Wekerle, C., Leung, E., Wall, A.M., MacMillan, H., Boyle, M., Trocme, N., & Wacechter, R. (2009). The contribution of childhood emotional abuse to teen dating violence among child protective services-involved youth. Child Abuse & Neglect, 33(1), 45-58.
Emotional abuse can be difficult to substantiate and may be considered inherent in more easily substantiated forms of maltreatment, such as physical or sexual abuse. Youth with a history of maltreatment are at heightened risk of dating violence as they learn to adapt to difficult circumstances (e.g., being hurt by loved ones, seeing maltreatment as a response to interpersonal problems). The authors theorize that emotional abuse will uniquely predict dating violence due to its negative impact on relationship development. Further, they predict that trauma will mediate or explain this association because of its link with the body’s stress response system. This study examined the rate of dating violence in youth involved with Child Protective Services (CPS), the unique effect of emotional maltreatment on the experience and perpetration of dating violence, and the potentially mediating effect of trauma.
A sample of 402 CPS-involved youth aged 14 to 17 from a large metropolitan area self-reported on measures of lifetime maltreatment, trauma symptoms, and previous year dating experiences. Most youth reported experiencing emotional maltreatment from parents (55% of males and 68% females). About 50% of youth had experienced moderate physical abuse and 30% reported more severe abuse. Physical abuse was most often perpetrated by the same-gender parent. Sexual abuse was reported by 30% of females and 8% of males. Significant trauma was reported by approximately 25% of males and females. Of the 85% of youth who reported dating (beginning, on average, at approximately age 13), victimization and perpetration of dating violence was endorsed by 49% and 44% of males and 63% and 67% of females.
A series of analyses were run to test the hypothesis that emotional maltreatment predicts dating violence (controlling for age, ethnicity, socioeconomic status, ward status, length of CPS involvement, and other maltreatment types) and to explore whether trauma symptoms might mediate the relationship between emotional abuse and dating violence. In preliminary statistical models, emotional maltreatment uniquely predicted dating violence victimization (for both males and females) and perpetration (for males only). However, in the final model in which trauma symptoms were added, emotional abuse was not found to predict dating violence; this indicates that trauma mediates, or explains, the relationship between emotional maltreatment and dating violence. This study demonstrates that dating violence may be common among youth involved with CPS and that emotional maltreatment, through its association with trauma, may be linked to future dating violence. Importantly, this study suggests that interventions that reduce trauma symptoms may also substantially reduce dating violence among CPS-involved youth.
The final sample used in this study may make the findings difficult to generalize to the broader population of CPS-involved youth. Three urban CPS agencies identified a random list of over 1500 potential participants; however, only 640 were eligible for the study and several hundred chose to not participate. Criteria that made youth ineligible included: out of age range, developmental delay, absent without leave, and deemed to be in crisis (i.e., suicidal, in detention, or in extended treatment). Participants were ethno-racially diverse, identifying as ‘Two or More Ethnicities’ (32%), White (29%), Black (25%), Latin American (3%), Native (1%), and Other (10%). The length of time youth were involved with CPS ranged from .25 to 17 years for an average of five to six years. Youth who were society and crown wards were overrepresented in the sample. Thus, the results may not generalize to youth who never or only briefly experience placement in out-of-home care. The final model accounted for over 30% of the variance in dating violence. The magnitude of variance in dating violence accounted for by this model is unusually large.