youth violence is a serious social problem that is often encountered in the criminal justice system. currently, there are very few instruments available that have been specifically designed to assess violence risk in juvenile offenders. this study examined the predictive and concurrent validity of a new instrument, the structured assessment for violence risk in youth (savry), in a sample of adolescent offenders. the savry was retrospectively coded using file information on 127 juvenile offenders aged 12 to 18 years. a follow-up period with a mean of 35.5 months (so = 14.8, range = 6.0 to 61.0) was used to determine general and violent reoffending. results substantiated the predictive validity of the savry total score for both general and violent reoffending with areas under the curve of .75 and .77, respectively. similarly, analysis of the predictive validity of the savry risk ratings indicated that youths in the higher risk categories were at significantly greater risk for both general and violent reoffending. as well, the savry was found to be superior to the youth level of service/case management inventory in the prediction of both general and violent reoffending. limitations of the current research are discussed.