Category Archives: Crime Maps

Where is #HannahGraham? #JesseMatthew #MorganHarrington

I have been asked this question by a number of posters. So here is my answer, with the following caveats:

harrington-matthew-graham_606

1.  I don’t pretend to know where Hannah Graham is. This is just my opinion. And judging on what I’ve seen so far? The police / FBI know what they are doing.

2. I am about to break the cardinal rule: Don’t chase suspects. Having said that, this is more about chasing – what we have been told is – evidence, not suspects. So I will avoid what I observe a lot of sleuthies are doing: trying to pin every missing persons case / unsolved murder on Jesse Matthew.

3. To the parents of Hannah Graham and Morgan Harrington: I know what you are going through because I have been through it. My sister’s murder is an unsolved cold case. My family lived with her being missing for 6 months. Your approach in this matter is absolutely right. Bond with each other, trust no one. The families possibly connected in my sister’s case are very close (Prior, Monast, Dube). We trust no one. Even the brother of Louise Camirand, who prefers to keep relatively anonymous: we talk when necessary. And we warn each other of the nut-jobs. (For more on those cases, click here.)  

With those caveats, a few more:

1. I am a relatively sane person with a normal life. I have a 9 to 5 job. I have 3 daughters. My life is rich and fulfilled. I do not need attention. 

2. I am not a professional investigator. My experience comes from living with a 35 year old cold case, and 15 years of semi-professionally studying the patterns of murder. Kim Rossmo – who invented geographic profiling – is a friend; we worked together on profiling the serial murders in Quebec, and he is always available to me as a confidante / consultant.    Rossmo spent time at the Police Foundation in Washington, DC, and has consulted with the FBI at Quantico, VA: I would be surprised if they haven’t yet consulted on these cases.  

3. 15 years has gained me access to a lot of material; I have a library of  over 1,000 photos of Quebec crime scene photos, I have pretty much exclusive access to Quebec cold-case files (no, you can’t know my source). I have frequently been asked for advice on cold cases (for better or worse).  I sometimes “think” better than law enforcement on these matters… that is no-knock to law enforcement. As I said, I’ve got 35 years of experience.

4. And, one more time: I’m a dad. I couldn’t post about this until now because I was making dinner. But I think of these things because last night my eldest daughter was out at a concert until 12:30 AM: it’s hard not to think of these things.

So with that long introduction, here’s what I think about #HannahGraham / #MorganHarrington / #JesseMatthew:

1. I think police are targeting the right areas. I think Hannah Graham is West or South of Charlottesville.  I believe this based on where Morgan Harrington was last seen and where she was found (that trajectory), where Matthew was born and recently lived, and because the North and East are more urbanized. Matthews has experience with the West (Harrington), and the the West and South are more rural. Also, for what ever reason, I have observed over time that predators tend to hunt near their living environment, and dump South. That’s not statistically significant, just something I’ve noticed in my experience.

2. Someone asked me – if we presume Matthew is responsible for the murder of Harrington and the disappearance of Graham – were the dump sites pre-meditated?  Given those assumptions, I say No. If we take the account of Jesse Matthew’s assault on a driver to be true, he is very impulsive (consider the 1,300 mile run to Galveston).   So, a guy who reacts, then figures out what he’s going to do later. No premeditation, though on the prowl. An opportunistic predator. Though I would say he’s learned from experience: impulsive in the heat of the moment, but over time, he’s learned to put contingencies in place.  

Tags:

Interactive Map #JesseMatthew #HannahGraham #MorganHarrington

I’m not seeing very helpful maps, so I made my own (just click and navigate). These are the facts. I may add some more speculative points later:

map

I have other crime maps. Mostly for Canadian cases, and the Rocky Mount Serial Killer case here in North Carolina (click on the PAGES link above).

 

Tags:

Toronto-area victim identified as Guang Hua Liu

The victim in the Toronto-area human remains case has been identified  as 41-year-old Guang Hua Liu of Scarborough Liu was last seen alive on Aug. 10 and was reported missing by friends the following day.

 

Bizarre and disturbing. The first set of remains were found near Hewick Meadows Park in Mississauga, not far from the Square One shopping mall. The second set of what appear to be human remains were found in a bag near Kennedy Road and BonisAvenue in Scarborough. That’s about a 30 minute drive using the 401 and 403.

Here’s the latest:

Peel Regional Police say they have identified Guang Hua Liu, 41, of Scarborough, as the victim.
Peel Regional Police say they have identified Guang Hua Liu, 41, of Scarborough, as the victim. (Toronto Police Service)

Peel Regional Police say the remains found in Mississauga and Toronto in recent days belong to a single mother who went missing earlier this month.

Insp. George Koekkoek named the victim as 41-year-old Guang Hua Liu of Scarborough during a news conference Tuesday afternoon.

