Interpersonal Victimization of Latino Youth: A Latent Class Analysis

Background and Purpose: The current study identifies latent classes among Latino youth by using data from a national sample of Latino youth 12-18 years old. The data offers unique benefits: 1) a particular focus on Latinos which allows examination of an understudied group, 2) inclusion of 23 violent acts spanning physical, sexual, psychological, and stalking violence, and 3) identification of perpetrators for each of these acts within the categories of dating partners, peers, youth family members, adult family members, and strangers. Latent class analysis is a useful person-centered approach for examining victimization profiles. We then compare these latent classes on demographic characteristics, mental health (depression, anxiety, and hostility), and delinquency.

Methods: We utilized data from the Dating Violence Among Latino Adolescents (DAVILA) study which surveyed 1,525 Latino teens and their caregivers across the USA by phone. Participants completed modified versions of the Juvenile Victimization Questionnaire and the Conflict Tactics Scales 2- Short Form. Nine variables were included in the latent class analysis which included victimization type (physical, sexual, stalking, and psychological) as well as relationship to perpetrator (dating partner, adult non-family member, peer non-family member, juvenile family member, and adult family member). The Frequency of Delinquent Behavior was used to measure self-reported delinquent behavior in the previous year and the Brief Symptom Inventory (BSI) scale assessed psychiatric symptoms in adolescents.

Results: A six-class latent class solution was found. The six classes were I) Multiform Victimization by Multiple Perpetrators (n = 184, 12.1%), II) Multiform Dating Violence (n = 99, 6.5%), III) Multiform Victimization by Youth & Adults (n = 236, 15.5%), IV) Physical and Stalking victimization by Non-Family, especially Peers (n = 127, 8.3%), V) Physical Violence Victimization by Juvenile Family Members (n = 93, 6.1%), and VI) Uninvolved (n = 786; 51.5%). Classes differed on some demographic variables, hostility scores, and the rate of delinquency. Using the Uninvolved class as reference, youth in classes I, II, III, and V were more likely to have higher hostility scores. Using the Uninvolved class as reference, youth in classes I, II, III, and IV were more likely to engage in physical delinquency compare to the involved classes. Youth in classes I, II and III were more likely to engage in drug delinquency compared to the Uninvolved class. Lastly, those in class I were more likely to engage in property delinquency than those in the Uninvolved class.

Conclusions and Implications: Our findings provide further evidence regarding the heterogeneity of victimization experiences among Latino youth. None of the classes centered on one victimization type or one perpetrator. LCA results also suggest that victimization occurs across a range of perpetrators, both inside and outside of the home. Three classes included a range of family and non-family perpetrators. Hostility and delinquency were central to differentiating the classes, indicating their relevance among polyvictimized Latino youth. This analysis provides further understanding of the various ways Latino youth experience victimization and what factors may differentiate the various groupings of victimization profiles.