NCJ Number
251442
Date Published
December 2017
Length
61 pages
Annotation
This report provides information on the history, scope, and data collection and processing of the National Crime Victimization Survey (NCVS), which is the only nationally representative household survey that provides data on criminal victimizations in the United States, whether or not they have been reported to police.
Abstract
This Technical Documentation for the NCVS provides NCVS data users with details on the survey's sampling, data collection, data processing, and weighting and estimation methodologies that remain generally stable unless there are significant sample or survey design changes by the U.S. Justice Department's Bureau of Justice Statistics (BJS), which sponsors the NCVS. Following a background chapter that describes the history, key measures, and supplements of the NCVS, a chapter reviews the sample design of the NCVS. This encompasses the sampling frame, stratification, first- and second-stage sample selection, sample sizes, and special sample situations. The third chapter focuses on data collection, which encompasses field operations, survey instruments, and non-response. The fourth chapter addresses the processing of survey data after its collection. This chapter covers such topics as coding operation, consistency edit and imputation, type of crime classification, and period-to-period recount. The remaining two chapters of this report explain weighting and estimation procedures. The appendices provide more details on terms and definitions, the chronology of NCVS question changes since January 2000, NCVS supplements conducted for the period 2000-2017, interview outcome codes, SAS code for direct variance estimation, illustrations of GVF variance estimation, and NCVS year-to-year correlations for 2014 and 2015. 17 references