Targeted Victimization: Exploring Linear and Curvilinear Associations Between Social Network Prestige and Victimization

Subjective well-being of children with special educational needs: Longitudinal predictors using machine learning

Children with special educational needs (SEN) are a diverse group facing numerous challenges related to well-being and mental health. Understanding the predictors of well-being in this population requires the incorporation of diverse factors along with approaches that can uncover complexity in how these factors work together to influence well-being. We longitudinally predicted subjective well-being in a group of children with diverse special educational needs (N = 499; M = 8.4 ± 0.9 years). Thirty-two variables - ranging from demographics to various categories of life experiences - were used as predictors for both nonlinear machine learning and classical linear classifiers. Nonlinear machine learning classifiers exhibited much performance in predicting subjective well-being (F1 score = 0.72 to 0.84) compared to traditional linear classifiers. Overall, across all children, prior subjective well-being, numeracy, literacy skills, and interpersonal dimensions played important roles. However, clustering further identified four distinct clusters sharing important predictors: a 'socializer' cluster dominated by interpersonal functioning predictors, an 'analyzer' cluster emphasizing academic skills predictors, and two clusters with more diverse sets of important predictors. Our research highlights the multiple pathways toward well-being in children with SEN as uncovered by machine learning, with implications for understanding and supporting their well-being.

When strength is a weakness

In general, individuals with low levels of coercive power are at greater risk of victimization than those who are more powerful. However, in some circumstances, superior coercive power increases an individual's vulnerability. In this paper, I show how coercive power can increase vulnerability (offsetting its protective effect) by its effects on targeting and tactics. Individuals with greater coercive power can increase their risk of getting targeted because they tend to be less vigilant and more likely to behave in ways that provoke others. They generate more grievances and have more enemies because they are less compliant and more verbally aggressive and confrontational. Powerful parties are also at greater risk of being targeted by adversaries seeking to gain status. An attack on a strong adversary is more likely to enhance status than an attack on a weaker adversary since it is a greater achievement. Individuals with coercive power are also at greater risk because of the tactics used by their weaker adversaries. Weaker parties are more likely to engage in pre-emptive attack and use weapons. They are better able to attract and rely on allies because of the norm of social responsibility, that is, the tendency to protect those in need. Finally, they are more likely to attempt to kill more power adversaries in order to incapacitate them and, thereby, avoid retaliation.

Sleepy and Secluded: Sleep Disturbances are Associated With Connectedness in Early Adolescent Social Networks

Studies in adults suggest that sleep disturbances predict poorer socioemotional skills and impaired social interactions. However, little is known regarding how sleep disturbances are associated with social processes during adolescence, a period when both sleep neurobiology and social relationships are undergoing dramatic developmental changes. The current study examined associations among sleep disturbances and peer connectedness in a sample of middle-school students (N = 213, 11-15 years old, 57% female) using a social network approach. Findings suggested that youth with greater sleep disturbances reported having fewer social connections, were rated as a social connection by fewer peers, and were less likely to have reciprocated nominations, even after controlling for sociodemographic characteristics and mental health symptoms.

Early Adolescents Demonstrate Peer-Network Homophily in Political Attitudes and Values

Research on political homophily has almost exclusively focused on adults, and little is known about whether political homophily is present early in life when political attitudes are forming and friendship networks are rapidly changing. We examined political homophily using a social network approach with rural middle school students (N = 213; mean age = 12.5 years; 57% female) from a remote U.S. community. Preregistered analyses indicated that early adolescents were more likely to spend time with people who shared similar political attitudes and values. These effects were most consistent for right-wing authoritarianism, patriotism, and anti-immigration attitudes. Our results show that political homophily is evident at an early age when young people are forming their political beliefs and making decisions about their friendships, suggesting that peer political-attitude socialization may emerge early in life.

