A photo of a 2024 event hosted by The Trace in Philadelphia. Trace staffers in the foreground and a person wearing a Trace backpack in the background.
Trace staffers at a 2024 event in Philadelphia. Meredith Edlow for The Trace

The Trace’s Gun Violence Data & Best Practices Guide

It is difficult to find gun violence data, and the data that does exist is often incomplete and unreliable. In this guide, we share data sources trusted by Trace staffers and examples from our archive to demonstrate how information can be turned into local and national stories.

If you know of any good data source that should be included — or you’re a newsroom interested in having gun violence data localized for your area — please send us a message at [email protected].

Also, be sure to check out our U.S. Gun Violence 101 sheet for an overview of the staggering numbers and key topics on gun violence that many Americans might not know.

By Brian Freskos and Ava SasaniPublished December 12, 2024

Table of Contents


Common Mistakes and Misconceptions

Mass Shootings

News consumers are understandably horrified and invested in the details of mass shootings, yet these large-scale tragedies account for a small fraction of the overall number of gun deaths in the U.S. In 2021, nearly 50,000 Americans died of gunshot wounds. Roughly 1% of those deaths were the result of mass shootings. Suicides regularly make up about half of all gun deaths, while an overwhelming portion of the other half are homicides of people of color.

Further, there is no consensus on the definition of “mass shooting.” Gun Violence Archive, among the most important data resources for covering this beat, defines “mass shooting” as an incident in which four or more people are shot, not necessarily killed, not including the shooter. The Department of Justice in 2013 considered a mass shooting “any incident in which at least four people are murdered with a gun.” Meanwhile, the Violence Prevention Project Research Center defines mass shooting as four or more people shot and killed, excluding the shooter, in a public location, with no connection to underlying criminal activity, such as gangs or drugs. For more information, see our story The Way We Define Mass Shootings Affects How We Respond.

Duped by Data

Data, especially crime data, can be easily misrepresented. Let’s say a politician, activist, or source is sharing data analysis about gun violence. Here’s a quick checklist for validating their findings:

  • Is the data examining a narrow time frame? Gun violence reached record peaks in 2020, during the height of the COVID-19 pandemic. If a politician wants to suggest that crime is down, they might look at the change in shootings between 2020 and 2023, citing the sharp decline in gun violence. But is gun violence really down, or is it just normalizing to pre-pandemic levels? Inclusion of a wider time frame will often show the latter.

  • Correlation vs. causation: Let’s say a state saw a decrease in fatal shootings between January 2017 and January 2018. That state’s governor might claim that the drop in shootings is thanks to a bill they signed into law. The police union could claim that shootings are down because they increased recruitment efforts, and more cops on streets leads to less violent crime. A social services nonprofit could say the drop in shootings is thanks to an increase in public housing, reducing the economic pressures that make gun violence more likely.

    In a vacuum, all three claims seem plausible. In reality, it is difficult to tease apart the reasons why gun violence dips or spikes. It’s rare to be able to accurately draw a causal link between the rate of shootings and a singular policy change.

    Remain skeptical of sources that claim credit or responsibility for changes in the gun violence rate.

  • Be wary of any data analysis that cites totals when a rate is more appropriate. Politicians and other partisan actors frequently use raw data about gun violence to portray American cities as dangerous epicenters of gun violence. The total number of gun violence incidents might be higher in certain urban areas, but that’s often because cities are more densely populated. A recent analysis by The Trace found that Selma, Alabama, had a higher per capita shooting rate than Chicago.

In 2016, The Washington Post wrote about Governor Mike Huckabee’s claim that U.S. police officers shoot more white people than Black people.

“U.S. police officers have shot and killed the exact same number of unarmed white people as they have unarmed black people: 50 each. But because the white population is approximately five times larger than the black population, that means unarmed black Americans were five times as likely as unarmed white Americans to be shot and killed by a police officer.”

Huckabee was technically correct, but comparing raw numbers is statistically meaningless, you need to account for population size.

Respect the people behind the numbers

Data only gets you so far, especially on the gun violence beat. Remember that data is not a substitute for talking to the people who are impacted by gun violence. The data we provide might indicate a particular city saw a decline in gun fatalities in 2022. That does not necessarily mean parents feel safe having their children walk to school in the morning. Today’s Texas gun violence data does not capture how the parents of children killed in the Uvalde massacre rallied together to launch a long-term campaign to reform the state’s gun laws. Chicago’s overall gun violence data indicates the city is becoming safer. Closer inspection reveals that four of Chicago’s 17 neighborhoods experienced a record spike in shootings, even as the city’s overall gun violence rate drops.

