homicide situations involve both structure and process. Their structures are defined by the particular combinations of offender, victim, and offense attributes that provide the physical and social context for the homicide. Their process involves the interpersonal and transactional dynamics that ultimately determine whether a homicide occurs within these particular contexts. Through the use of a comparative method (QCA) and actual narrative accounts of homicides, this study was designed to identify the unique and common structures and processes that underlie homicide situations across time and different subgroups.

This final chapter summarizes our major findings about homicide situations and their structural and processual features. After reviewing these findings, we discuss their implications for (1) dispelling various myths and misconceptions about homicide; (2)theoretical developments; and (3) public policy issues and violence intervention efforts. We conclude with a discussion of the limitations of the current study and new directions for future research.

Major Findings

Our analysis of homicide situations reveals the following major findings about the nature of homicide situations and subgroup variation in them over time:

• Homicide situations in the United States from 1976 to 1998 are empirically concentrated among a small subset of possible structures. There are 32,768 possible structures of homicide situations involving the simultaneous classification of 15 dummy variables (215), but only 618 of these structural contexts are found to represent at least 5 homicides per year. These 618 distinct homicide situations account for 82% of all homicides from 1976 to 1998. This pattern of situational clustering (i.e., few configurations account for a large proportion of all cases) holds true for each decade.