ABSTRACT
Crime Statistics from the National Crime Victimization Survey (NCVS), like all statistics, are affected by measurement error and missing data. This chapter looks at those sources of error and the steps taken to assess and reduce them. The NCVS obtains information from a sample of households and thus has sampling variability: if a different sample were selected the estimates would be different. The NCVS statistic for number of burglaries estimates the number of victimizations that would be counted if every household in the US population answered the survey questions about burglary—not necessarily the same thing as the “true” number of burglaries. Most NCVS respondents report two or fewer victimization incidents during their interview. The NCVS selected new samples of counties in 2006 and 2016. The NCVS margin of error measures uncertainty about the estimates from sampling variability, that is, the error caused by taking a sample instead of a census of the population.
