THE UNIFORM CRIME REPORTS, BASED ON SYSTEMATIC REPORTS BY POLICE DEPARTMENTS TO THE FEDERAL BUREAU OF INVESTIGATION (FBI), AND VICTIM SURVEYS BY THE U.S. CENSUS BUREAU, BASED ON INFORMATION FROM A REPRESENTATIVE SAMPLE OF THE POPULATION WHO REPORT BEING A VICTIM, ARE THE INDICES EXAMINED. THE EIGHT FACTORS THOUGHT TO INFLUENCE THE INCIDENCE OF CRIME WHICH ARE USED IN THE STUDY ARE: POPULATION DENSITY, HOUSEHOLD CROWDING, UNEMPLOYMENT, PROPORTION OF THE POPULATION WHICH IS BLACK, MARITAL INSTABILITY, PROPORTION OF THE POPULATION WHICH IS UNDER 18 YEARS OF AGE, RESIDENTIAL TRANSIENCY, AND SIZE OF COMMUNITY RESOURCES ALLOCATED TO PUBLIC SAFETY. TWENTY-SIX U.S. CITIES IN WHICH VICTIMIZATION SURVEYS HAVE BEEN CONDUCTED CONSTITUTED THE UNITS OF ANALYSIS. DATA FOR THE CITIES WERE OBTAINED FROM LEAA PUBLICATIONS, FBI UNIFORM CRIME REPORTS, AND THE U.S. COUNTY AND CITY DATA BOOK. DATA ON ROBBERY, BURGLARY, AND AUTO THEFT WERE SELECTED FOR ANALYSIS. THE ANALYSIS OF THE EFFECT OF THE EIGHT DIFFERENT FACTORS ON THE INCIDENCE OF ROBBERY, BURGLARY, AND AUTO THEFT SHOWED THAT QUITE DIFFERENT CONCLUSIONS WERE REACHED DEPENDING UPON WHETHER THE MEASURE USED WAS THE UNIFORM CRIME REPORTS INDEX OR ONE BASED ON VICTIMIZATION SURVEYS. THE INCONSISTENCIES REVEALED BY THE ZERO ORDER CORRELATION ANALYSIS ARE SUPPORTED BY THE MULTIPLE REGRESSION ANALYSIS. IT IS CONCLUDED THAT, LACKING EVIDENCE SHOWING ONE MEASURE TO BE THE MORE VALID INDICATOR OF CRIME, NEITHER IS A SATISFACTORY INDEX OF CRIME FOR THE PURPOSE OF EXPLAINING THE CAUSES OF CRIME. IT IS SUGGESTED THAT BOTH INDICES NEED TO BE REFINED TO MAKE THEM MORE USEFUL INDICATORS. TABULAR DATA ARE INCLUDED. (RCB)