Koekkoek said Liu was last seen alive on Aug. 10 and was reported missing by friends the following day.

Liu, a Canadian citizen of Chinese descent, was a single mother of three.

Until recently, Liu was employed as “the owner of a now-defunct spa” on Eglinton Avenue in Scarborough, Koekkoek said.

Koekkoek said police have executed search warrants in their investigation.

“We are working on suspect information,” he said.

The investigation that led to Liu’s identification began when a human foot was found in the Credit River in Mississauga’s Hewick Meadows Park last Wednesday. Further remains were found nearby.

On the weekend, Toronto police were alerted to the discovery of remains on back-to-back days in the West Highland Creek.

While police had previously said there were “obvious similarities” between the finds in both cities, Koekkoek said Tuesday that police had “forensically linked” the recovered remains.

————————————–

A severed head, a foot and two hands discovered in Mississauga, Ont., last week and human remains found in Toronto on the weekend, are likely from the same victim, police said Monday.

Police say they can’t be 100 per cent certain, but there are “obvious similarities.”

Forensic scientists will make the final determination but Peel Regional Police said Monday morning their investigators believe the gruesome discoveries in the nearby cities over the past week are from the same victim.

“It is fair to say that … somehow parts of a person’s body have been discovered in two different locations,” said Const. Peter Brandwood.

On Saturday and Sunday remains were found in a creek in Toronto’s east end.

On Monday Brandwood confirmed those remains are human.

“Those remains that were discovered in Toronto have now been confirmed to be identified as human remains. Investigators are convinced that there are obvious similarities between the body parts in our investigation here and the human body parts and the recovery of those human body parts in the Toronto jurisdiction,” he said.

 

Tags:

The Serial Killer Ate My Homework

Watching some of these investigative reporters attempt to solve crimes gets as boring as watching American league baseball. No-one wants to single and bunt their way to victory, it’s all about the DH bases loaded home run, let’s hang it all to a serial killer and solve five crimes at once.

Israel Keyes

Take the case of Nancy West writing in a recent New Hampshire Sunday News article about murder suspect Israel Keyes. Keyes is being held in Alaska for the alleged kidnapping and murder of 18-year-old Samantha Koenig from an Anchorage coffee shop. Keyes is also apparently  a person of interest in the slaying of Essex, Vermont couple, Bill and Lorraine Currier, who were randomly abducted and murdered in June 2011 (apparently Keyes has told investigators the bodies could be found in a Vermont landfill).  The article (and apparently an impatient public, and capitulating law enforcement agencies) then attempts to tie Keyes to the disappearance and murder of Celina Cass, whose body was retrieved from a local river over a year ago, less than a quarter of a mile from her home. The evidence? Cass disappeared the month after the Currier murders.

Never mind that where the Curriers live in Essex, Vermont is a good five-hour drive on rough roads to where Cass disappeared in New Hampshire. Never mind that the psychological profile of someone who robs and kills a couple in their 50s is vastly different from someone who murders an 11-year-old. Investigators also note that Keyes owned a cabin near the Canadian border in Constable, New York. Let me put that in perspective for you; that’s three states, and over 300 miles. It’s like saying a person from Cornwall, Ontario is a suspect in a Sherbrooke, Quebec murder simply because there once was a penitentiary in Cornwall.

It gets better. The article also notes that the cases of Maura Murray and Louise Chaput remain unsolved in New Hampshire. 

Celia Cass

I’ll make this really easy for everyone. There is no evidence that Israel Keyes murdered Celina Cass (or Murray or Chaput). Cass was found a quarter of a mile from her house and was most likely murdered by a family member.

As I wrote about in my last post, through a long process of trial and error I have become a disciple of the least effort principle of Occam’s Razor. By all means keep your mind open for the unexpected, but also keep it simple, let the facts speak for themselves. The pressure and temptation to throw everything into some great unifying theory in criminal investigation is strong. I remember back in the Summer of 2005 I was working with NBC television to do a story for Dateline NBC on my sister’s murder. The producers were interested in exploring an angle between her case, and the then two new investigations into the twin disappearances of Briana Maitland and Maura Murray. The producers wanted myself and Geographic Profiler, Kim Rossmo to go on record and suggest that all the cases might be related, that their was a possibility that a serial killer had been operating across the American-Canadian border over a period of three decades. There was absolutely no evidence to support this theory. Rossmo explained that when establishing locus and territoriality in geographic profiling, the span of a serial predator quickly diminishes at a point of say, 30 miles. For someone to be operating in a playing field of several hundred miles is very rare, if not impossible. Some might cite Ted Bundy, but that was never really the case: Bundy travelled. In the case of the Green River Killings one of the major inhibitors to resolving that investigation was the temptation to tie too much together (to essentially make Gary Ridgeway and Robert Pickton one person). When we told the producers at NBC that there no evidence to support such a sensational theory they didn’t care. They wanted us to say it anyway.

