FBI crime data undercounts killings in Chicago, other big cities, analyst says

The sizable drop in killings in the FBI’s national crime report comes with an asterisk as poor data collection methods saw homicides be undercounted in some major cities, according to analysts.

The FBI reported a nearly 12% decline in homicides throughout the country in its 2023 Crime in the Nation report. However, those official numbers didn’t include 27 uncounted killings in Baltimore, an extra 50 slayings that took place in Dallas and a whopping 118 additional homicides in Chicago that were part of the Major Cities Chiefs Association’s data.

The discrepancy was highlighted by the Law Enforcement Legal Defense Fund, an advocacy group for police officers and public safety causes.

According to the MCCA data, Chicago ended last year with 617 killings, compared to the 499 documented by the FBI.

Dallas saw 292 killings, while the FBI recorded 242. Baltimore suffered 260 homicides in the MCCA data, but 233 in the FBI's numbers.

Sean Kennedy, the policy director of the Law Enforcement Legal Defense Fund, said the discrepancy is not the product of anything “nefarious or conspiratorial” on the FBI’s part.

Instead, he said there’s a “fat finger” problem where the local police departments aren’t reviewing the data they send to the federal agency, and the FBI isn’t pressing the departments to clarify their submissions when they come across anomalies.

He said the FBI’s defensiveness about the data’s quality only creates further distrust in the agency’s representation of national crime trends.

“It breeds a sense of paranoia that someone is cooking the books, when they're not cooking the books — they're just not following a recipe very strictly,” Mr. Kennedy said.

The Washington Times reached out to the FBI for comment.

Outside of some lethargy on the part of government agencies, Mr. Kennedy said another challenge is the FBI’s reliance on the National Incident-Based Reporting System (NIBRS).

The relatively new method, which was first mandated in 2021, is far more intensive than most local departments require for their record keeping.

The analyst said it creates further disparities because short-staffed police departments — and the people they serve — prefer to have officers on the street instead of having them crunch numbers behind a desk.

“The average person would rather have police solve the crime or prevent the crimes that could be committed than produce high-quality data about the time of day that a burglary happened in the southwest side of a city on Tuesday,” Mr. Kennedy said.

Not all the FBI’s miscounts are underselling violent crime.

Federal numbers recorded nearly 36,000 aggravated assaults in New York City last year, while the NYPD documented roughly 28,000.

Mr. Kennedy said the difference is because the FBI is counting multiple offenses within one incident, such as a street brawl where one person attacks three people, while NYPD is likely only counting that as a singular assault.

He said that level of specificity is a credit to NIBRS but can also be a challenge when trying to collect accurate crime numbers.

The analyst said one solution is for the feds to consider awarding these departments grants so they can dedicate more resources to gathering good data without taking cops off patrol.

Another solution is for the FBI to consult with these departments and determine which parts of the NIBRS standards are getting in the way of sending reliable data.