Published on: Friday, December 8, 2017

On December 7th, the Justice Department’s Bureau of Justice Statistics issued a press release announcing it released estimates of crime from the 2016 National Crime Victimization Survey (NCVS).  The report, Criminal Victimization, 2016, estimates crime differently from the crime data reported by the FBI’s Uniform Crime Reporting Program—the NCVS includes a different set of offenses, crimes that were reported and not reported, and collects data from residents instead of law enforcement. 

Among other things, the survey found in 2016 “U.S. residents age 12 or older experienced 5.7 million violent victimizations—a rate of 21.1 victimizations per 1,000 persons age 12 or older.”  This reflects a rise from 18.6 victimizations per 1,000 persons age 12 or older found in the 2015 NCVS.  Although the 2016 NCVS appears to confirm there was a rise in violent crime nationally, the NCVS itself gives reasons to be cautious about such a conclusion:

The ability to compare 2016 estimates of crime to 2015 or other years was limited due to a redesign of the NCVS sample. In 2016, BJS introduced new areas to the NCVS sample to reflect population changes based on the 2010 Decennial Census and to produce state- and local-level victimization estimates, which will be released in early 2018. Among sampled areas that did not change, there was no measurable difference in rates of violent or property crime from 2015 to 2016.

Indeed, Jeff Asher at fivethirtyeight.com has an article explaining Why We Can’t be Sure If Violent Crime Is On The Rise based on the NCVS’s redesign and other reasons.  An article in the Marshall Project titled How Hard is it to Count Violent Crime by Yolanda Martinez also discusses how the NCVS's new methodology makes it difficult to measure crime.