Results indicate that when controlling for other incident-level variables, homicide incidents with non-White offenders are more likely to be cleared than are incidents with White offenders, regardless of the victim's race. This finding is consistent with the proposition that offender's status is inversely related to likelihood of arrest as well as with the focus of social reaction theorists on racial notions of the typical criminal. The results provide some support for the argument that because non-White homicide offenders are perceived as greater threats to society, the police are more willing to invest effort in pursuing such offenders. Such pressure might be stronger for homicide than for lesser offenses because homicide investigations receive more public or media attention. For aggravated assault, the results indicate that incidents between Whites were most likely to be cleared, followed by interracial incidents. Aggravated assault incidents between non-Whites were least likely to be cleared. Results may indicate police devaluation of incidents involving people from lower social positions and reluctance to investigate such incidents for less serious offenses that receive less public attention. Data were collected from the National Incident Based Reporting System (NIBRS) the largest-scale incident-level dataset collected by the Federal Bureau of Investigation and was developed to eventually replace Uniform Crime Reports. Tables, appendix A-B, notes, and references