Predicting How Well Adolescents Get Along with Peers and Teachers: A Machine Learning Approach

How well adolescents get along with others such as peers and teachers is an important aspect of adolescent development. Current research on adolescent relationship with peers and teachers is limited by classical methods that lack explicit test of predictive performance and cannot efficiently discover complex associations with potential non-linearity and higher-order interactions among a large set of predictors. Here, a transparently reported machine learning approach is utilized to overcome these limitations in concurrently predicting how well adolescents perceive themselves to get along with peers and teachers. The predictors were 99 items from four instruments examining internalizing and externalizing psychopathology, sensation-seeking, peer pressure, and parent-child conflict. The sample consisted of 3232 adolescents (M = 14.0 years, SD = 1.0 year, 49% female). Nonlinear machine learning classifiers predicted with high performance adolescent relationship with peers and teachers unlike classical methods. Using model explainability analyses at the item level, results identified influential predictors related to somatic complaints and attention problems that interacted in nonlinear ways with internalizing behaviors. In many cases, these intrapersonal predictors outcompeted in predictive power many interpersonal predictors. Overall, the results suggest the need to cast a much wider net of variables for understanding and predicting adolescent relationships, and highlight the power of a data-driven machine learning approach with implications on a predictive science of adolescence research.

Power in Aggressor-Victim Relationships: Exploring Social, Physical, Gender- and Ethnicity-Based Power

Theory highlights power in aggressor-victim relationships, yet empirical work assessing dyadic power is largely absent. Variability in power balance versus imbalance within aggressor-victim dyads (based on social, physical, gender- and ethnicity-based power) was explored. Participants (N = 952; grade 6-8; 50% girls, 44% Hispanic/Latina/o) nominated aggressors and victims (4662 aggressor-victim dyads; 642 strong dyads [based on reputational strength]; 169 sustained dyads [based on longevity]). Dyadic social power (social network centrality and prestige) was calculated from friendship nominations. Self-report was used for dyadic physical (body mass index), gender- and ethnicity-based power. Across power indicators, there were more power-balanced than imbalanced dyads (particularly for strong and sustained dyads). The findings challenge theoretical notions that aggressors are more powerful than their victims and have implications for aggressor-victim relationships.

Peer likeability and victimization in young adolescents: Moderating effects of descriptive and status group norms for aggression and prosocial behaviour and network density

INTRODUCTION

Previous research has shown that peer victimization can be highly responsive to variables at the classroom level. Aggressive and prosocial norms may promote or reduce its status in classrooms. However, yet there is an apparent lack of success to explain which types of norms are more influential. This study examined the role of aggressive and prosocial descriptive and status norms in the peer victimization-status link. It also explores how the network density increases adherence to the prevailing norm in the classroom and its effect on the status of the victims.

METHOD

Data on peer acceptance and rejection, victimization, prosocial behaviour, and aggression were collected with sociometric methods in a sample of 6,600 students (M = 13.1 years, SD = 0.6; 49.2% girls), from 269 classrooms in 81 secondary schools in Spain. Group norms for aggression and for prosocial behaviour were assessed in three ways, the behaviour of all peers (class-norm), the behaviour of most-liked peers (likeability-norm), and the behaviour of most salient peers (visibility-norm).

RESULTS

Multilevel regression analyses revealed that the negative impact of victimization on peer likeability was moderated by the classroom's norm for prosocial behaviour, by the status norm of most visible peers' norm for prosocial behaviour and for aggression, and by the group's network density. The behavioural status norms of most likeable peers had no significant effect.

CONCLUSION

These results underscore the overall importance of group context as a moderating factor of the relation between victimization and peer status in adolescents, and add to the growing body of knowledge driven by the socio-ecological approach to the study of peer relations in developmental psychology. As implications for education, these results affect the importance of considering socio-emotional variables in the formation of class groups in order to reduce victimization.

