Columbus residents recently heard they live in the most dangerous city in America.
Residents of cities in Michigan, Illinois and New Jersey have heard the same thing.
The challenges of deciphering the label were apparent a few days ago, when a blog post titled "Five Most Dangerous Cities In America" started making the rounds on Facebook. It ranked Columbus No. 1.
Some people panicked. So much, in fact, that they failed to notice the list focused on property crimes and was published in 2012. Its author? LiveWatch.com/SafeMart.com, which sells security systems.
"Most dangerous" depends on how you define "dangerous."
"Many people probably don't consider property crimes to be the most dangerous crimes," said Kim Love-Myers, associate director of the University of Georgia Statistical Consulting Center. Property crime generally includes burglary, larceny and motor vehicle theft.
When it comes to violent crime, Flint, Mich., is widely listed as America's "most dangerous" city, based on Federal Bureau of Investigation data. LiveWatch.com/SafeMart.com has another list of "most dangerous" cities that includes violent crime. Flint tops that list, and there's no mention of Columbus.
Flint's title isn't absolute.
A 2013 post from Information Please Database online separates dangerous cities by population, offering three most dangerous cities: Camden, N.J., Flint, Mich., and Detroit. A list dated 2013 on the Curiosity Aroused website calls East St. Louis, Ill., the most dangerous city.
A 2012 Forbes story says Saginaw, Mich., is the country's most dangerous city for women.
And since the Internet isn't big on expiration dates, one search might yield a 2011 reference to a U.S. News and World Report story that calls St. Louis the most dangerous city. Or a 2009 Yahoo! Voices piece that says Cincinnati has the most dangerous neighborhood in America.
Still worried about Columbus? Shortly after a story about the LiveWatch.com/SafeMart.com list was posted at Ledger-Enquirer.com, another security company emailed another study. SimpliSafe has a blog post titled "FBI City Crime: 6 Most Burgled Cities, Is Yours One?"
Houston topped the list, which is dated 2012. Columbus didn't make an appearance.
Love-Myers emphasized the importance of analyzing research beyond its headline.
"Always make sure you understand the motivation of the source of your data," she said. "You need to read it, not just go with the big headline."
The numbers listed by LiveWatch.com/SafeMart.com require some disclaimers.
The LiveWatch.com/SafeMart.com post that ranks Columbus the No. 1 "most dangerous" city is dated 2012. It relies on the FBI's metropolitan statistical area property crime data from 2010 .
LiveWatch.com/SafeMart.com has a more recent list of "most dangerous" cities, which also focuses on property crime, that's dated 2013. Columbus ranks No. 5 on that list, which relies on FBI data from 2011.
Preliminary FBI crime statistics for 2012 suggest Columbus property crime has declined between 2011 and 2012. The preliminary numbers "indicate that when compared to data for 2011, the number of violent crimes reported by law enforcement agencies around the country increased 1.2 percent during 2012, while the number of property crimes decreased 0.8 percent."
Columbus Police Department data indicates property crime has declined in the city of Columbus -- from 13,219 in 2010, to 12,268 in 2011, to 11,234 in 2012.
The police department's numbers highlight another important point. When you're looking at local crime data, you need to ask if it's for an individual city, or the city's metropolitan statistical area. The FBI data used by LiveWatch.com/SafeMart.com defines the Columbus MSA as including Russell County, Ala., and Chattahoochee, Harris, Marion and Muscogee counties.
To make matters more complicated, Love-Myers noted laws can vary between places within an MSA -- adding confusion to the statistics. And, of course, these statistics only include reported crimes.
In addition, some people criticize the method of comparing cities per 100,000 people. It's better to find a city with a similar population and compare accordingly, they say.
Under that method, let's compare the Columbus MSA with the Macon, Ga., MSA, using 2012 Georgia Bureau of Investigation statistics . The Columbus MSA includes four counties and a total population of 249,500, according to the statistics. The Macon MSA includes five counties and a total population of 235,405.
The Columbus MSA had 17 murders in 2012, the statistics note. Macon's MSA had 26.
Columbus had 31 rapes in 2012. Macon had 61.
Columbus had 425 robberies in 2012. Macon had 305.
Columbus had 8,065 larcenies. Macon had 7,296. Macon also appears on the LiveWatch.com/SafeMart.com "most dangerous" lists based on property crime rates.
Outside Georgia, there are some cities with population numbers that compare with the Columbus population listed in the preliminary FBI crime statistics for 2012 .
Those preliminary statistics list a 196,178 population for Columbus, along with 994 violent crimes and 11,266 property crimes for 2012. According to the same preliminary 2012 statistics:
Little Rock, Ark., had a population of 196,055, and 2,579 violent crimes and 15,804 property crimes.
Amarillo, Texas, had a population of 196,576, and 1,278 violent crimes and 8,895 property crimes.
Akron, Ohio, had a population of 198,390, and 1,759 violent crimes and 10,034 property crimes.
(The population numbers "are FBI estimates based on provisional data from the U.S. Census Bureau.")
Something else to remember: Discrepancies in local and FBI crime statistics may result from human data input.
Challenges of comparisons
Even the FBI seems to warn against ranking cities based on crime statistics.
"Don't draw conclusions from the data by making direct comparisons between cities or individual agencies. Valid assessments are only possible with careful study and analysis of the unique conditions that affect each law enforcement jurisdiction," the FBI notes in a June 2013 news release accompanying the preliminary 2012 crime statistics.
Columbus Mayor Teresa Tomlinson was critical of the security company's references to Columbus, noting, "Columbus is the victim of random and disfavored statistical methodology."
She added, "Basically, it's a marketing ploy."
Brad Morehead, CEO of LiveWatch.com, the parent company of SafeMart.com, said he's aware of the discussions the "most dangerous" lists have sparked in Columbus.
"As a security company, we felt there was a void in the market for a focused discussion on property crime," Morehead said. "It's been great to see that discourse and discussion from the residents of Columbus."
So, is there any method of comparing cities' crime rates without excessive footnotes, complicated methodologies and debatable conclusions?
Go to Google, type "compare cities' crime stats," and click on the first link that appears: "Crime Rate Comparison."
It offers handy little spaces to type the names of the cities you'd like to compare. But then there's this clarification: "Based on the final 2006 FBI Crime Statistics."
Back to the drawing board.
Mike Owen and Tim Chitwood contributed to this report.