The Development of Adolescents’ Friendships and Antipathies:A Longitudinal Multivariate Network Test of Balance Theory

Peer relationships constitute an important context for the social and emotional development of adolescents (Furman & Rose, 2015). Most attention has been drawn to the emergence and maintenance of positive relationships (Veenstra et al., 2013). However, adolescents can also be tied negatively to peers. It is likely that positive ties (e.g., friendships) and negative ties (e.g., dislike) are interrelated, affecting each other over time (Berger & Dijkstra, 2013). The current study was aimed at examining the simultaneous development of friendship and antipathy networks and their interplay. We used (structural) balance theory (Cartwright & Harary, 1956; Heider, 1958) to understand the interplay between adolescents’ friendship and antipathy networks. According to the theory, relationship constructs are balanced when they are characterized by three positive ties or by two negative ties and one positive, resulting in stable relationships among the three individuals in the group (see Fig. 1-4) Based on these principles, it was expected that friendships would be formed or maintained when two adolescents disliked the same person (shared enemy hypothesis), that friends would tend to agree on whom they disliked (friends’ agreement hypothesis), that adolescents would tend to dislike the friends of those they disliked (reinforced animosity hypothesis), and, finally, that they would become or stay friends with dislikes of dislikes (enemy’s enemy hypothesis). Data came from two US middle schools (Grades 6 to 8; 11-14 years) participating in a larger longitudinal study on the social and academic development of youth. 480 students (NSchool1 = 220; NSchool2 = 260) participated in all three grades and were included in the current analyses. Friendship and antipathy networks were constructed based on peer nominations (Who is your best friend?; Who do you like the least?). Peer nominations were also used to assess prosocial behavior and antisocial behavior (Who cooperates, shares, and helps others?; Who starts fights, say mean things, and teases others?). A longitudinal multivariate network approach was used (i.e., RSiena; Snijders et al., 2010), which allowed examining the development of adolescents’ friendship and antipathy networks simultaneously over time, while controlling for individual dispositions (i.e., gender and ethnicity), network characteristics (e.g., density and reciprocity), and between-networks effects (i.e., interplay between both network types). Support was found for the first three hypotheses, and partially for the fourth hypothesis (Table 1, Fig.1-4). Taken together, the findings highlight that friendships are formed and maintained when adolescents dislike the same person; that adolescents agree with their friends on whom to befriend and dislike; and finally, that adolescents disagree with their antipathies on whom to befriend and dislike. The present findings show that the use of multivariate networks in combination with the predictions of balance theory can be fruitful for our understanding of the formation and maintenance of positive and negative relationships among adolescents. Future researchers, particularly investigating behaviors that are inherently relational such as bullying, may benefit from incorporating mechanisms of triadic interdependence. By more comprehensively mapping the adolescent peer ecology, we may be better able to reduce negative peer relationships and promote positive peer relationships among adolescents