A Critical Gap in Gun Violence Data

In order to effectively study anything, researchers need accurate, comprehensive data. That’s true with infectious diseases. It’s true with environmental threats, and with drug addiction. And it’s true with gun violence. 

Unfortunately, we have a critical gap in the data available on gun violence: there is no standardized data on non-fatal shootings collected on the local level. No studies of state or federal gun laws have examined the impact of those laws on the number of people shot, because the data on nonfatal gunshot wounds over time and jurisdiction aren’t available. And many studies of local initiatives to curb gun violence rely upon data that don’t record non-fatal criminal shootings.

Currently, fatalities are the only direct measure of local-level gun violence involving people being shot in a criminal act. Local health departments record such data from death certificates and are mandated to report that data to the CDC. The CDC makes county-level mortality data available online for specific causes of death, including homicide, though it does not include data when there are few deaths reported in a jurisdiction during a given reporting period. The CDC also makes available national estimates of people treated in hospitals for nonfatal traumatic injuries, including gunshot wounds,  but hospital-based surveillance systems that track nonfatal gunshot wounds at the local level are rare. There has been considerable investment to expand and improve data on fatalities – for example, the CDC just recently announced new grant recipients to expand its National Violent Death Reporting System to all 50 states – but there hasn’t been a similar focus on improving our ability to gather non-fatal shooting data.

There is a potential solution to this data gap: The FBI could add non-fatal shootings to its Uniform Crime Reporting (UCR) program. 

The FBI has maintained the UCR program since 1930. Though it is a voluntary reporting system, the vast majority of local law enforcement agencies participate in the UCR by sending monthly statistics on certain crimes known as Part 1 offenses. These include murder and non-negligent manslaughter, rape, robbery, aggravated assault (an assault that results in serious injury to victim or involves the use of a deadly weapon), burglary, motor vehicle theft, larceny-theft, and arson. 

For murders, non-negligent manslaughters, and legally justifiable homicides, local jurisdictions throughout the U.S. report UCR data that include the type of weapon used in these fatal incidents. While current UCR reporting methods for robberies and aggravated assaults typically include information on the type of weapon used, reporting on firearm-involved incidents does not distinguish which cases involved victims with gunshot wounds from incidents in which victims were shot at but not hit, or from incidents in which someone uses a firearm to threaten another person without firing it. 

These distinctions are important, and not only because the outcomes differ greatly. Crimes in which no victim is shot or seriously wounded are vulnerable to inconsistencies in reporting. Nearly all studies of the effects of policies designed to reduce violence are based upon comparisons across place and time.  However, statistical analyses estimating the impact of policies rest on assumptions that there are no systematic errors in outcome measures of gun violence.  

For a crime to be recorded in the UCR, typically someone must call the police, and police must respond to that call, determine that a crime had been committed, and complete necessary reports. Crimes committed with firearms that do not result in gunshot wounds and occur in areas in which gun violence is relatively common are less likely to be reported to and recorded by the police than those occurring in areas where gun violence is less common. Consistency in reporting is also likely to differ between periods when gun crime is trending high and more peaceful times, as political pressures stemming from high rates of gun crime create incentives for police to under-report crimes in which there are no serious injuries to victims.  

Some might argue that findings from studies of lethal gun violence are sufficient because whether a shooting is fatal is due, to some degree, on chance. But studies reveal that local trends for fatal and nonfatal gun violence can differ, and interventions can influence fatal and nonfatal shootings differently. That may be true for the effects of state gun laws as well if, for example, certain gun laws are more effective in preventing shootings stemming from spontaneous altercations or conflicts, which are more likely to be nonfatal, than in preventing premeditated acts to kill, which are more lethal.  

By adding non-fatal shootings to the UCR as sub-categories of robberies and aggravated assaults, the FBI would facilitate local tracking of gun violence that agencies can use to combat gun violence. It would also allow for much richer research on the epidemiology of gun violence and more accurate evaluations of interventions to reduce gun violence.