Police Responses to Cases of Officer-Involved Domestic Violence: The Effects of a Brief Web-Based Training

Background and Purpose: Little is known about domestic violence perpetration by law enforcement officers or “officer-involved domestic violence” (OIDV). The purpose of this study was to examine officers' responses before and after participation in the National Prevention Toolkit on Officer-Involved Domestic Violence (“Toolkit”), a brief, web-based training focused on warning signs and recommended responses to OIDV. Methods: We recruited potential participants through professional listservs and organizations, as well as through presentations at national and international conferences. Using a pre-post design, officers were provided two case scenarios of OIDV and asked to indicate how likely they were to take particular actions on a scale from 0% to 100% likelihood. Officers were also asked several questions regarding the training's impact on family safety. A factor analysis revealed potential actions formed five reliable subscales: assisting a victim in finding help, referring the perpetrator for help, recommending couples counseling, telling the couple to separate, and arresting the perpetrator. Subgroup analyses were also used to compare change scores across demographic and professional groups. Results: Only officers who answered 90% or more of the questions were included in the analyses (N = 852). Officers were from all regions of the U.S., but primarily from the Southeast (87%). Most officers were male (86.8 %) and Caucasian (82.2 %), though 11% identified as Black/African American. One-fourth (25.4%) held supervisory positions. Change scores with moderate to large effects included: 1) Assisting victim with obtaining help, rising from 41.4% to 74.9% average likelihood (d=.90), 2) Recommending perpetrator see department's chaplain, rising from 38.5% to 53.5% average likelihood (d=.67), and 3) Arresting perpetrator, rising from 26% to 47% average likelihood (d=.55). In addition, officers reported an increased likelihood of recommending couple's counseling post-training (55.5% to 62.9%, d=19). Few significant differences emerged among subgroups (10 of 121 tests). Non-supervisors had larger, positive change scores than supervisors for reporting the perpetrator to a supervisor and to internal affairs. Female officers had a larger increase change score for arresting the perpetrator when compared to male officers. Older officers increased their likelihood of assisting the victim more than did younger officers, although increases for both groups were significant. Officers also indicated the Toolkit would influence them to intervene in suspected cases of OIDV. Conclusions and Implications: Results showed that after training officers increased their likelihood of choosing victim-supportive behaviors, recommending help for the perpetrator, and formally reporting him. Although these responses were apparently intended to be helpful, future trainings must further emphasize limitations associated with helping professionals who are often not specially trained in domestic violence (e.g., chaplains). Future trainings must also clarify that couples counseling is most appropriate for less severe cases and must focus explicitly upon ending the violence (Stith, McCollum, & Rosen, 2011). In addition to resources such as the Toolkit, supervisors seem well-situated to provide training on OIDV. Domestic violence advocates are also encouraged to provide training to officers regarding OIDV and social workers, specifically, are urged to lead efforts to develop and reform training materials and policies related to OIDV.