The Effect of Counting Rules on Cross-National Comparisons of Homicide

Abstract Cross-national crime comparisons often rely on police-recorded data. Most frequently, such comparisons focus on homicide, as it is expected to be the crime type least likely to go undetected. This study examines how different statistical counting rules and legal definitions employed across European countries affect the reliability of cross-national homicide comparisons. We analyze homicide data from 41 European countries (1998-2022) obtained from Eurostat and compare them with three sources of vital statistics from the World Health Organization and the Institute for Health Metrics and Evaluation. We assess correspondence rates between police-recorded homicides and vital records descriptively and graphically to identify cross-national and temporal variations. Additionally, we estimate within-between models to quantify the impact of statistical counting rules and legal definitions on cross-national homicide comparisons. Statistical counting rules and legal definitions for homicide vary widely across countries, shaping the likelihood of homicides being recorded in police statistics and compromising cross-national comparability. Countries that record data when crimes are first reported to the police have approximately 13-15% higher homicide counts than those using process-based or output-based systems. Additionally, broader definitions of homicide (e.g., those including terrorism-related deaths) are associated with higher recorded homicide. National counting rules and legal definitions substantially impact the reliability of cross-national homicide comparisons based on police data. This challenge is likely even greater in regions with less standardized counting rules and legal frameworks or for crime types more susceptible to under-recording, posing a significant challenge for comparative criminological research, international policy benchmarking, and resource allocation decisions.