May 01, 2003

Prof maps potential for criminal activity

In a less ominous version of Minority Report, a CSUS professor has found a way to help police officers predict who might commit what sort of crimes and where.

Criminal justice professor Donald Dixon, a former crime analyst for the Dallas Police Department, studied five years of arrest data for violent juvenile crime in Dallas. He found that when he combined modern crime mapping techniques and with low-tech, readily accessible census data, he was able to pinpoint not only where criminal activity was occurring but possible underlying causes. He studied areas where the offenses were happening, where the offenders live and where the victims live.

“I wanted to identify neighborhoods where violent juvenile crime was likely to occur but I was most interested in what it is about an area that may contribute to behavior,” Dixon says. “Working with the police department, I constantly got inquiries from people who wanted information about crime rates by zip code. But that information is meaningless because zip codes are arbitrarily configured and have no relationship to any idea of ‘community.’

“When juveniles live in an area long enough they’re influenced by the neighborhood – the same schools, the same peers. They’re not influenced by artificial boundaries like zip codes.”

Dixon mapped five years worth of juvenile arrests for violent crime in the city of Dallas and identified “hot spots” where the worst concentrations of this type of crime were found. Of the 3,600 arrests within those five years, one-sixth of the arrests were in just four neighborhoods which comprised less than two percent of the city’s geographic size. “They were that concentrated,” Dixon says.

He then compared the hot spots with data from the most recent census – number of households, number of people in the family, males vs. females, race, etc. Among the findings was that the two biggest hot spots differed in their racial makeup and the types of violent crimes that the juvenile offenders were committing. Both of these issues have implications for police officers working in the area.

“If I’m a good community-oriented policing police officer, I would want to know what it is about these kids that makes them likely to commit the offenses they are committing. From there I know what to do to plan an intervention,” he says.

Dixon found elements that emerged – such as income, number of two-parent households and education level – seemed to confirm what criminal justice professionals have long believed to be potential predictors of criminal activity. In the hot spots he found, for example, 49 percent of the residents had stopped their education at high school.

“That’s not a surprise but this is a new way of looking at it. I was able to pinpoint these areas with precision,” Dixon says. ”I now have a lot more confidence in previous findings.”

Crime mapping has become increasingly common nationwide. The Sacramento Police Department, for example, has a mapping program on its website that allows citizens see the location and type of crimes committed in a neighborhood such as Campus Commons or Midtown over a three-month period.

Other characteristics of the Dallas neighborhoods that also seemed to play a role were the presence, or absence, of parks, churches, liquor stores and bars. Hot spots tended to be near liquor stores and bars, Dixon says, which in Dallas are highly concentrated. They tended to not be near parks and churches.

Dixon plans to do follow-up work on adult crime and expects some of the same topics to arise. “Home ownership, education level, income are issues these kids are facing,” he says. “When I look at the adults, I will probably see an overlap.”

He is also interested in replicating this research in the Sacramento-Northern California area.