Atlanta ranks 25th in crime ranking

Atlanta may be the city too busy to hate. But for crime, we make time. 

The city ranks 25th of 400 cities in what's widely described as the annual "Most Dangerous Cities" list. St. Louis is No. 1.

Last year, Atlanta was 18th. So while we still rate among high-crime cities such as Detroit and Baltimore, we're safer than we used to be.

Or are we? In terms of crime, is Atlanta's really getting better? And was it that bad to begin with?

We looked at the same 2009 FBI data used by the crime ranking. These numbers predate high-profile attacks in the Virginia-Highland neighborhood and elsewhere in recent weeks. 

Violent crimes reported to or uncovered by Atlanta police dropped from about 14 percent between 2008 and 2009. Property crime fell by about 12 percent. Crime is dropping nationally, too, but more slowly, the data said.

The problem is this: The FBI warns against ranking cities. Doing so leads to "simplistic and/or incomplete analyses that often create misleading perceptions adversely affecting communities and their residents," its website says.

CQ Press' list has been at the center of the ranking controversy.

Until recent years, Morgan Quitno Press, a small Lawrence, Kan., publisher, released the ranking along with a thick book of crime statistics. The publisher called it the "Most Dangerous Cities" list.

Although there's little consensus among criminologists on how to compare crime between cities, leaders of the American Society of Criminology agreed on this: The ranking deserved condemnation.

The group passed a resolution in 2007 that called it "invalid, damaging, and irresponsible."

"They fail to account for the many conditions affecting crime rates, the mismeasurement of crime, large community differences in crime within cities, and the factors affecting individuals' crime risk," it read.

What's worse, Morgan Quitno kept its methodology secret. Criminologists had no way of verifying or analyzing the results.

That same year, CQ Press, which is known for printing authoritative material on government and policy, bought Morgan Quitno. It responded to criticism by making big changes, including hiring Rachel Boba, a Florida Atlantic University criminologist.

CQ dumped the ranking's "Most Dangerous Cities" name. It also revealed its methodology.

A city's crime score is based on murders, rape, robbery, aggravated assault, burglary and motor vehicle thefts reported in cities with populations of 75,000 or more.

CQ computes the differences between city and national crime rates for each offense, then basically adds them together to determine a city's score.

A higher score means higher crime. This year, Atlanta's score is 168.53. First-place St. Louis scored 381.62.

Although this is a very basic type of analysis, it can be useful, Boba said. Since it compares city rates with the national average, it can help agencies flag issues that need further scrutiny.

When combined with other statistics in the 388-page book, the rankings are also a useful entry point for people who want a snapshot of a city's crime, she said. 

But these numbers need to be used very carefully, Boba stressed.

The ranking's old name was incorrect. It does not show whether a city is "safe" or "dangerous," she said, or that a resident of one city would be safer if he moved elsewhere.

A person's risk of becoming a victim often depends on whether he frequents high-crime neighborhoods and how he spends his time, Boba said.

Richard Rosenfeld, former head of the American Society of Criminology and a professor with the University of Missouri-St. Louis, goes further. Some city streets are as safe as suburban ones, and what city you're in has little to do with whether you become a crime victim.

"You learn next to nothing of a person's risk for becoming a victim of crime by what city they're in," Rosenfeld said.

Although the book addresses this issue, the news release for the 2010 ranking does not expressly discuss the reason for the name change or warn media against calling the ranking the "Most Dangerous Cities" list.

CQ's ranking also does not say why crime is higher or lower from one community to the next. Those issues are "beyond the scope of this book," as Boba wrote in its introduction.

Plus, cities with different rankings may have very similar crime rates. Atlanta would have dropped off the Top 25 entirely if its score were just four points lower.

The list has other limitations.

We discussed CQ's methodology with Rosenfeld and Georgia State University professor Robert R. Friedmann, who studies ways to improve crime statistics.

Friedmann, who has been critical of rankings, said that CQ's general approach has been around for some time. 

It's not invalid, but specialists are dissatisfied with it. The CQ rankings report is often misused or misconstrued and does not tell a city what it needs to do to lower crime.

"I cannot say what they say is true or not true," Friedmann said. "The question is whether what they do is meaningful."

The ranking weighs auto theft equally with homicide, Rosenfeld notes.

Also, many cities would fare much better or worse if their boundaries included larger or more representative portions of the metro area.

Rosenfeld and Friedmann issue their own ranking, which adjusts a city's homicide rate for socioeconomic factors such as age, the number of single-mother heads of household and poverty rate that are common in neighborhoods with high crime.

Their ranking shows whether a city's homicide rate matches expectations and tells policy-makers what social problems they need to tackle to help prevent violence. 

To sum up:

CQ's analysis is neither invalid nor incorrect. Using its math, Atlanta does rank as 25th among cities with the highest scores for crime rate.

But as one of its authors acknowledges, the number does not mean Atlanta's residents are in more danger than those in other cities. She and other experts emphasize that a person's safety is more dependent on his lifestyle and what neighborhoods he frequents than the city he's in.

Although the list's name was changed, the book addresses this point while its news release did not. And the message has not stuck with the public.

Because the news release does not tackle the list's misuse head on, we conclude the statement meets AJC PolitiFact Georgia's definition of Half True. It was accurate but left out important details or took things out of context.

Featured Fact-check

Threads posts

stated on October 22, 2024 in a Threads post

“The cheat has begun” because “Michigan, Georgia and Pennsylvania have officially announced that they will not have election results available on election night.”