state of continuously changing friendship networks to act as a dynamic constraint for changes in smoking behaviour, while allowing current smoking behaviour to be simultaneously a dynamic constraint for changes in friendship networks.Longitudinal design with four measurements.Nine junior high schools in Finland.A total of 1163 adolescents (mean age = 13.6 years) who participated in the control group of the ESFA (European Smoking prevention Framework Approach) study, including 605 males and 558 females.Smoking behaviour of adolescents, parents, siblings and friendship ties.Smoking-based selection of friends was found in male as well as female networks. However, support for influence among friends was found only in female networks. Furthermore, females and males were both influenced by parental smoking behaviour.In Finnish adolescents, both male and female smokers tend to select other smokers as friends but it appears that only females are influenced to smoke by their peer group. This suggests that prevention campaigns targeting resisting peer pressure may be more effective in adolescent girls than boys.

Das sozialwissenschaftliche Interesse an dynamischen Netzwerkfragestellungen ergibt sich aus dem oft vorlaufigen Charakter von Analysen einzelner Messungen eines Netzwerks. Solche Querschnittsanalysen lassen im Allgemeinen wenig kausale Ruckschlusse zu und bieten daher lediglich ein eingeschranktes Verstandnis des Netzwerkes. So kann man mit herkommlichen (d.h., beschreibenden) Querschnittsverfahren beispielsweise den Zusammenhang zwischen strukturellen Eigenschaften der Akteure im Netzwerk und ihren individuellen Eigenschaften studieren. Beispiele hierfur finden sich uberall in der Netzwerkliteratur, zu denken ware etwa an den Zusammenhang zwischen strukturellen Lochern in Netzwerken von Managern und deren Leistung (Burt 1992), zwischen strukturell kohasiven Netzwerkregionen und Kooperation der Akteure (Gould 1993) oder zwischen Segregation von Netzwerken und Ethnizitat, Geschlecht oder Verhaltensdimensionen der Akteure (Moody 2001; Vermeij et al. 2009). Solche beschreibenden Analysen einzelner Netzwerke werfen aber vor allem die Frage auf, wie sich der gefundene Zusammenhang erklaren lasst. Oft gibt es verschiedene Erklarungen fur denselben Sachverhalt. Beim Beispiel der Theorie struktureller Locher ware die Frage zu klaren, ob tatsachlich die Fahigkeit eines Manager, unverbundene Netzwerkregionen zu vernetzen, zu seinem Berufserfolg beitragt – oder ob nicht vielmehr umgekehrt der Berufserfolg ihm erst eine Reputation zuteil werden lasst, aufgrund derer andere Netzwerkakteure ihn als Mittelsmann akzeptieren. Empirisch lassen sich diese Erklarungen nur dann voneinander unterscheiden, wenn man longitudinale Daten zur Verfugung hat. Wie das Beispiel verdeutlicht, sind es oft gerade die Prozesse, die zur Entstehung eines beobachteten Netzwerks fuhrten, die von inhaltlichem Interesse sind. In vielen Fallen lassen sich Netzwerkdaten daher als Momentaufnahme eines Evolutionsprozesses verstehen. Dem beschreibenden Schritt der Analyse einer solchen Momentaufnahme folgt dann auf naturliche Weise der im Verstandnis tiefer gehende Schritt der Analyse von generierender Netzwerkdynamik, den wir in diesem Kapitel behandeln wollen.