"Profiles of Children’s Prosocial Behaviors and its Longitudinal Associations with Peer Aggression, Bullying, and Victimization"
SRCD 2019 Biennial Conference
March 21-23, 2019
Title: Profiles of Children’s Prosocial Behaviors and its Longitudinal Associations with Peer Aggression, Bullying, and Victimization (Poster)
Session: Social Relationships
Authors: Ting-Lan Ma, Su Jiang, Sandi Simpkins, Deborah Vandell, Nicole Zarrett
Abstract: Relations between children’s prosocial behaviors and other aspects of their relationships with peers is a topic of considerable interest (Eisenberg, et al., 2015; Templeton, 2013). Children who are prosocial are less involved in bullying and aggression, as well as less likely to be victimized (Perren & Alsaker, 2005; Closson & Hymel, 2016). However, less is known about if subtypes exist in children’s prosocial behaviors, and whether these subtypes are associated with subsequent peer relationships in distinct ways. In this study, we identify distinct profiles of prosocial behaviors in middle childhood (4th grade) and examine how these profiles are associated with subsequent peer relationships in early adolescence (6th grade).
The longitudinal data were part of the NICHD Study of Early Child Care and Youth Development, a birth cohort of 1340 children recruited at 10 sites in the U.S.. Teacher reports of nine prosocial behaviors were collected at 4th grade (n=914) using the Child Behavior Scale (Ladd & Profilet, 1996). Peer relationships 6th grade were measured by both self-reported bullying and victimization (Kochenderfer & Ladd, 1996), and teacher-reported social exclusion, general aggression, relational aggression and peer victimization using CBS.
To examine the latent profiles emerged from the nine prosocial behaviors, we conducted Latent profile analysis in Mplus 8.0 using full information maximum likelihood estimation with robust SEs. To determine the best solution of number of classes, we examined multiple fit indices including the BIC, AIC, CAIC, SABIC, LMR-aLRT, and cmP which demonstrated the four-class solution as the best fit. The four-class pattern consisted of 67% (n=612) of children rated as highly prosocial (labeled high prosocial), 7% (n=73) who are rated as highly kind but less friendly (labeled kind only), 8% (n=73) were rated as highly friendly but less kind (labeled friendly only), and 18% (n=165) were rated as low in all prosocial behaviors (labeled low prosocial).
To examine whether class membership is associated with subsequent peer relationships at 6th grade, we conducted a series of ANCOVA controlling for gender to compare the mean differences in self-reported and teacher-reported aggression, bullying, and victimization across the class membership obtained at 4th grade. The results showed that children in the low prosocial group were more likely to be victimized and being aggressive, compared to the other three groups. Friendly only children were reported by teachers to exhibit higher relational aggression compared to high prosocial children. Friendly only children were less likely to be socially excluded and to bully others than low prosocial children. Kind only children were not significantly different from low prosocial children in the likelihood of being excluded or bullying others.
Our results suggest several different patterns of prosocial behavior, which are differentially related to children’s later relationships with peers. Although children who were consistently low on indicators of prosocial behaviors in middle childhood had the most problematic relationships with peers in early adolescence, children who were friendly (but not kind) and children who were kind (but not friendly) also displayed some problematic peer relationships