Conflict theory postulates that lower class criminal offenders receive the most severe legal sanctions. The empirical literature that reports on the testing of the validity of this proposition, however, is equivocal. The study reported in this article extended previous research in this area in three significant ways. First, it reduced the influence of random sentencing variation by examining sentencing practices in a State with legally mandated sentencing guidelines. Second, it examined individual offense categories, a strategy that provides a more comprehensive profile of the factors that are relevant in the sentencing of a particular offense. Third, the study explored several different regression specifications for each offense category, providing a more accurate assessment of those factors that were consistently important in determining sentence severity. The study sample consisted of 2,760 offenders selected randomly from the admissions population committed to the custody of the State's department of corrections during fiscal year 1985. The study hypothesized that the severity of imposed legal sanctions would depend on the interplay between an offender's socioeconomic status and offense type. Results from four different regression models provide some support for this hypothesis. A significant inverse relationship was observed between socioeconomic status and length of sentence for manslaughter and the possession of narcotics. Findings also show that extralegal factors had a greater role in the sentencing of violent and moral-order offenders, and prior criminal record was more salient in the sanctioning of property offenders. Additional offense-specific analyses are required to provide information on the relationship between socioeconomic status and criminal sentencing. 2 tables, 1 figure, 4 notes, and 86 references