Adolescent Victim Types Across the Popularity Status Hierarchy: Differences in Internalizing Symptoms

Previous studies have called attention to the fact that popular youth are not immune to peer victimization, suggesting there is heterogeneity in the popularity of victims. Yet, no study to date has determined whether victims with different levels of popularity status can be identified using person-oriented analysis. Such analysis is critically needed to confirm the existence of popular victims. Further, there remains a paucity of research on internalizing indices of such popular victims, especially compared to other victim and non-victim groups. To address this gap in the research literature, the current study used latent profile analysis to identify subgroups of victims based on victimization (self- and peer-report) and popularity (peer-report). This study sought to verify the existence of popular victims and to compare victim subgroups on loneliness and self-esteem. Participants were 804 Dutch adolescents (50.2% boys, M_age = 13.65 years, ranging from 11.29 to 16.75 years). The results revealed six subgroups, including a group of popular self-identified victims. Popular self-identified victims were generally less lonely than other victims, but had higher loneliness and lower self-esteem than non-victims. Implications are discussed for understanding the victimization experiences of high-status youth.

Civic Development within the Peer Context: Associations between Early Adolescent Social Connectedness and Civic Engagement

Social connectedness is theorized to contribute to civic development and in turn, civic engagement is thought to cultivate social connectedness. Little research has examined how social connectedness is linked with early adolescent civic engagement. The current study used a social network research design to examine associations between early adolescent social connectedness via their position within their peer network and their civic engagement. Middle-school students (N = 213) aged 11-15 years (M = 12.5; 57% female) were recruited from a remote US community and provided nominations for peer connections and reported on multiple aspects of civic engagement. Early adolescents who had identified more peer nominations had higher civic efficacy. Youth who had fewer connections with different peer groups and fewer connections with popular peers were more engaged in political behavior. Greater popularity was associated with higher political engagement for boys, but not girls. Greater connections with different peer groups was associated with greater environmentalism for younger but not older teens. These findings highlight the need to consider adolescent civic development within the peer context.

Bullying and Victimization Trajectories in the First Years of Secondary Education: Implications for Status and Affection

Bullying is known to be associated with social status, but it remains unclear how bullying involvement over time relates to social position (status and affection), especially in the first years at a new school. The aim of this study was to investigate whether (the development of) bullying and victimization was related to the attainment of status (perceived popularity) and affection (friendships, acceptance, rejection) in the first years of secondary education (six waves). Using longitudinal data spanning the first- and second year of secondary education of 824 adolescents (51.5% girls; M_age T1 = 12.54, SD = 0.45) in the SNARE-study, joint bullying and victimization trajectories were estimated using parallel Latent Class Growth Analysis (LCGA). The four trajectories (decreasing bully, stable high bully, decreasing victim, uninvolved) were related to adolescents’ social position using multigroup analysis that examined differences in slope and intercepts (T1 and T6) of social positions, and indicated that the relative social position of the different joint trajectories was determined at the start of secondary education and did not change over time, with one exception: adolescents continuing bullying were besides being popular also increasingly rejected over time. Although bullying is functional behavior that serves to optimize adolescents’ social position, anti-bullying interventions may account for the increasing lack of affection that may hinder bullies’ long-term social development.

Dyadic nominations of bullying: Comparing types of bullies and their victims

Previous studies indicate that when identifying individuals involved in bullying, the concordance between self- and peer- reports is low to moderate. There is support that self- and peer- identified victims constitute distinct types of victims and differ in adjustment. Likewise, differentiating between self- and peer- reports of bullying may also reveal distinct types of bullies. The goal of this study was to examine differences between types of bullies identified via dyadic nominations (self-identified, victim-identified, and self/victim identified). First, we examined the concordance between dyadic nominations of bullying and traditional measures of bullying (i.e., self- and peer-reports). Second, we compared the behavioral profiles of the bully types to nonbullies, with a focus on aggressive behaviors and social status. Third, we examined whether the types of bullies targeted victims with different levels of popularity, as well as the role of their own popularity and prioritizing of popularity. Participants were 1,008 Dutch adolescents (50.1% male, M_age  = 14.14 years, standard deviation [SD] = 1.30) who completed a classroom assessment of dyadic nominations, peer nominations, and self-report items. Results indicated that victim identified and self/victim identified bullies were more aggressive, more popular, and less socially preferred than self-identified bullies and nonbullies. Self/victim identified bullies targeted victims with the highest social status. The association between bully type and victims' popularity was further qualified by bullies' own popularity and the degree to which they prioritized popularity. Implications for the implementation of dyadic nominations are discussed.

