Investigating the Cycle of Violence: The Role of Attachment Style in the Relationship Between Child Maltreatment and Intimate Partner Violence

Purpose: Maltreated children are more likely than their peers to become involved in intimate partner violence (IPV) in young adulthood, but the mechanisms relating child maltreatment and IPV are not well understood. This study examines the relationship between the attachment styles of former foster youth who have been maltreated and IPV involvement in early adulthood. We adopt a person-centered approach (latent class analysis) that captures distinct profiles of the poly-victimization of youth in the Midwest Evaluation of the Adult Functioning of Former Foster Youth (Midwest Study). We then conduct a test of joint significance to examine the effects of maltreatment on IPV through attachment style.

Methods: Our sample included 551 23/24-year-old former foster youth living in three Midwestern states. Latent class analysis (LCA) models were fit using Mplus Version 7.  We used eight indicators of maltreatment that were included in the baseline survey to identify latent classes of maltreatment experience prior to foster care entry. We then conducted a test of joint significance to examine the effect of maltreatment class on IPV through attachment style (a test of indirect mediation). This test was conducted in two steps: (1) conducting a MANOVA to evaluate the extent to which maltreatment class influenced attachment style and (2) conducting a multinomial logistic regression to evaluate the influence of attachment style on IPV. This model was tested in the full sample then by gender.

Results: A three-class solution was selected as the best fitting model. 64.1% of the sample was placed in class 1. This class showed elevated levels of neglect and was called the “predominant neglect class” (PNC). 18.1% of the sample was placed in class 2. This class showed elevated levels of sexual maltreatment and was called the “sexual maltreatment class” (SMC). 17.8% of the sample was placed in class 3. This class showed elevated levels of neglect, physical maltreatment and sexual maltreatment and was called the “multiple maltreatment class” (MMC). Youth in the MMC had significantly higher avoidant attachment scores than youth in the PNC (p = .011). Further, youth with higher avoidant attachment scores were more likely to be in violent partnerships than non-violent partnerships (p = .016). Therefore both step (1) and step (2) showed significant effects; child maltreatment indirectly affected IPV involvement through attachment style. These findings only held when the mediation model was restricted to females (and the significance was marginal, p = .06).

Implications: Study findings enable a greater understanding of the active mechanisms linking child maltreatment and IPV involvement: the avoidant attachment styles held by multiply maltreated emancipated foster youth may influence the likelihood that they are involved in IPV. Despite the marginal significance, it was likely females who were driving the indirect mediation effect in this sample. Foster care administrators would be wise to target prevention and intervention efforts towards females who present with patterns of though, speech and behavior that could be characterized as “avoidant”.