Victims of Crime - The National Crime Victimization Survey

In 1972 the Law Enforcement Assistance Administration established the National Crime Victimization Survey (NCVS). The survey is an annual federal statistical study that measures the levels of victimization resulting from criminal activity in the United States. The survey was previously known as the National Crime Survey, but it was renamed to emphasize the measurement of victimization experienced by citizens.

Sponsored by the BJS, the survey was created because of a concern that the FBI's Uniform Crime Reports (UCR) did not fully portray the true volume of crime. The UCR provided data on crimes reported to law enforcement authorities, but it did not estimate how many crimes went unreported.

The NCVS is designed to complement the FBI's Uniform Crime Reports. It measures the levels of criminal victimization of persons and households for the crimes of rape, robbery, assault, burglary, motor vehicle theft, and larceny. Murder is not included because the NCVS data is gathered through interviews with victims. Definitions for these crimes are the same as those established in the FBI's UCR.

Many observers believe the NCVS is a better indicator of the volume of crime in the United States than the FBI statistics. Nonetheless, like all surveys, it is subject to error. The survey depends on people's memories of incidents that happened up to six months before. Many times, a victim is not sure what happened, even moments after the crime occurred.

Errors can come from other factors as well. Individuals who have been repeatedly victimized—by spousal or parental abuse, for example—may not remember individual incidents, or may remember only the most recent event. For instance, the NCVS found that a disproportionately large number of incidents are reported to have occurred at the end of the time period covered by the survey when the victim's memory was perhaps fresher. In addition, the NCVS limits the data to victims age 12 and older—an admittedly arbitrary age selection. Despite these factors, however, the BJS claims a 90 to 95 percent confidence level in the data reported in the NCVS.

The NCVS and the FBI's UCR are generally considered the primary sources of statistical information on crime in the United States. Like all reporting systems, both have their shortcomings, but each provides valuable insights into the status of crime in the United States. Over the years some significant differences have occurred in their findings. For example, the Uniform Crime Reports saw a 15 percent increase in crime from 1982 to 1991, while the NCVS reported a leveling off and, in 1990, a decrease in crime. These differences require the reader to evaluate both sets of statistics carefully, not relying solely on one or the other.

Redesigned Survey

Beginning in 1979, the NCVS underwent a thorough, decade-long redesign. The new design was expected to improve the survey's ability to measure victimization in general and particularly difficult-to-measure crimes, such as rape, sexual assault, and domestic violence. Improvements included the introduction of "short cues" or techniques to jog respondents' memories of events. Generally the redesign, as anticipated, resulted in an increased number of crimes counted by the survey. Therefore pre-1992 data cannot be directly compared to the later data.