Data were obtained from the records of the U.S. Sentencing Commission made available on the Internet through the Inter-Consortium for Political and Social Research, which consists of information on all defendants convicted of Federal offenses. The study used the drug-offender portion of these data for fiscal year 1995. The drug offenders in the study were all sentenced under Federal statutes and were from various social backgrounds and judicial districts. An overwhelming majority received imprisonment as a sentence. Due to the use of determinate sentencing by the Federal courts, the data permitted the macro analysis of mandatory minimum sentencing guidelines for drug offenses. The study used ordinary least squares to determine whether race/ethnicity had a direct effect on sentencing, controlling for other factors. Length of imprisonment, measured in months, was the dependent variable. Independent variables pertained to demographic, case-processing, and offense-specific variables. The study findings confirm those of other studies, in that case-processing and offense-specific characteristics were the most important social predictors of sentencing outcomes for drug offenses. Varying regional behaviors and stereotypes, as well as different judicial subcultures and the use of discretion in sentencing affected the penalties given to Federal drug offenders. The characteristics of the theoretical "villain" in drug offenses (those offenders who deserve longer sentences) differed according to region. Offenders who received shorter sentences (perceived by the court as "victims") tended to be non-Hispanic, accepted responsibility for their offense, accepted a plea bargain with acknowledged guilt, used no firearm, and trafficked a small amount of the illegal substance. Although the "villains" (those who received longer sentences) of drug offenses tended to be the potentially violent, high-level traffickers, and career criminals, the label of "villain" was also applied to others, e.g., Hispanics and/or the individuals who defended their right to trial. Criminal justice practitioners should be made aware of the tendency to view certain ethnic groups, notably Hispanics, as primarily "villains" rather than "victims" when preparing presentence reports and deciding on departures from sentencing guidelines. 5 tables, 3 notes, and 48 references