Overkill, we know it when we see it: examining definitions of excessive injury in homicide research

The purpose of this paper is to examine how the term “overkill” is used in the homicide literature to identify definitional issues that may interfere with reliable data coding across studies. This preliminary examination of the concept can guide future studies seeking to develop a standard definition.

To identify issues inherent in the term “overkill,” three definitions – ranging from broad and unclear to more specific and objective – were extracted or adapted from the existing literature. Using closed, homicide case files, nine coders were tasked with coding for the presence of overkill according to one of the definitions across two rounds of coding. Definitional components that made the coding of overkill difficult were identified using a qualitative sorting task to separate items into themes that represented similar issues; basic inter-rater agreement patterns were examined using pairwise percent agreement.

Based on coder feedback, two problems were identified: conceptual issues with the definitions and logistical issues with coding. However, feedback also suggested that increasing the objectivity of the overkill definition led coders to feel the intended meaning of the term was lost. Two out of three groups showed an increase in coder agreement between the two phases of data collection, illustrating how increased training is useful in certain situations.

This study is the first in-depth methodological and empirical examination of how the term “overkill” has been operationalized in the literature, raises key questions that may help with more clearly coding this variable, and outlines issues that may add difficulty to the development of a standard definition.