UCLA

A three year longitudinal study was conducted to deconstruct the temporal directionality between peer victimization and depressive symptoms in middle school youth. Analyses tested whether peer victimization predicted depressive symptoms; depressive symptoms predicted peer victimization; or the two constructs transacted reciprocally across time. To extend previous research, these relationships were simultaneously tested in one model that used a structural equation framework to specifically delineate the directional relations. The moderating effects of gender and ethnic differences on these relationships were also investigated. The sample was comprised of 2307 ethnically diverse students (54% female) followed from 6th through 8th grade and drawn from 11 middle schools. The ethnic breakdown was 44% Latino, 26% African-American, 10% Asian and 10% White. Students completed self-report surveys at six time points, corresponding to the six semesters constituting 6th through 8th grade. The Children’s Depression Inventory – Short Form (CDI-10) was used to assess self-reported depressive symptoms and the Peer Victimization Scale (PVS) to asses self-reported victimization The results showed that depressive symptoms were a stronger and more continuous predictor of victimization than was victimization a predictor of depressive symptoms for both genders, all ethnic groups, and across most of middle school. Gender effects indicated that boys reported fewer depressive symptoms than did girls, but the predictive impact of depressive symptoms on victimization was notably higher for boys, even as much as one year later. Ethnic group differences indicted that depressive symptoms consistently predicted changes in victimization throughout most of middle school for African-American and Latino students, but not as consistently for Asian and White students. Analyses further revealed that in girls and Latino students, victimization mediated the association between depressive symptoms from 6th to 7th grade and 7th to 8th grade, revealing a feedback mechanism between the two constructs. The opposing reciprocal association was also found to be true for Latino students. This study clearly elucidates that contrary to ongoing beliefs that victimization leads to mental health issues, depressive symptoms may precede and increase risk for victimization, and, in some groups, may lead to a feedback mechanism. The implications of these findings are that among other efforts to reduce victimization, interventions must be implemented to reduce depressive symptoms in youth that may make them more vulnerable to victimization by others. Recommendations for intervention strategies and future directions for research are discussed.