Since the beginnings of criminal statistics in the early nineteenth century their use as instruments for measuring the amount of crime has been controversial. Nevertheless, they are still important to the present day. These data are used to justify policies, to decide on measures of crime prevention and control and to make statements about the performance of the criminal justice system. Despite all methodological doubts, these figures are still referred to in criminology, as there are no real alternatives to them. Even crime or victim surveys originally intended to replace recorded crime statistics can only complement them. Therefore it is important to tackle and attempt to solve all those methodological problems that have emerged since the days of Quetelet and other pioneers in the field of criminal statistics.

If these observations are already true for every national situation to which criminal statistics are normally related they are all the more valid when one tries to compare statistical figures produced in different countries. Criticisms that comparisons across jurisdictions are virtually impossible or at least misleading given the many differences in legal definitions, reporting and recording practices are as old as the earliest writings of Quetelet in comparative criminology. The difficulties to overcome these problems may have led to the fact that international comparative data in criminal justice are relatively scarce compared to other fields of public interest. However, there are now signs that an upturn of international research in criminology is taking place. In view of an increasingly interlinked world and an ever converging Europe, not only are economic and social comparisons needed, but it is also necessary to compare the development of crime figures as well as criminal policies and criminal justice systems.

The European Sourcebook of Crime and Criminal Justice Statistics is a prominent example of such efforts to collect and evaluate national data on all levels of the criminal justice chain and to bring them together in a European comparative perspective. In contrast to other data collections, it has established a common understanding of concepts, furthered comparability by using standard definitions for crime types and penal measures or at least improved the evaluation by demonstrating the deviations across jurisdictions. The various articles of this volume refer to the recent edition of the European Sourcebook. They focus on the question how the given data can be used for criminological analyses especially of crime and criminal justice trends, while also taking into account the methodological difficulties and the problems caused by lacking data.

This issue starts with a comprehensive approach to describing and evaluating “Crime and Justice Statistics collected by International Agencies” (Chris Lewis), thus relating the European Sourcebook to other international data collections on crime and justice. The article discusses the advantages and disadvantages of the different collections and states that those associated with international bodies, like the UN Crime Trends Survey (UNODC), EuroStat Data Collection, CEPEJ and SPACE of the CoE on criminal justice and prison systems, are more likely to continue whereas other collections based on the work of a small number of experts, e.g., the European Sourcebook, could be taken over by EuroStat in the long run.

A crucial issue when internationally comparing crime and criminal justice data are the different legal concepts of offences defined by the national legislator. The European Sourcebook has tried to find a solution for improving the comparability of offence data by establishing standard definitions and breaking down the offence types in special items to be included or excluded. Thus for every offence type and for each country one can identify to what extent national data correspond to the standard definition. The article by Stefan Harrendorf describes the methodology, the questionnaire design and the validation process referring to offences and analyzes the degree of conformity and deviation in terms of the standard definitions as such and the special items in particular. In this respect some problematic offence definitions can be identified. But overall the offence definitions turn out to work well and the conformity rates tend to be high.

A comprehensive and complex evaluation of European Sourcebook data is presented by Paul Smit/Anneke Zuiderwijk-van Eijk/Rob Decae. They take into account the figures for all levels of the criminal justice system concerning the total of offences and some special offence groups for all countries concerned. In order to observe a long term period, data from the recent editions are used as well as from former data collection waves. Thus figures from 1990 to 2006 are available, but of course problems of missing or outlying data had to be solved. For the sake of reducing complexity the authors have clustered the European countries in North-West, South, Central, and East Europe. The article not only measures crime rates, but relates them to suspects, convicted persons and prisoners using specially developed indicators: the relation between suspected and convicted persons (conviction ratio) and the relation between sanctioned persons and prisoners (so called punitivity). Among a multitude of observations a decreasing trend of punitivity in East Europe and an increasing one in North-West Europe can be observed, but nonetheless significant differences in the levels remain.

The article by Beata Gruszczynska/Markku Heiskanen is based on figures from crime statistics and focuses on some common crimes: homicide, assault, rape, robbery, car theft, domestic burglary, and drug offences. The 10 year trends (1998 to 2007) are shown for European countries, clustered in Western Europe on the one hand and on the other in Central and Eastern Europe where as the authors demonstrate significant socio-political and legal changes have occurred. Overall property offences, homicide and robbery have decreased whereas drug offences and assault have increased which cannot be explained by increasing reporting activities of victims. Besides these common trends the levels of crimes differ considerably between Western and Eastern Europe, this is especially true for homicide and assault.

Whereas trend analyses are usually based on police statistics data the paper by Marcelo Aebi refers to the court level. It thoroughly analyzes the reliability and validity of conviction statistics data and draws the conclusion that conviction statistics are less valid but more reliable than police statistics as a measure of crime. Therefore the article does not use conviction data as a measure of crime level, but of crime trends. Of course, there have been some changes in legal definitions, e.g., concerning rape, but - as far as the authors could ascertain – they mostly had only spot effects that did not alter the overall trends of convictions for each type of offence. The major findings for the given period are a decrease of theft and an increase of robbery and assault. The latter is broadly discussed: is this a social artifact because of rising reporting rates or a real increase? At the end the article integrates these findings in a broader view of long term crime trends, discusses theoretical explanations and proposes an explanatory multifactor model inspired by opportunity-based theories.

To conclude, the paper by Jörg-Martin Jehle deals with a specific subject: the attrition or loss of cases from one level to the other within the criminal justice chain, particularly concerning sexual offences. Because of the lack of flow statistics one cannot follow cases from level to level, but only relate quantities at police and court level. The author compares the rates of offences, suspects and convicted persons as well as the so called conviction ratio (relation of convicted to suspected persons) between the countries selected (England/Wales, France, Germany, the Netherlands, Poland, and Sweden). Partly due to the different national legal concepts of rape, sexual assault and sexual abuse of minors the input level in the form of recorded offences shows big differences which decrease somewhat at the following levels.

The various papers published in this issue demonstrate that the European Sourcebook provides a plentitude of data which can be analyzed from different points of view and via different approaches. Since an open electronic access to the data set is offered every interested researcher may use these data for testing his/her own assumptions or for comparing them to other data sources.