Poster #112 - Investigating the Role of Friends in the Developmental Trajectory of Maltreated Adolescents

• In Event: Poster Session 17

Abstract

Despite extensive evidence that friendships play an integral role in adolescent development (Prinstein & Giletta, 2016), we know relatively little about the friendships of maltreated adolescents, a group that is at significantly elevated risk for experiencing future adverse outcomes (Cicchetti & Toth, 2016). The existing literature regarding the influence of friendships on maltreated youth’s development is highly conflicting (Merritt & Snyder, 2015) and limited by investigations that do not account for the full complexity of friendship experiences. Thus, there is a critical need to study how maltreated adolescents’ friendships might mitigate or exacerbate their risk for proceeding on negative developmental trajectories.

The study aims were to (1) investigate how the friendship experiences of maltreated and non-maltreated adolescents differ and (2) tests for the potential mediating effects of friendship experiences on the association between maltreatment and psychopathology/peer victimization (see Figure 1) with a plan to examine an alternative hypothesis that friendship experiences exhibit moderating effects. This investigation utilized data from two prospective longitudinal studies of maltreated and non-maltreated adolescents (N = 514 for the Female Adolescent Development Study [FADS) and N = 1,354 for the Longitudinal Studies of Child Abuse and Neglect [LONGSCAN]). Maltreatment was assessed using Child Protective Service records. Other variables were assessed via adolescent self-report. Data sets were analyzed separately. Regression was used to test for differences in friendship experiences and structural equation modeling to test for mediating and moderating effects.

Preliminary findings revealed that in LONGSCAN, maltreatment significantly predicted greater engagement with close friends involved in risky behaviors (β = .08; p = .03), but no differences in other friendship experiences. Notably, experiencing child sexual abuse (CSA) uniquely predicted shorter friendships (β = .07; p = .04). FADS analyses revealed that maltreated adolescents reported less alienation in their friendships (β = -.10; p = .03) and having less friends (β = -.18; p < .001). CSA uniquely predicted having lower quality friendships (β = -.15; p = .001) in FADS. None of the friendship experience variables exhibited a mediating effect on the associations between maltreatment and psychopathology/peer victimization in either dataset. However, in FADS, overall friendship quality significantly mediated the association between CSA and total internalizing symptoms (standardized indirect effect estimate = .03; 95% non-standardized CI [0.18, 1.29]; p = .009). Exploration of the alternative hypothesis testing for moderating effects demonstrated that in LONGSCAN, maltreatment predicted more internalizing/externalizing symptoms and peer victimization for adolescents with low, but not high levels of positive friendship qualities (see Figure 2) and that maltreatment predicted more post-traumatic stress symptoms for adolescents engaged with peers involved in high, but not low levels of risk behaviors. These findings emphasize that it would be beneficial for practitioners to work with maltreated adolescents to establish high quality friendships with friends who are not engaged in risky behaviors to protect them from experiencing psychopathology and peer victimization. These findings also indicate that sexually abused adolescents in particular may have more pronounced difficulties with establishing meaningful friendships, which has direct implications for their mental health.

Authors

• Michelle P Brown, University of South Carolina
  Presenting Author

• Jennie G. Noll, Pennsylvania State University, University Park