What makes gun violence newsworthy?

A 2019 shooting in Brownsville, Brooklyn – a majority Black neighborhood in New York City – forever altered 11 lives and ended one more. The fatal incident received scant media attention. The families of the victims were left to mourn alone, without the outpouring of public sympathy and support that typically follows large-scale shootings. Just two of six leading national newspapers devoted space on their websites’ front pages to the Brownsville shooting for more than a few hours in the days that followed, according to an analysis by The Trace.

Journalists who want to cover this beat need to confront uncomfortable yet common racist biases about what kind of gun violence deserves media coverage. Too often, shootings in wealthier, whiter parts of the country are covered with a reverence befitting tragedies. Shootings in places like Brownsville are shrugged off as an intractable problem, the inevitable reality of living in a place that is Black, poor, and urban.

^ Top

Have a suggestion for other resources? Use the help desk.

Help Desk

Data Sources

Bureau of Alcohol, Tobacco, Firearms and Explosives

Listing of Federal Firearms Licensees

Data on gun stores, wholesalers, manufacturers, and other types of federally licensed gun sellers in the United States

The data is updated monthly and is useful for analyzing state and national trends. It also includes each seller’s phone number, so you contact them for interviews.

Trace Coverage

^ Top

U.S. Firearms Trace Data

Statistical reports with information about guns used in crimes

Every year, the ATF traces hundreds of thousands of crime guns recovered by local, state, and federal authorities. While the ATF is legally barred from publicly releasing most trace information, the agency does publish statistical reports showing how many crime guns were recovered and traced in each state, the states in which those crime guns were first purchased, the types and calibers of those crime guns, and time-to-crime rates, among other information. You can use these reports to analyze gun trafficking trends.

Trace Coverage

Pro Tip: FOIA ATF Inspection Records

After the ATF inspects a gun store, the agency produces a report detailing the violations and resulting penalties. You can file a Freedom of Information Act request with the ATF for copies of these reports. If you want to see an example of an inspection report, check out this database The Trace published in 2021. It contains nearly 2,000 inspection reports that our reporters analyzed for our Off Target project in partnership with USA Today.

^ Top

International Firearms Trace Data

Statistical reports about crime guns recovered in other countries

The ATF conducts traces on behalf of law enforcement agencies in Mexico, Canada, Central America, and the Caribbean. You can use these reports to analyze trends associated with gun trafficking from the United States to those countries and regions.

Trace Coverage

^ Top

Firearms Compliance Inspection Results

Data on ATF gun store inspections

The ATF inspects gun stores and other federally licensed gun sellers and publishes data about the results online. This data is updated monthly and shows how many inspections resulted in the agency revoking a gun store’s license to operate.

Trace Coverage

^ Top

Annual Firearms Manufacturers And Export Reports

Data on firearms manufactured and exported by federally licensed gun sellers in the United States.

AFMERs include the type and number of firearms produced by each manufacturer. You can use this data to analyze manufacturing and exporting trends nationwide or by state. Each AFMER is intentionally published with a yearlong delay to protect industry secrets.

Trace Coverage

Other Recommend Reading:

^ Top

Firearms Commerce Reports

Statistical reports with historical data about the gun industry

Firearms Commerce Reports are good for analyzing trends in gun manufacturing, imports and exports, and federally licensed gun sellers. You can, for instance, look at the change in the number of gun dealers nationally and by state. These reports also contain national and state-level data about weapons registered under the National Firearms Act, a federal law governing civilian possession of machine guns, silencers, short-barreled rifles, grenades, and other weapons deemed by Congress to pose an acute threat to public safety. These reports used to be published annually. However, it appears that they were quietly discontinued. The last report was published in 2021.

Trace Coverage

^ Top

Federal Firearms Licensee Statistics Theft/Loss Reports

Data on guns lost by or stolen from gun dealers and other federally licensed firearm sellers

These reports include the type of guns that went missing and the type of theft (i.e., robbery, burglary, or larceny. For definitions, see here). The numbers are also broken down by state, so if you’re a local reporter, you can analyze trends in your area.

Trace Coverage

Pro Tip: Local Police Records

When gun dealers suffer a theft, they are required to report the incident to both the ATF and their local police department. You can file a public records request for the police report to learn more details about the theft and obtain contact information for potential sources. In 2019, The Trace used information from police reports to piece together how dozens of firearms stolen from North Carolina gun shops wound up being used in shootings, robberies, and other crimes in multiple other states.