Prestigious Youth are Leaders but Central Youth are Powerful: What Social Network Position Tells us About Peer Relationships

Measures of social network position provide unique social and relational information yet have not been used extensively by researchers who study peer relationships. This study explored two measures-social network prestige and social network centrality-to improve conceptualization of their similarities, differences, and meaning within a peer relationships context. Prestige and centrality were computed from friendship nominations (N = 396 6th graders; 48% girls; 49% White) and participants nominated peers on several social indicators (e.g., aggressive, popular). Two example classroom networks were examined to visually depict social network position. Associations between measures of social network position and social indicators were examined using correlations and latent profile analysis. Latent profile analysis identified three profiles based on the social indicators, which differentially related to prestige and centrality. Overall, prestigious youth were generally well-liked, prosocial, and leaders, whereas central youth were powerful and aggressive. The results strengthen the conceptualization of these network-based measures, allowing them to be more readily used by peer relationships researchers to understand youth's interaction patterns and behaviors.

Does an aggressor's target choice matter? Assessing change in the social network prestige of aggressive youth

Based on a social dominance approach, aggression is conceptualized as a strategy used to gain position, power, and influence within the peer network. However, aggression may only be beneficial when targeted against particular peers; both victims' social standing and the number of victims targeted may impact aggressors' social standing. The current study examined associations between aggressors' targeting tendencies (victims' social standing and number of victims) and aggressors' own social standing, both concurrently and over time. Analyses were conducted using three analytic samples of seventh and eighth grade aggressors (Ns ranged from 161 to 383, 49% girls; 50% Latina/o). Participants nominated their friends; nominations were used to calculate social network prestige. Peer nominations were used to identify aggressors and their victim(s). For each aggressor, number of victims and victims' social network prestige were assessed. Aggressors with more victims and with highly prestigious victims had higher social network prestige themselves, and they increased more in prestige over time than aggressors with fewer victims and less prestigious victims (though there were some differences across analytic samples). Findings have implications for the need to extend the social dominance approach to better address the links between aggressors and victims. Aggr. Behav. 43:364-374, 2017. © 2017 Wiley Periodicals, Inc.

Commentary: On the importance of looking at nonlinearity and developmental effects - a reflection on Flom et al. (2017)

By examining both linear and curvilinear associations between mental development and activity level, the study by Flom et al. (Journal of Child Psychology and Psychiatry, 2017) highlights the importance of going beyond linear associations in psychological fields of research. Results from Flom et al. (Journal of Child Psychology and Psychiatry, 2017) also raise interesting questions for future research. First, studies should look at variables that may explain the associations between activity level and mental development, such as self-regulation and attention. Second, longitudinal changes in the strength of the association between activity level and mental development should be examined to determine when this association is at its strongest. Finally, longitudinal research looking at bidirectional effects is needed to confirm the direction of the associations between activity level and mental development. Answers to these questions will allow the identification of the best targets and developmental periods for interventions to take place.

Reciprocal Associations between Delinquent Behavior and Social Network Position during Middle School

Delinquency and social standing are linked within middle school. Yet, theoretical explanations are generally unidirectional, and prevailing models are somewhat contradictory in terms of the directionality of the link between delinquency and social standing. The current study aimed to expand upon our current understanding of the social nature of delinquency by examining reciprocal associations between delinquency and social standing. We conceptualized social standing using two indices of social network position: social network prestige (how important or influential one is within the peer network) and social network centrality (how well-connected one is to peers in the network). We also assessed gender differences in associations. Ethnically diverse middle school students were followed longitudinally across one year (three waves; N = 516, M _age  = 11.91 years at the first wave; 47% girls; 55% Latina/o). Participants reported on their delinquent behavior and nominated friends within their grade; friendship nominations were used to calculate social network prestige and centrality. Results indicated that both indicators of social network position were associated with increases in delinquency across school years, but not within the school year. Further, delinquency was associated with increases in social network prestige but not social network centrality (again, only across school years). Similarities across gender were found. The findings highlight the need to expand upon current, generally unidirectional theories of the social nature of delinquency, and suggest important differences between change within vs. across the school year.