Nevertheless, enough evidence of disparity exists to justify a search for contingencies under which discrimination is more or less likely to occur. This study used data from 200 cases in each of 31 Florida counties to test 5 contingent hypotheses. In general, data suggest that discriminatory processing is a relatively rare occurrence in Florida. Of 90 tests, only 13 yielded statistically significant findings. While there was a small but significant overall disparity by race in the total sample, it was sustained in only 11 of 29 subgroup analyses and was large in only 2 instances. In addition, socioeconomic status and family structure had no influence on dispositions. There was no systematic variation by degree of penetration into the system, rationality of court organization, wealth of the population of the jurisdiction, or seriousness of the offense (except by sex). Results suggest that differential sanctioning is contingent on whether the individual sanctioned is a member of an aggregate that poses a threat to elites. The relevant threats are not strictly political, but are primarily of a symbolic, social-psychological nature wherein white adults react differentially to resented youthful traits often stereotypically associated with nonwhites. 6 tables, 3 notes, and 88 references. (Author abstract modified)