How Statistical and Legal Counting Rules Shape Cross-National Comparisons of Homicide

• In Event: Trends in lethal and non-lethal violence

Abstract

Cross-national crime comparisons often rely on police statistics. Most commonly, such comparisons focus on homicide, as it is expected to be the crime type least likely to go undetected. Here, we examine how different statistical and legal counting rules employed across European countries may still affect the reliability of cross-national crime comparisons. We accessed data on homicide from 41 European countries (1998–2022) from Eurostat and compared it with three independent sources of vital statistics recorded by the World Health Organization and the Institute for Health Metrics and Evaluation. Correspondence rates between police-recorded homicides and vital records were assessed descriptively and graphically to identify cross-national and temporal variations. Two sets of within-between models were estimated to explore how statistical and legal counting rules affect cross-national homicide comparisons. Statistical and legal counting rules for homicide vary widely across countries, affecting the likelihood of homicides being recorded in police statistics and distorting cross-national comparisons. Countries recording data when crimes are first reported to the police often show higher homicide records than those using process-based or output-based systems. Additionally, more expansive definitions of homicide (e.g., including terrorism in measures of homicide) correlate with higher homicide records. National counting rules significantly affect the reliability of cross-national homicide comparisons relying on police data. Cross-national comparisons of other world regions with less standardized counting rules and legal frameworks, or for crime types more prone to under-recording, are likely to face this problem, posing a major challenge for comparative criminological research.

Authors

• Jose Pina-Sánchez, University of Leeds

• David Buil-Gil, The University of Manchester

• Marcelo F. Aebi, University of Lausanne