^ Top

National Firearms Commerce and Trafficking Assessments

In 2021, the Biden administration ordered the ATF to issue annual reports on gun trafficking, something it hadn’t done in two decades. The agency has issued three such reports as of November 2024, linked below. These reports are chock full of useful data pertaining to gun manufacturing, ghost guns, machine gun conversion devices, trafficking investigations, and a host of other gun-related topics. In their second report, the ATF released more detailed gun theft and crime gun trace data for specific states, cities, and counties.

National Firearms Commerce and Trafficking Assessment: Firearms in Commerce – Volume One

National Firearms Commerce and Trafficking Assessment (NFCTA): Crime Guns – Volume Two

National Firearms Commerce and Trafficking Assessment (NFCTA): Firearms Trafficking Investigations – Volume Three

^ Top

Centers for Disease Control and Prevention

WONDER database

One of the best resources for data on gun deaths

WONDER is compiled using death certificate information that states pass along to the CDC. You can use WONDER to analyze gun deaths by type (homicide, suicide, unintentional shooting), age of victims, as well as state and region.

Trace Coverage

^ Top

Injury Statistics Query and Reporting System

Data on gun-related injuries, but accuracy is suspect

Be careful not to confuse WONDER with the CDC’s Web-based Injury Statistics Query and Reporting System, or WISQARS, a database that tracks gun-related injuries. Unfortunately, the gun injury estimate is one of several categories of CDC data flagged with an asterisk indicating that, according to the agency’s own standards, it should be treated as “unstable and potentially unreliable.”

Unlike WONDER, which calculates the number of gun-related fatalities based on death certificate data, WISQARS is based on a survey of 100 U.S. hospitals, known as the National Electronic Injury Surveillance System, or NEISS. Of these, 66 — less than 2 percent of all hospitals in the United States — are used to collect additional data that includes gunshot wounds. For comparison, the emergency room injury database produced by the Agency for Healthcare Research and Quality surveys nearly 1,000 hospitals. This small sample size is part of why Trace staffers and some leading public health researchers are skeptical of the accuracy of WISQARS estimates. There have been some attempts to improve CDC nonfatal injury data.

Trace Coverage

^ Top

Federal Bureau of Investigation

Background Checks & Gun Sales

Statistics on the number of background checks conducted on prospective gun buyers, a proxy for gun sales

The FBI runs the National Instant Criminal Background Check System, or NICS. This is the system used to run background checks on prospective gun buyers. The bureau publishes national and state-level data on the number of checks conducted each day, month, and year. It also releases data on checks that resulted in prospective gun buyers being denied a sale because of felony convictions, mental health adjudications, or other records that legally disqualify people from owning firearms.

Trace Coverage

With no national database of gun sales, NICS is one of the best tools available for estimating the number of firearm purchases and analyzing trends over time. Background checks do not correspond one-to-one with sales, but in most cases, background checks for firearms transfers are conducted by licensed firearms dealers at the point of sale. As a result, you can use these transfer checks to produce estimates of gun sales, taking into account the share of transactions that include multiple guns. You can learn more about this by checking out the Trace Coverage and Other Resources below.

Trace Coverage

Other Resources:

^ Top

Uniform Crime Reports

Data on reported crimes — including homicides — in the United States

There is reason to be skeptical of this data because it is based on voluntary submissions from local law enforcement agencies, and not all of them participate. Still, the data can be useful for exploring trends in violent crime. UCR also collects information about killed and assaulted law enforcement officers and police use-of-force. You can analyze the data using the FBI’s Crime Data Explorer.

Trace Coverage

^ Top

Federal Judicial Center

The Federal Judicial Center, the education and research agency for the U.S. federal court system, publishes a comprehensive and interactive spreadsheet of every civil and criminal court case dating back to 1970. The FJC calls this dataset the Integrated Database. It enables users to filter cases according to circuit, district, case disposition, nature of proceeding, and nature of offense. The nature of offense category can be used to filter for gun-related charges, giving users a view of all the defendants charged with particular gun crimes, the disposition of those cases, and other related information like docket numbers. This data is released on a quarter time lag.

Trace Coverage

^ Top

Gun Violence Archive

GVA is a nonprofit organization that tracks gun injuries and deaths in the United States. GVA gathers information about gun violence via news coverage, social media, and police reports. That collection methodology has obvious limitations – it tends to drastically underestimate nonfatal gun violence incidents that don’t make headlines. Still, GVA is one of the best data sources we have for understanding U.S. gun violence in near-real time.

You can use GVA data to analyze gun deaths and injuries, mass shootings, the use of assault weapons in shootings, road rage shootings, shootings at or near schools, and other topics.

