Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
âBâ Principal Findings and Recommendations of the National Research Council (1976b) Study Bâ1 FINDINGS 1. The design of the NCS generally is consistent with the objective of producing data on trends and patterns of victimization for certain cat- egories of crime. 2. Conceptual, procedural, and managerial problems limit the potential of the NCS, but the panel considers that, given sufï¬cient support, the problems ought to be amenable to substantial resolution in the long run. 3. A major shift of resources to analytic and methodological research is essential in order for the NCS to yield data useful for policy formula- tion. This shift should be accompanied by the development of admin- istrative mechanisms to enhance this large and complex seriesâ capacity for self-correction. 4. The primary uses envisioned originally for the NCS were of a social and policy indicator nature. The panel agrees that a subsequent objec- tive of producing operating intelligence for jurisdictions is inconsistent with the original purposes of the NCS and with the design informed by 149
150 SURVEYING VICTIMS those purposes, except insofar as operating intelligence is a by-product of understanding broad trends and patterns of victimization. Bâ2 RECOMMENDATIONS 1. A substantially greater proportion of [Ofï¬ce of Justice Programs] re- sources should be allocated to delineation of product objectives, to managerial coordination, to data analysis and dissemination, and to a continuing program of methodological research and evaluation. We are concerned about the current balance between resources allocated to data collection and resources allocated to all other aspects of the victimization survey effort. 2. The staff providing managerial and analytic support for the NCS should be expanded to include the full-time efforts of at least 30 to 40 professional employees. Without this expansion, the NCS cannot be developed to achieve its potential for practical utility. 3. A coordinator at the Bureau of the Census should be appointed whose responsibilities would crosscut the various Census operations that sup- port the NCS. 4. The staff that performs NCS analysis and report-writing functions, whether LEAA employees or otherwise, should have an active role in the management of the NCS. Speciï¬cally, the analytic staff should par- ticipate in the development of objectives for substantive reports and publication schedules. Once analytic plans are formulated, the analy- sis staff should have autonomy in specifying tabulations to be used in support of the analysis, and it should have direct access to complete NCS data ï¬les and to data processing resources. It should be the ana- lytic staff âs responsibility to formulate statistical or other criteria used in hypothesis testing. Finally, a feedback mechanism should be insti- tuted through which the staff can inï¬uence decisions on the content of survey instruments, on ï¬eld and code procedures, and on analytic and methodological research to be undertaken. 5. Resources now used for the nationwide household survey and for the independent city-level household surveys should be consolidated and used for carrying out an integrated national program. The inte- grated effort could produce not only nationwide and regional data, but, on the same timetable, estimates for separately identiï¬able Stan- dard Metropolitan Statistical Areas (SMSAs) and for at least the ï¬ve largest central cities within them. For some purposes, it would be prac- ticable and perhaps useful to combine data for 2 or more years and to
APPENDIX B 151 show separate tabulations for a large number of cities and metropoli- tan areas. 6. A review and restatement of the objectives of the commercial surveys should be conducted and data collection should be suspended, except in support of experimental and exploratory review of these objectives. 7. Five percent of the NCS sample in the future should be available to in- terview in order to explore different forms and ordering of questions, and for pretesting possible new questions. . . . 8. Routine NCS tabulation should include results on the risk of victim- ization, where the unit of analysis is the surveyed individual, and that analysis of risk should be a signiï¬cant part of NCS publications on a recurring basis. If the NCS data are coded and tabulated so as to yield a cumulative count of personal and household victim experiences of all surveyed respondents, analyses of multiple victimization, including events now excluded as âseriesâ incidents, could and should be routine components of ofï¬cial publications. 9. A major methodological effort on optimum ï¬eld and survey design for the NCS should be undertaken. Toward this goal, high priority should be given to research on the best combination of reference period, fre- quency of interview at an address, length of retention in the sample, and bounding rules. Part of the recommended research in this area should be a new reverse record check study in order to assess: (a) dif- ferential degrees of reporting for different types of victimizations and different classes of respondents, (b) problems of telescoping and decay, and (c) biases in the misreporting of facts. 10. Local interest in victimization patterns should be addressed through LEAA-Census joint development of a manual of procedures for con- ducting local area victimization surveys. The federal government should produce reports on the NCS that contain detailed analyses of patterns and trends of victimization so as to allow law enforcement personnel, the public, and policymakers to draw inferences that might be applicable to the issues with which they are concerned. Informing the public and their policymakers of the distribution and modiï¬ability of risk should be the primary objective of the NCS. SOURCE: National Research Council (1976b:3â5).