A prospective investigation of peer-victimization and loneliness in best-friend dyads : the effects of gender and friendship stability

Fox, Claire and Hunter, Simon and Jones, Di Sian (2017) A prospective investigation of peer-victimization and loneliness in best-friend dyads : the effects of gender and friendship stability. In: World Anti-Bullying Forum 2017, 2017-05-07 - 2017-05-09, Quality Hotel Friends. (http://www.wabf2017.com/wp-content/uploads/2017/04...)

[thumbnail of Fox-etal-WABF-2017-Peer-victimization-and-loneliness-in-best-friend-dyads]
Preview
Text. Filename: Fox_etal_WABF_2017_Peer_victimization_and_loneliness_in_best_friend_dyads.pdf
Final Published Version

Download (801kB)| Preview

Abstract

Associations have been identified between peer-victimization and loneliness. However, little is known about the ways in which these associations operate within best-friend dyads. Indeed, friendships are recognised as important relational contexts for young people’s development. A shortterm prospective design and the Actor-Partner Independence Model (APIM: Cook & Kenny, 2005) was used to model the extent to which children’s loneliness is influenced by (i) their own experiences of peer-victimization and (ii) their best friend’s experience of peer- victimization. Participants were 1,234 young people aged 11-13 years, completing a battery of self-report measures (including peervictimization, loneliness, and depression) on two occasions approximately 6 months apart. From this group, 443 reciprocal best friend dyads were identified at T1 (i.e. pairs of children who chose one another as their ‘best friend’) and of those, 87 were dyads who remained reciprocal best friends at Time 2. Cross-sectional results indicated that loneliness and peer-victimization were positively correlated at the individual level (r = .35, p < .001). In addition, levels of best friends’ loneliness were also positively correlated (r = .19, p < .001), as were their levels of peer-victimization (r = .13, p < .01). Turning to the longitudinal data, the APIM model allowed us to control for earlier levels of loneliness as well as earlier depression scores, and we investigated whether gender or stability of friendship moderated any of the paths of interest. The analyses revealed a small effect of peervictimization on later loneliness, but only among girls. In addition, there was a medium size effect indicating that a best friend’s levels of peer-victimization were positively associated with young people’s own later loneliness. It was also interesting to note that a best friend’s loneliness was positively associated with young people’s own later loneliness, with this effect most obvious among stable best friend dyads. These results indicate that, even after controlling for earlier levels of loneliness and depression, girls’ self- reported loneliness is associated with the degree of peer-victimization that their bestfriend reports. The implications this has for future avenues of research will be discussed. In addition, the finding that the loneliness of a best-friend can, for stable best-friend dyads, impact on a partner’s level of loneliness implies that these friendships should not be viewed as a panacea for young people’s problems. The potential for negative outcomes on feelings which we would expect bestfriendships to reduce (loneliness) should be investigated further.