Prospects, Problems, and Pitfalls in Comparative Analyses of Criminal Justice Data

Abstract

Official crime and criminal justice data are influenced by different substantive (e.g., victims’ reporting rates), legal (e.g., offense definitions), and statistical (e.g., counting rules) factors. This complicates international comparison. The UN Crime Trends Survey, Eurostat’s crime statistics, and the European Sourcebook of Crime and Criminal Justice Statistics try to enhance comparability and document remaining differences. The UN survey and Eurostat rely on the International Classification of Crimes for Statistical Purposes, which has potential but is not yet satisfactorily applied. The European Sourcebook provides the most detailed and best-verified data among the three. Even standardized data need to be compared with extreme caution. Crime levels are not a valid measure of crime in different countries, with the possible exception of completed intentional homicide. Total crime rates depend mainly on the internationally differing quality of police work. Comparisons of crime trends are less problematic but depend on the offenses under comparison being not defined too differently. Indicators expressed as ratios of different system-based values have increased comparability. Owing to immense differences in crime rates and criminal justice variables, mean crime rates for the world or Europe cannot be calculated. Country clusters need to be built very carefully.