THE FOCUS OF THE STUDY WAS ON THE RELATION BETWEEN SUBJECTIVE AND OBJECTIVE CRITERIA, BUT OBJECTIVE PERFORMANCE INDICES WERE VIEWED AS COMPONENT PARTS (PREDICTORS) OF SUBJECTIVE RATINGS TO DETERMINE THE EXTENT OF THE RELATION. TWO HYPOTHESES WERE TESTED: (1) SUPERVISORY RATINGS ARE LINEARLY PREDICTABLE, GIVEN KNOWLEDGE OF RATEE OBJECTIVE PERFORMANCE; AND (2) SUPERVISORY RATINGS ARE MORE STRONGLY RELATED TO OBJECTIVE INDICES OF PERFORMANCE FOR MINORITY EMPLOYEES THAN FOR NONMINORITY EMPLOYEES. THE TOTAL STUDY SAMPLE CONSISTED OF 952 POLICE OFFICERS FROM A LARGE METROPOLITAN POLICE DEPARTMENT IN DADE COUNTY, FLA. THERE WERE 795 INDIVIDUALS IN THE NONMINORITY GROUP AND 157 INDIVIDUALS IN THE MINORITY GROUP. IN ADDITION TO AGE AND JOB TENURE, DATA WERE AVAILABLE ON 35 OBJECTIVE AND 8 SUBJECTIVE CRITERION MEASURES. THE EIGHT SUBJECTIVE MEASURES WERE JOB KNOWLEDGE, JUDGMENT, INITIATIVE, DEPENDABILITY, DEMEANOR, ATTITUDE, RELATIONS WITH OTHERS, AND COMMUNICATIONS. STATISTICAL ANALYSIS TECHNIQUES WERE EMPLOYED TO ANALYZE DATA. THE FIRST HYPOTHESIS WAS SUPPORTED IN BOTH MINORITY AND NONMINORITY CROSSVALIDATION SUBGROUPS, WITH AND WITHOUT THE EFFECTS OF AGE AND JOB TENURE. THE SECOND HYPOTHESIS WAS NOT SUPPORTED, AND REASONS FOR THE LACK OF STATISTICAL SIGNIFICANCE OF DATA REGARDING THIS HYPOTHESIS ARE OFFERED. THE FINDING THAT SUPERVISORY RATINGS WERE LINEARLY PREDICTABLE FROM OBJECTIVE PERFORMANCE INDICES FOR MINORITY AND NONMINORITY SUBORDINATES COMPORTS WITH CIVIL RIGHTS LEGISLATION AND U.S. SUPREME COURT DECISIONS. SUPPORTING DATA AND REFERENCES ARE PROVIDED. (DEP)