Trace Coverage

^ Top

International Trade Commission

The U.S. International Trade Commission tracks data on firearms, firearm accessories, ammunition, and other weapons imported into and exported from the United States. This data can be used to provide information about how many firearms have been exported abroad, particularly to countries engaged in conflict or unrest. The data is updated regularly with a lag time of about three months. It is accessed through dataweb.usitc.gov, for which an account is required. The following HTC codes are relevant for civilian firearms; however, additional HTC codes can be used for other weapons, including some military weapons:

Trace Coverage

^ Top

Medical Examiners and Coroners

Local medical examiner and coroner offices can be great resources for data on gun deaths. This data can provide insights into the age, race, and gender of victims and can be filtered for specific subcategories of gun death (e.g., homicide, suicide, accident). You can use it to analyze, for example, the percentage of suicides that involved guns, the race of gun homicide victims, or the median age of accidental shooting victims in your area.

Trace Coverage

^ Top

National Center for Education Statistics

The NCES publishes statistics on guns and violence at schools. This includes data on guns brought to campuses, gun access among students, and hate crimes. In June 2024, The Trace used NCES EDGE geographic datasets — which contain geocodes for public, private, and postsecondary schools nationwide — to analyze how many shootings occurred near K-12 campuses.

Trace Coverage

^ Top

OpenSecrets and the Federal Election Commission

Want to know how much gun control or gun rights groups spend on lobbying and elections? You can find the answer at OpenSecrets, a great resource for political finance data. OpenSecrets has data on federal campaign financing back to 1979, lobbying to 1998, and personal financial disclosures to 2004. The group even allows you to make a custom data request, though payment is required. Information on election spending is also available from the Federal Election Commission. You can use OpenSecrets and FEC data to look at how groups’ spending has changed over time, how much money they have given to particular candidates or spent opposing particular candidates, and even the amount they’ve spent on vendors.

Trace Coverage

^ Top

Police and Sheriffs

Local police departments and sheriff’s offices can be great sources of local gun violence data. Here are a few specific aspects of gun violence that this data can help analyze:

Shooting Deaths & Injuries

Police departments and sheriff’s offices compile data on shooting deaths and injuries in their jurisdictions that is often more detailed and reliable than the federal government’s. You can file a public records request for this information, though many police departments now publish at least some of it on their websites. Check out Jeff Asher’s YTD Murder Comparison dashboard, which contains links to many police departments’ online crime data portals.

Trace Coverage

^ Top

Stolen and Recovered Guns

Earlier in this guide, we covered data on firearms stolen from gun stores and other federally licensed gun sellers (see Federal Firearms Licensee Statistics Theft/Loss Reports). But there is another category of gun theft: Guns stolen from people’s homes, cars, and other locations. These thefts also get reported to local police, and many departments can provide the data in spreadsheet form, allowing you to analyze the types of guns that were stolen, the types of locations they were stolen from (a home, car, or business, for example), and other angles.

Local police also maintain data on guns that they have recovered at crime scenes. You can request this information to explore the most common makes and models of firearms used by criminals, or whether there has been an increase in the recovery of assault weapons, high-capacity magazines, or even ghost guns. You can also compare it against stolen gun data to determine whether any pilfered firearms turned up at crime scenes.

In 2017, The Trace, in partnership with NBC TV, used stolen and recovered gun data to identify more than 23,000 firearms recovered by police between 2010 and 2016. The vast majority of those guns were recovered in connection to crimes, including carjackings and kidnappings, armed robberies at stores and banks, sexual assaults and murders, and other violent acts committed in cities from coast to coast.

Below are a few of the spreadsheets we obtained for that project, which we called Missing Pieces. We also published a guide on how to investigate the relationship between gun theft and violent crime in your community:

Pro Tip: Key Fields for Stolen and Recovered Gun Data

Police departments store and collect data differently. When requesting stolen and recovered gun data, there are a few key fields that you should try to get.

Case Number
Each stolen and recovered gun typically has a case number assigned to it. If there is a specific gun you’re interested in and want to find out more details about the incident in which it was involved, you can use the case number to request the police report.

Property ID
A unique identifier assigned to each gun. This helps make sure you don’t count the same gun more than once in your analysis.

Date
The date on which each gun was reported stolen or recovered. It’s useful for analyzing trends by month and year.

Make
Essentially the brand of the gun, e.g., Smith & Wesson, Taurus, Colt, etc. You can use this to determine what makes are being recovered most often in your area. It also helps if you want to identify stolen guns that went on to be used in crimes (see Serial Number below).

