It's not quite the stuff of science fiction, but Penn’s Jerry Lee Center of Criminology does seem to be taking a page from Steven Spielberg’s “Minority Report”: Working to identify potential murderers before a homicide can occur.

The Center’s new Homicide Prevention Unit, a joint initiative with the First Judicial District of Philadelphia, has put Penn’s acclaimed scholars to work in identifying those individuals in Philadelphia most likely to commit a murder. Through the use of advanced data-mining and risk-analysis methods, Center experts have helped authorities sort through more than a half-million probation cases in Philadelphia.

Now, a select group of specially trained probation officers—some of whom will likely be trained through Penn’s M.S. program in criminology­—will focus their efforts on these high-risk individuals and, hopefully, keep them on the right side of the law.

“If we can identify those who are more likely to commit crimes, what we can do is try to stop them,” says Lawrence Sherman, director of the Jerry Lee Center. “It’s not a matter of locking them up—they haven’t committed a crime yet. But you have to ask them to turn their lives around.”

The program was launched with $1 million in funding from city, state and private funding, with the Jerry Lee Foundation contributing $500,000.
Sherman says the initiative wouldn’t have been possible if Penn hadn’t recently recruited Professor of Criminology and Statistics Richard Berk.
A former UCLA statistics professor, who has used statistical methods to track everything from crime to weather, Berk was the driving force behind the initial data-mining operation, says Sherman.

“He’s worked in the field of crime and justice for more than 30 years now,” Sherman says. “He was the person who was able to take these 500,000 probation cases and analyze them … We wouldn’t be doing this if it wasn’t for recruiting Professor Berk to the project, and then to Penn.”

Berk helped authorities sort through 519,783 cases dating all the way back to 1969—the largest database of criminal records ever reviewed for the purpose of predicting future crime. From that huge initial sample, the Penn team identified several hundred on which to focus their efforts.
Such simple steps as providing adequate access to treatment for alcoholism, drug addiction, depression or other conditions could go a long way to helping these individuals toward a better life, Sherman says.

“Now we’re into a new phase, where if we can identify who is more likely to commit these crimes, we can figure out what to do to stop them,” Sherman says. “We can talk to them about their lives and how we might help them. We don’t tell them they’re on a list. That’s counterproductive.”
Sherman admits the list isn’t foolproof. But even if only a few of the high-risk individuals heed the advice of their probation officers, the program could prove to be a valuable tool in Philadelphia’s ongoing fight against crime.

“We aren’t necessarily claiming there’s a big difference between anyone who is on the list and those who aren’t,” Sherman says. “There’s a margin of error we’re aware of. But while you can’t do something intensive for 55,000 people, you might be able to do something intensive for a couple hundred.”

The real emphasis here is on helping people.”

Originally published on September 7, 2006.