Emphasis on crime numbers spurs debate

Once a month on the top floor of Nassau police headquarters in Mineola, commanders from all precincts, units, bureaus and squads take turns stepping into a crucible.

The commanders' crime-fighting successes and failures are beamed for all to see. On a projection screen in the front of an auditorium are the raw crime numbers and statistical percentages that can help make or break a lifelong police career.

Each commander steps to the front to justify why crime is up or down. The chief of department, the force's highest-ranking uniformed officer, presides over what can be a grilling for the unprepared. "If you're not showing any improvements, it'd get ugly," said one former precinct commander who attended these meetings for years and spoke on condition of anonymity.

Meetings like Nassau's occur in departments across the country and are a point of contention in policing: While some officials and experts think the meetings help hold commanders accountable, others argue they place too much emphasis on statistics and encourage manipulation.

Suspicions raised

It was at July's meeting where Nassau police brass heard Thomas DePaola, the commander they had appointed just weeks earlier to oversee the Fifth Precinct, boast of considerable drops in grand larceny and overall crime in that short time, officials have said.

So a team of chiefs was commissioned to audit DePaola's statistics and discovered that he'd been falsifying 18 or so months of crime figures -- not just at the Elmont-based Fifth Precinct, but mostly at the Manhasset-based Sixth Precinct, which he'd commanded just prior.

Now, about 170 case reports, filed under DePaola's supervision between January 2011 and several weeks ago, are being reclassified, officials have said. DePaola has since been demoted and stripped of his command.

Nassau's monthly meetings, called Nass-Stat, are modeled after a New York Police Department crime-mapping and statistics program called Comp-Stat.

The NYPD's Comp-Stat meetings are credited by some with bringing down crime, and notorious to others for scenes of browbeating top bosses berating commanders who fail to perform.

Nass-Stat, according to interviews with current and former commanders, is more collegial and less combative than New York City's. But Nassau's can still be grueling for a presenter with bad crime numbers and no plan to turn them around, those commanders say.

"If you're making an attempt, things were good," the former commander said. "If you were making feeble attempts, then you were called on the carpet."

John Eterno, a former NYPD captain and now a Molloy College professor who co-wrote a study this year concluding that the NYPD has a statistics-manipulation problem, says Comp-Stat and Nass-Stat encourage even well-meaning officers to falsify numbers.

"These pressures from the top can cause even a saint to do corrupt and illegal things. When you put that pressure on a human being -- 'You will do this, You will do that, otherwise you won't have a career' -- the pressure is very great to make changes," Eterno said.

The study, which anonymously surveyed thousands of retired NYPD cops, found that crime-stat manipulation was part of the fabric of the department. Eterno and a colleague, Eli Silverman, wrote a book: "The Crime Numbers Game: Management by Manipulation."

The NYPD disputes the study's conclusion and methodology.

No 'undue pressure'

Supporters of Comp-Stat, and its progeny such as Nass-Stat, say pressure to reduce crime is exactly what holds police managers accountable.

"If it's pressure to do your job, yes. Undue pressure? I don't think so," said James Lawrence, the Nassau police commissioner from 2002 to 2007.

Lawrence helped pioneer Nass-Stat after serving in New York City as a Comp-Stat expert and chief for the NYPD.

"Naturally when you're aware of things that are going on, you feel pressure to address it," he said. "That's your job. That's what you're getting paid for."

Heather Mac Donald, a scholar at the think-tank the Manhattan Institute for Policy Research, disputes that the programs encourage statistical manipulation. If anything, she says, they instill "a sense of life-or-death urgency about fighting crime."

"The question is, are we better off doing without those accountability mechanisms, and I would say no," she said.

Nassau's first deputy police commissioner, Thomas Krumpter, said that as part of the DePaola audit, the chiefs scrutinized more than 3,000 crime reports from across all precincts and found all but DePaola's to have been properly classified.

Nassau said DePaola kept his numbers low by downgrading grand larcenies to petty larcenies, classifying theft cases as "lost property" and omitting from statistics crimes where victims requested no arrests.

A message left at DePaola's home was returned by the department's chief spokesman, who said DePaola, on the job 27 years, would have no comment.

Experts praised how the department handled the scandal, including demoting DePaola two ranks, to captain from inspector, on orders of Nassau Police Commissioner Thomas Dale.

"I'm impressed with the fact that they saw it and they apparently acted quickly," said Samuel Walker, a professor emeritus of criminal justice at the University of Nebraska-Omaha and an expert on police accountability.

DePaola's alleged distorting methods are typical of the way others have manipulated statistics around the country, experts said.

Eterno said the distortion is especially prevalent with agencies that employ a Comp-Stat model. "We found it all over the world -- we've documented it in London, New South Wales, Paris, Philadelphia, Broward County, Baltimore. I could go on and on," Eterno said.

Rate of misclassification

An acceptable misclassification rate is between 2 percent and 4 percent, said Daniel Bibel, a past president of the Association of State Uniform Crime Reporting Programs.

DePaola's rate had been 12 percent, police said.

It's impossible to quantify how often crime statistics are falsified, criminologists say.

"We have no idea how common it is because a whole bunch of cases may have occurred that we don't find out about," Walker said.

Nationally, recent cases of crime-stat fraud have been more egregious and systemic:

The Milwaukee police admitted in June that 20 percent of aggravated assaults -- about 5,300 cases since 2006 -- had been misreported as lesser offenses.

Baltimore city cops agreed to reform rape investigations after a 2010 report showed that they consistently kept rape statistics low by bullying victims and improperly classifying many legitimate rapes as "unfounded."

The Federal Bureau of Investigation, to which more than 18,000 U.S. police departments are legally required to submit their crime data, audits only a tiny fraction of submitted data, making it important for individual municipalities to have internal checks and balances in place, experts said.

"Let's look at the bright side here -- that Nassau took it upon themselves and said, 'These numbers don't sound right, and we're going to investigate this,' " Bibel said.