THE REPORT OPENS WITH A DESCRIPTION OF THE FRAMEWORK FOR SOCIAL INDICATORS PROPOSED BY THE ORGANIZATION FOR ECONOMIC CO-OPERATION AND DEVELOPMENT (OECD) WORKING PARTY FOR SOCIAL INDICATORS AND THE CONSEQUENT NEED FOR VALID INDICATORS OF VICTIMIZATION. REQUIREMENTS IMPOSED BY THE CRITERION OF VALIDITY AND THE CHOICE OF APPROPRIATE MEASURES ARE ALSO DISCUSSED. A SURVEY OF MEDICAL AND POLICE STATISTICS CONCLUDES THAT BOTH OFFICIAL SOURCES OF DATA RELIABLE FOR MEASURING DEATHS FROM INTENTIONAL VIOLENCE BUT THAT NEITHER IS RELIABLE AS A DATA BASE FOR CONSTUCTING INDICATORS OF INJURY OR MENTAL SUFFERING FROM VIOLENCE. THE POSSIBILITY OF USING SURVEYS OF SELF-REPORTED DELINQUENCY AND SURVEYS OF VICTIMIZATION IS CONSIDERED. SELF-REPORT SURVEYS ARE CONCLUDED NOT TO PROVIDE RELIABLE INFORMATION ON THE ACTUAL HARM SUFFERED BY INDIVIDUALS AS A CONSEQUENCE OF VICTIMIZATION. DESPITE METHODOLOGICAL AND TECHNICAL PROBLEMS, VICTIM SURVEYS ARE CONCLUDED TO BE THE BEST POTENTIAL SOURCES FOR SUCH INFORMATION. THE PROBLEM OF COMBINING POLICE STATISTICS WITH VICTIM SURVEYS IS EXAMINED, AS IS THE PROBLEM OF WEIGHTING DATA. DIRECTIONS FOR FURTHER RESEARCH ARE SUGGESTED. OECD COUNTRIES ARE URGED TO COOPERATE IN ATTEMPTS TO SOLVE METHODOLOGICAL PROBLEMS WITH VICTIM SURVEYS. A LIST OF REFERENCES AND SUMMARY DESCRIPTIONS OF MAJOR VICTIM SURVEYS CARRIED OUT IN OECD COUNTRIES ARE INCLUDED. (LKM)