Model
Just like Honda has the Accord and Toyota the Highlander, gunmakers give names to the different models they produce. Smith & Wesson makes the Bodyguard 2.0 pistol, for instance. Like make, model can also be useful for identifying stolen guns used in crimes.

Serial Number
Gun manufacturers are legally required to put serial numbers on their products. However, these serial numbers are only unique to the manufacturer. That means Smith & Wesson and Colt can each produce a firearm with the same serial number. So if you’re looking at your data and you see a stolen gun and a recovered gun with identical serial numbers, you can’t say for certain whether they are the same firearm unless they also share other identifiers, such as make and model.

Caliber
The diameter of the bullet that the gun is designed to fire, e.g., 9mm, .40, etc. Gunshot wounds from medium- to large-caliber rounds are significantly more likely to kill victims than wounds from smaller-caliber bullets. You can use caliber to determine whether the guns used in crimes in your community are getting deadlier.

Type
Broad groupings of firearms based on shared characteristics, like pistols, rifles, and shotguns. Many departments use National Crime Information Center codes to designate types. For example, the value “P” indicates the weapon is a pistol, “R” indicates a rifle, and “S” a shotgun. More weapon type codes can be found in the NCIC manual.

Category or Description
Could be thought of as the subtype. This field typically captures functional characteristics or specific action types, like semiautomatic or automatic. Agencies often use NCIC codes for this field as well, and sometimes join them with the type codes. For example, the value “PI” indicates a pistol (P) that is semiautomatic (I).

Capacity
The number of rounds the firearm and its magazine can hold. In 2017, The Trace analyzed the capacity of guns recovered in Baltimore and found that criminals were increasingly using so-called high-capacity magazines despite a ban on their sale.

Offense
The crime associated with the firearm. This can be helpful in figuring out whether a gun was stolen from a car or some other location, or whether a recovered gun was used in a homicide versus a robbery.

Trace Coverage

^ Top

Transactional Records Access Clearinghouse

TRAC, a project of Syracuse University, is a source of data on the federal government’s staffing, spending, and enforcement activities, including federal prosecutions. The Trace has used TRAC data to report on crackdowns against domestic abusers who unlawfully keep guns, unlicensed private gun sellers, and other gun offenders. TRACFED, which requires a paid subscription, allows for more granular and custom data analyses.

Trace Coverage

^ Top

The Violence Prevention Project Research Center

Formerly The Violence Project, The Violence Prevention Project Research Center at Minnesota’s Hamline University is a tremendous source of data on homicides at K-12 schools, college and university campuses, and houses of worship. The center also collects data on the trends, locations, circumstances, and characteristics of U.S. mass shootings and the firearms used in those shootings, including how the firearms were obtained. For more information about how the center defines mass shooting, see the Mass Shootings section below.

Trace Coverage

^ Top

Washington Post Police Shooting Database

In 2015, The Washington Post launched a database to record every person shot and killed by an on-duty police officer in the United States. As of October 2024, the database contained more than 10,000 fatal police shootings and can be downloaded on The Post’s website. You can use this data to analyze trends in the number of police shootings over time; the age, race, and gender of victims; and how many shootings were mental illness-related. The Post also includes links to its sources so you can find more details about an individual shooting.

Trace Coverage

^ Top

Think we missed something? Reach out!

Contact Us

Public Records Requests

The Trace’s Data Hub would be happy to assist you in crafting public records requests about gun violence data – just create a ticket using our help desk. Before you begin the process of a records request, check to see what data is currently available in your city or state. In many larger cities like Philadelphia and Atlanta, law enforcement and city officials publish a treasure trove of valuable gun violence data online.

When crafting a public records request, here are some guidelines and best practices to keep in mind:

  • It’s notoriously difficult to get public records from the ATF. That doesn’t mean you shouldn’t try, but be prepared for a long and difficult process: Federal data shows the ATF has one of the worst track records when it comes to producing public information.

  • Check if the data is already out there. Before you begin to request records about gun violence in your community, be sure to check in with the Data Hub staff: We might already have the records that you’re looking for.

  • Be specific. When requesting records, be sure to include as much detail as possible about the information you’re interested in.

  • Get ready to clean. Gun data is complicated and any records you get back will likely be filled with errors. You will need to clean the data before doing your analysis.

  • Get familiar with the agency you’re requesting from. Many law enforcement agencies across the U.S. have publicly available guidelines, handbooks, and other manuals dictating their rules for data entry. These rules can help hone your request. Remember that you want to limit the work of the FOIA officer handling your request. You can use these data entry instructions to find key terms that are easily searchable in police records, raising your chances of success.

^ Top