The key findings of this study are that selected elements on which SM is based do contribute to the prediction of reoffending and that open prisons are associated with a reduced probability of reoffending during the first year after release. In Scotland, SM has been adopted in order to engage long-term prisoners (those serving 4 years or more) in the identification of their needs and then to direct resources to address those needs. SM is intended to provide a record of change toward reducing the risk of recidivism; collecting data on the effectiveness of interventions; and assessing levels of various types of offender needs so as to inform the development of prison regimes and enable the prioritization of resources. The SM data may be used to allocate resources toward those offenders statistically more likely to reoffend and as a measure of the future effect of initiatives to address reoffending, such as the recently introduced Integrated Cases Management process and the Home Detention Curfew system (electronic tagging) within the Scottish Prison Service. SM data, however, must first be improved in order to facilitate sustained analysis and use. The dependent variable for this study was reoffending compared with no reoffending. The independent variables were derived from the computer records for all released long-term prisoners since 2002. Data were in seven domains: demographic, socioeconomic, offense history, SM automatic referrals, SM psychometric and behavioral tests, SM risk and needs assessment, and SM supervision levels. 8 tables, 1 figure, and 37 references