Data were obtained from the computerized databanks of the Community and Youth Corrections Division of Manitoba. The data encompassed the Manitoba risk/needs classification instrument; criminal history; personal demographics; and a breakdown of aboriginal offenders into metis, "treaty-on," and "treaty-off." "Treaty-on" refers to offenders with registered Indian status as defined under Canadian Federal legislation and who were living on a reserve at the time of admission to probation. "Treaty-off" refers to offenders with registered Indian status who were living off a reserve at the time of admission to probation. Metis/non- status refers to the group without registered Indian status who rarely live on a reserve at any time. The aboriginal categories were all self-reported at the time of admission to probation. The database consisted of approximately 11,600 adult probationers from 1986 to 1991. Random samples of 200 probationers were drawn from each of the 6 years. There were 513 non-aboriginal (56.8 percent) and 390 (43.2 percent) aboriginal offenders. The major finding was that a risk/needs classification instrument originally developed on a sample of non-aboriginal offenders showed predictive validity among aboriginal offenders. Establishing the validity of the Manitoba Risk-Needs Scale with aboriginal offenders also implies that the risk factors are similar for aboriginal and non-aboriginal offenders. Although some of the individual items did not predict as consistently for the "treaty" group as for the metis/non-status group, important factors such as criminal history, substance abuse, and criminal peers showed good predictive validity. The findings support a social psychological perspective of criminal conduct that views risk-needs factors as the same for groups regardless of culture and race. 3 tables, 3 notes, and 27 references