Data for this study were collected during the Mississippi DUI Probation Follow-Up Project conducted in 11 cities between 1975 and 1981. To identify distinctive arrest profiles within the DUI sample, the study used a Q mode factor analysis followed by a discriminant function analysis. Five subgroups emerged: a low offense group of 202 individuals; 63 offenders in a mixed group with a higher number of total arrests than the low offense group and diverse types of offenses; a young traffic group of 58 persons with many other hazardous moving violations other than DUI; 16 persons in the public drunkenness category with high numbers of alcohol-related offenses, particularly public drunkenness offenses; and 14 persons with the highest numbers of license and equipment violations called the license group. The two last subgroups differed on demographic and drinker status variables. The public drunkenness group had the highest accident rate. A discussion of treatment implications concludes that programs focusing exclusively on changing alcohol consumption behavior are not likely to reduce accident risk for some offender groups and suggests matching programs to offender type. Tables and 18 references. (Author abstract modified)