Recidivism after a community service is compared to recidivism after imprisonment in a matched sample of adult Dutch offenders. We use longitudinal, official record data to compare recidivism over a maximum of eight years. ‘Propensity score matching’ and ‘matching by variable’ are used to take selection-effects into account. After a community service offenders turn out to recidivate less than after a prison sanction.

How many criminal suspects become victims themselves? A high correlation of offence and victimisation is to be expected because of a marked similarity between offenders and victims. We assume that offenders strongly fulfil the lifestyle and environment criteria, associated with an increased risk of victimisation. By studying self-reported victimisation data from 39.978 respondents, matched to registered criminal suspects, we verified our assumption that criminal suspects are at considerably higher risk of being victims. This correlation is especially high for crimes of violence and for people suspected of multiple crimes.

Developmental trajectories are estimated of 147 high-risk girls from 12 to 32, based on conviction data. Four trajectories are identified: adolescence-limited, low frequency desisting, high frequency desisting, late onset. Despite their high delinquency risk, only few girls developed a long-term criminal career. Girls in the trajectories differ with regard to intelligence, self-concept, social skills, aggression, personality disorders, and divorced parents. For each trajectory a profile is given, based on type of offences, personality and background characteristics.

Filicide, the murder of a child by a parent, is a dramatic event. The gravity increases when the perpetrator resorts to committing or attempting suicide. It is assessed to what extent filicide-(para)suicides can primarily be understood as homicidal or suicidal behaviour, or as a separate category of lethal violence. Parents committing filicide-parasuicide differ from filicidal parents and other suicidal parents in sociodemographic, individual and offence-related characteristics. Filicide-(para)suicide seem to constitute a category of lethal violence, different from both filicides and parasuicides.

Using data from judicial documentation on adult adjudicated offenders in the Netherlands, an actuarial risk prediction instrument has been developed. StatRec estimates the four year reconviction rate of adult offenders, based on a limited number of static factors. The metric qualities of the scale are good. StatRec produces a precise estimate of the base rate in the group of individuals with the same combination of background characteristics as the offender. The scale does not use dynamic and situational factors that may influence the risk of reoffending. Taking them into account only slightly enhances the predictive power. StatRec can be used to validate more specific risk assessment tools.