A 2-year evaluation was initiated in 1974. The surveys included a national sample of 60,000 households and of 39,000 commercial establishments, samples of 10,000 households within specified municipal city limits, and corresponding samples of 1,000 to 5,000 businesses in the same cities. Researchers found that the initial, intensive research programs examining the surveys' quality had declined in activity and that the survey's objectives had changed. The surveys were originally intended as social and policy indicators but have subsequently been used, inappropriately, for producing operating intelligence for jurisdictions. In addition, serious technical difficulties in the measurement of the relatively rare and ill-defined events covered in the surveys were identified. These included numerous telescoping and memory decay factors coupled with differentials in underreporting of victimization events that vary substantially across categories of respondents, crimes, and offenders. Modifications in the survey instruments were suggested to include more independent variables -- such as 'opportunity' and 'vulnerability' -- that are related directly to the phenomena of victimization. Other changes might include the elimination of certain concepts used in the Uniform Crime Reports and a reassessment of the category of 'series victimizations.' Also criticized were the lack of sufficient personnel for managing and operating the National Crime Surveys, the use of different methodologies for the national and city surveys, and the limitations of the commercial surveys. Footnotes are included.