Behind the 911 calls: understanding arrest predictors for domestic violence in Kansas City, Missouri using arrest data

Abstract

Domestic violence (DV) is a global health epidemic that has a wide-reaching negative impact on many individuals. This phenomenon is also highly complex. Police are often the first point of contact for many DV victims and offenders. This makes police response to DV an important factor to study and understand. The primary research question for this research project is the following: “what is the impact of victim, offender, and situational characteristics on police arrest decisions?” The data being utilized are Kansas City, Missouri’s (KCMO) police arrest data from July 2019 to July 2023. A logistical regression analysis was performed to analyze the relationship between the independent variables and arrest outcome. This analysis found that victim sex, type of relationship, uniform crime reporting (UCR) offense code, and arrestee weapon use were the only independent variables found to be statistically significant with the dependent variable. The project concludes that it is difficult to gain a deeper understanding of police arrest decision-making in DV cases using quantitative data alone. It is suggested that future research be conducted using quantitative arrest data alongside qualitative data. This ensures that demographic characteristics of the victim/offender and situational characteristics of the incident are included in the same study as contextual factors of police decision making that can only be adequately understood through qualitative methods.