From Snapshot to Full-length Film: the "Living Environment and Security" Survey in the Victimization Data Series

Victimization surveys are most valuable if one can compile their data into the longest possible series. This goal is pursued by all countries that conduct such surveys. As an illustration, our article studies the linkage between the new "Living Environment and Security" Surveys (Cadre de Vie et Sécurité: CVS) and pre-existing nationwide victimization data in France, including from surveys by the Centre for Sociological Research on Law and Penal Institutions (Centre de Recherches Sociologiques sur le Droit et les Institutions Pénales: CESDIP) and by INSEE on "Household Living Conditions" (Enquête Permanente sur les Conditions de Vie des Ménages: EPCVM). The linkage is feasible despite several difficulties and a number of unresolved issues. The exercise confirms the continued decline in property crime. For personal crime, the figures show stagnation, at a modest level, in physical assault and a steep rise in verbal assault. We compared the victimization surveys with police statistics. These consist of series compiled by the Central Directorate for Criminal Investigations (Direction Centrale de la Police Judiciaire: DCPJ) from Gendarmerie and police reports on crimes and offences. The comparison shows a certain consistency--at least in trends--for property crime, but not for personal crime. In absolute-value terms, the police data, in our view, overestimate serious violence and underestimate milder violent crime. The divergence between the two sources is probably due to three factors: (1) the statistical effects of legislative changes that have broadened the scope of definition of assault and battery; (2) victims' low propensity to report less violent offences; (3) the exclusion of the most benign offences--traffic fines--from the scope of police statistics.