Unpacking Selection and Peer Influence Pathways in Bullying and Friendship Networks

• In Event: 3-100 - Peer Network Dynamics and Child Development: Appraising the Evidence and Charting New Directions

Integrative Statement

There is wide consensus among scholars that bullying is a group process (Salmivalli 2010). This idea stems from the participant roles approach (Salmivalli et al. 1996), which emphasizes the social nature of bullying and its strong relation with status in the group. Yet, research on network-behavior dynamics found no indication of selection and influence processes for bullying in friendship networks (Veenstra et al. 2013). These null results may have methodological underpinnings: treating bullying as individual behavor, as most previous studies did, neglects its relational component, namely who is targeted and by whom (Rodkin et al. 2015). Research that considered victimization in a relational framework found that bullies, who targeted the same victims, defended each other when their victims retaliated (Huitsing et al. 2014). This suggests that having a common victim may foster positive affect between individuals, such as a friendship (describing a selection process: Fig. 1: a → c). Friendships, in turn, are an important source of peer influence (Brechwald & Prinstein 2011): individuals more readibly conform to peers’ behaviors (and ties) through direct contact and interaction with them (describing an influence process: Fig. 1: b → c), particularly when behaviors are status-enhancing. In this study, we investigate to which extent selection and influence processes explain bullying relations by examining them together with friendship relations in childhood.
Classrooms were drawn from the first three waves of the KiVa study (May 2012, October 2012, and May 2013), a program aimed to reduce school bullying among children in elementary education (8-12 years) in the Netherlands (Veenstra 2015). For the current study, 19 classrooms (481 students) containing friendship relations (“Who is your friend?”) and bullying relations (“Who do you bully?”) are available. We analyzed each classroom separately with longitudinal bivariate social network analysis using RSiena (Snijders et al. 2013) and summarized the results in a meta-analysis (Viechtbauer, 2010).
We tested the dependence of one type of network on another type by including two mixed triadic multiplex effects (and their counterparts), controlling for dyadic and degree-related multiplex effects to account for main effects of one network type on another and dynamics in in- and out-degrees between network types. In support of the bully selection hypothesis (Fig. 1: a → c), the bullying the same victim to friendship effect positive and significant: when two children bullied the same victim they tended to become friends (Table 1: Est. = 0.41, p < .05). In addition, in support of the bully influence hypothesis (Fig. 1: b → c), the friend bullying to bullying the same victim effect was positive and significant: children started to bully the classmates who were bullied by their friends (Table 1: Est. = 0.74, p < .001). This study provides evidence for group processes in bully-victim networks in childhood. One the one hand, bullies select each other as friends; on the other hand, bullies influence their friends to start bullying. These novel findings highlight how group processes in bullying operate through children’s peer networks in the classroom.

Authors

• Ashwin Rambaran, University of Groningen
  Presenting Author

• Jan Kornelis Dijkstra, University of Groningen
  Non-Presenting Author

• Rene Veenstra, University of Groningen
  Non-Presenting Author