Poster #227 - Sex differences in delinquency selection similarity during the pre-adolescent years

• In Event: 1-061 - Poster Session 02
  In Poster Session: PS 02 Section - Social Relationships

Integrative Statement

Adolescents closely resemble their friends. (e.g., Linden-Andersen et al., 2009), a fact that may be of particular importance given the prevalence of problem behaviors at this life stage (Allen, Moore, & Kuperminc, 1997). Current literature suggests that similarity in friendships may be most evident in aggression, deviancy, and delinquent behaviors; socialization processes have been credited as major contributors to such similarity (e.g., Brechwald & Prinstein, 2011). However, it is likely that both selection and influence processes play a role (DeLay et al., 2016). Effects of peer influence and selection on delinquency similarity are inconsistent (Knecht et al., 2010). It is unclear to what extent selection would affect delinquency similarity. Delinquency is more prevalent among boys, but gender does not differ as to the peer influence on those behaviors (Laursen & Hafen, 2012). Age-related change demonstrates that delinquent behaviors rise in pre-adolescence and peak in mid-adolescence (Richmond, Kerr, Stattin, & Laursen, 2018). The current study explored the selection effects on delinquency, specifically whether adolescents already resembled each other on delinquency one year before they became friends, and whether age and gender moderated that prior similarity.
Data was drawn from a 5-year longitudinal study that examined a community sample of 1805 adolescents attending public schools in Sweden. For this study, we focused on 5th to 10th grade adolescents (M=11.76, SD=1.623) who participated at two consecutive years. Friendship dyads were created by pairing adolescents with their same-age, same-sex reciprocated friends at the second wave of data. Non-friend dyads were created by pairing children with three randomly assigned peers (matched on sex and age). A total of 536 friend dyads and 3006 dyads of nonfriends were created. Dyadic similarity in delinquency behavior was examined a year before adolescents became friends (see Table 1).
Two-level hierarchical linear modeling (HLM) was conducted to explore whether similarities in delinquency predicted friendship reciprocity one year later. At level 1, within-dyad delinquency similarity was examined. At level 2, age and gender were included as within-person variables. Results of the HLM analyses are reported in Table 2.
Our results show that younger dyads are more similar on delinquency than older adolescents. Gender moderated the association between delinquency similarity and friendship. Boys selected friends who were more similar in delinquency than did girls. However, age did not affect the association between delinquency similarity and friendship. Delinquency similarity was equally important across age groups in the selection of friendships.
Our findings suggest that similarities in delinquent behaviors may play a more important role in boys’ friend selection process than in girls’. This makes sense based on previous research on gender difference in delinquency (e.g., Mears, Ploeger, & Warr, 1998). Boys are more apt to be involved in delinquent behavior, and therefore, it is more salient to the selection and establishment of friends for boys than girls. Age, in turn, may not matter.

Authors

• Li Wei, Florida Atlantic University
  Presenting Author

• Olivia Maria Valdes, Florida Atlantic University
  Non-Presenting Author

• Amanda Gaudree, Florida Atlantic University
  Non-Presenting Author

• Brett Laursen, Florida Atlantic University
  Non-Presenting Author

• William M. Bukowski, Concordia University
  Non-Presenting Author