Monday, January 23, 2012

Why Do Reporters Sometimes Have Such Trouble With Crime Stats?

As a crime analyst I have a professional interest in news articles about crime stats even when they aren't about the city where I work. When the article is about the city where I work, I have a vested interest in them as I often have provided the stats that the article refers to.

Yesterday afternoon, I was checking articles from the local media outlets when I found one from Philip Jankowski over at the Killeen Daily Herald. The piece titled: "Killeen homicides rise in 2011" looked at homicides that occurred in Killeen in 2011. I have worked for the Police Department in Killeen for nearly 21 years so this was going to be article that covered a topic that I was very familiar with. In the story there was a statement I found very troubling:

"Aside from the Luby's massacre in 1991 when 26 people were killed at once, the last time Killeen had 15 homicides was in 1979.

Since the city had about one-third of its current population more than 30 years ago, the homicide rate nearly tripled in 2011. There were 12 reported homicides in 2010."

Source: http://www.kdhnews.com/news/story.aspx?s=63745 as retrieved 1/22/2012 at 5:06 PM CDT.

The part that troubled me was that the statement that homicide rate in Killeen had "nearly tripled in 2011" as compared to 1979 was flat out wrong. Not just a little bit off, but completely and totally wrong. In this case, I provided the numbers that were ostensibly used to write that part of the article. Let's look a little deeper at these numbers.

In 1979 the Killeen Police Department reported 15 murders to the FBI's Uniform Crime Reports (UCR) Program. In 2011, the department also reported 15 murders to UCR.

Obviously, the population of Killeen has changed from 1979 to 2011. In order to directly compare these numbers we need to calculate the murder rate.

Rate is the number of times an event occurs in a given population. In the case of crime rates, they are normally calculated as the number of crimes reported per 100,000 population. This allows us to determine if a crime occurs more or less often in given population even if the original population figure is different from the later.

You can think of crime rate this way, if you were to line up 100,000 residents of a city, the rate would tell you how many of them had been victim of a particular crime in a given year.

To calculate rate the formula looks like this:

(Number of Crimes / Population) x 100,000 = Crime Rate Per 100,000 population

In 2010, the population of Killeen was reported to be 127,921. In 1980, the population was reported to be 46,292. We'll have to use these population figures to calculate the rates because they are the closest available numbers to the years we are interested in. (As a side note, sometime around May of 2012 the Census Bureau will release a 2011 population estimate for Killeen. When those numbers become available I'll recalculate our crime rates.)

To calculate the murder rate for 1979 the formula would look like this:

(15/46,292) x 100,000 = 32.4 murders per 100,000 population

To calculate the murder rate for 2011 the formula would look like this:

(15/127,921) x 100,000 = 11.72 murders per 100,000 population

Clearly we see that the murder rate in 1979 is nearly three times higher than the murder rate in 2011. (If you're interested it works out to be a 63.82% decrease in 2011 as compared to 1979.) This is the exact opposite of what Jankowski reported in his story.

Now when a reporter gets crime statistics wrong in a story, there are a few possibilities about what might have happened. One, the reporter may not have understood what he or she was trying to report. Another possibility is that he or she might have been careless about his or her reporting. The other is that the facts are being deliberately being misrepresented.

In this case, the story indicates that the murder rate is much higher in 2011 as compared to 1979. However, the truth is that a citizen of Killeen was around three times more likely to be murdered in 1979 than he or she would be in 2011.

This is a huge problem. The public's perception of crime is influenced by what they see reported in the media. This perception often has significant consequences. If a city gets a reputation as having a high crime rate, how many new businesses are likely to relocate there? How, many people are going to look at that community as a place they want to live and work (and buy newspapers)?

This is one of the reasons I am passionate about accurately reporting crime statistics. When accurately counted and properly used, these statistics can tell you a lot about your community and the effectiveness of your crime fighting efforts. As I am often fond of saying: "In order to know how to get where you are going, you have to know where you have been." And you can quote me on that.

1 comment:

  1. On Monday evening after the error was pointed out to the reporter who wrote the Killeen Daily Herald story referenced above, they changed the offending paragraph in the online version to read:

    "Since the city had about one-third of its current population more than 30 years ago, the homicide rate was one-third of the 1979 rate. There were 12 reported homicides in 2010."

    Of course, the damage done by the print version is already done. If someone read the print version do you think they go to the website to check and see if there were any changes?

    I have not seen a retraction of the original story printed anywhere.

    ReplyDelete

I reserve the right to remove defamatory, libelous, inappropriate or otherwise stupid comments. If you are a spammer or are link baiting in the comments, a pox be upon you. The same goes for people trying to sell stuff. Your comment will be deleted without mercy.