Workshop participants looked at the origin of concern about race-crime statistics, whether these statistics should be collected and if so how often, the politics of race-crime statistics, general issues in the collection of such statistics, and purposes and usefulness of the statistics. It was generally agreed that race-crime statistics need to be collected in order to understand whether the criminal justice system treats people of different racial or ethnic groups fairly. The statistics can be useful in uncovering discrimination, determining what kinds of programs would be most useful in the correctional system, and identifying groups most likely to enter the criminal justice system. On the other hand, race-crime statistics collected by the police should not be generally used as indicators of various groups' level of involvement in crime. Problems in the collection of race-crime data are noted such as systematic bias in the assessment of race and ethnicity, high levels of unreliability or random error if data are collected from the perspective of criminal justice system decisionmakers, and the reluctance of both victims and offenders to disclose data. Appendixes contain a list of workshop participants and information on the racial origin variable in Canada's Uniform Crime Reporting.