A STUDY OF SENTENCING DISPARITY BETWEEN 50 SECOND CIRCUIT JUDGES INVOLVED TWO PHASES: (1) JUDGES WERE REQUESTED TO RENDER SENTENCES ON THE BASIS OF 20 PRESENTENCE REPORTS, AND (2) SUBJECTS WERE DIVIDED INTO TWO GROUPS, WITH EACH ASKED TO RENDER SENTENCES ON A BASIS OF 10 SLIGHTLY DIFFERING PRESENTENCE REPORTS. RESULTS SHOWED EXTENSIVE SENTENCING DISPARITY, BUT NO PATTERN WAS EVIDENT WHICH COULD EXPLAIN THE SENTENCING VARIATIONS. IN THE STUDENT-JUDGE STUDY, PRESENTENCE REPORTS WERE DISTRIBUTED TO THE 95 STUDENTS, ALONG WITH ALL FACTS RELATING TO THE CASES. THE SUBJECTS WERE RANDOMLY DIVIDED INTO TWO GROUPS, WITH EACH BEING GIVEN ONE VERSION OF FOUR CASES. CHARACTERISTICS FEATURED IN THE CASES WERE THE SEX, RACE, CRIMINAL RECORD, AND ETHNIC SURNAME OF THE DEFENDANTS, AND THE NATURE OF THE OFFENSES (BLUE-COLLAR OR WHITE-COLLAR). EACH STUDENT WAS ASKED TO RENDER A SENTENCE IN ALL FOUR CASES, WITH THE CRITERIA FOR SENTENCING COINCIDING WITH THE INDETERMINATE SENTENCING RANGE USED BY THE SECOND CIRCUIT JUDGES. SUBJECTS WERE ALSO REQUESTED TO IDENTIFY THEMSELVES BY SEX, RELIGION, RACE, AND FAMILY INCOME. THE KOLGOMOROV-SMIROV TWO-SAMPLE TEST WAS USED BECAUSE IT IS SENSITIVE TO DIFFERENCES IN THE DISTRIBUTION OF THE TWO GROUPS BEING TESTED. THE 2-SAMPLE MEDIAN TEST WAS ALSO PERFORMED ON THE DATA. THE KOLGOMOROV-SMIROV TEST SUGGESTS THAT THE BLACK AND WHITE STUDENT-JUDGES DIFFERED SIGNIFICANTLY IN THEIR SENTENCES (P=.032), WITH THE BLACK STUDENTS TENDING TO BE MORE LENIENT WITH THE UNDERPRIVILEGED OFFENDERS. THE LOWER INCOME GROUP WAS ALSO SIGNIFICANTLY LESS SEVERE IN THE SENTENCING OF THE UNDERPRIVILEGED THAN THE HIGHER INCOME GROUP. NO OTHER STATISTICALLY SIGNIFICANT CORRELATIONS WERE EVIDENCED. EXTENSIVE SENTENCING DISPARITY WAS REVEALED IN BOTH THE REAL JUDGE AND STUDENT-JUDGE STUDIES, BUT WITH THE EXCEPTIONS NOTED IN THE STUDENT GROUP, DISPARITY COULD NOT BE EXPLAINED BY THE INTERACTION OF THE CHARACTERISTICS OF THE JUDGE WITH THE CHARACTERISTICS OF THE CASE OR DEFENDANT INVOLVED. REFERENCES ARE PROVIDED. (RCB)