Eventually we backed away from the Dateline story, and the producers were not interested in doing a show that stuck with the facts. I will admit that the temptation to give them what they wanted was there. Regardless if it was true, a Dateline story would have given my sister’s case International exposure. It could have led to information that could have solved the case. But the premise wasn’t true, it could have done more damage than good. And anyway, an American audience would do little to shed light on events of 3o-years-ago; what ultimately was needed was a program in the French language, produced for locals, by locals (which is ultimately was what we got).

In November 1999, 16-year-old Julie Surprenant disappeared from a Montreal bus-stop. Less than two years later, 14-year-old Julie Bureau went missing

Julie Surprenant’s father, Michel

from her home near Sherbrooke, Quebec. Then ten months later the body of 27-year-old Julie Boisvenu was found in a ditch near Sherbrooke. She had been raped, beaten and strangled to death. The press quickly tried to suggest that the cases were somehow linked. Their evidence? The girls were all named Julie. I’m not joking. I remember the La Presse headline, Les Trois Julies, and I myself got caught up in this hysteria. So what happened? Julie Surprenant was abducted and killed by serial offender Richard Bouillon who, on his prison death bed, confessed to a nurse that he killed her. Her body has never been found. Julie Bureau was a runaway who resurfaced three years later, apparently living under everyone’s noses in Sherbrooke.  Julie Boisvenu was murdered by Hugo Bernier, who is currently serving a life sentence. Bernier was a repeat offender, but not a hardened criminal like Bouillon.

This brings me full circle to the cases on Briana Maitland and Maura Murray. Both disappeared

Maura Murray

within a month of each other eight years ago. Both disappearances involved abandoned automobiles on lonely forested highways. Both were young, attractive women with their whole lives ahead of them. For years investigators, the media and the public have tried to link the cases. It took the first year and a half before investigators officially dismissed any connection, wasting valuable resources and time.

The cases are vastly different.

Murray appears to have been under numerous stressors that could have given her a reason to runaway. She may be living somewhere else, or she may have been in despair and perhaps died in the woods. Maitland’s disappearance seems to be linked to foul play. Friends and associates to this day are not talking. She may have been in the wrong place at the wrong time. Where Maitland’s investigation appears to have stalled, the Murray case has received fresh interest with the creation of a blog by investigative journalist James Renner  ( apparently to the dismay of the family unfortunately). Nevertheless, Renner appears clear-headed and dedicated to sticking to the facts of the case. I hope both cases soon find their resolutions.

Tags:

Pickton inquiry hears from serial killer profiler

Kim Rossmo was one of the first officers to warn that a serial killer could be responsible for the disappearance of women in Vancouver's Downtown Eastside.

Kim Rossmo was one of the first officers to warn that a serial killer could be responsible for the disappearance of women in Vancouver’s Downtown Eastside. (The Magazine of Simon Fraser University)

A renowned criminologist who warned the Vancouver Police Department that a serial killer might be at work while women went missing in the Downtown Eastside is scheduled to testify Tuesday at the missing women inquiry.

Kim Rossmo, a geographic profiler who was on the force at the time, was part of a working group formed as public pressure mounted for police to solve cases of missing sex workers.

Rossmo is credited as being among the first officers to warn about the possibility of a serial killer.

In 1998, he and another officer were preparing to issue a news release that said, in part: “The objective of this group is to determine if a serial murderer is preying upon people in the Downtown Eastside and, if so, what murders and disappearances are linked together.”

It would have marked the first time Vancouver police had publicly acknowledged the possibility of a serial killer, but just two weeks before the news release was scheduled to be issued, it was scrapped by the head of the force’s major crimes section and the working group was disbanded.

Systemic failures

In a brief address prior to the opening of the inquiry Tuesday, commissioner Wally Oppal compared the Pickton investigation to other serial killer cases including Clifford Olson, Ted Bundy and Gary Ridgway, known as Green River Killer.

Even though the cases spawned their own investigations and inquiries, Oppal said the same problems keep cropping up — issues of leadership, morale and resources within the policing community.

Oppal said he has to ask himself what he can do differently if previous reports failed to affect change.

Oppal said his final report will examine the systemic failures in the policing environment, including the relationship between police and the victims, and the failures in the organization itself.

Report due in June

Pickton wasn’t arrested until February 2002, five years after his name first surfaced as a suspect in the disappearance of sex workers in Vancouver’s Downtown Eastside, when officers showed up at his Port Coquitlam farm with a search warrant related to illegal firearms and stumbled upon the belongings and remains of missing women.

Pickton was convicted in 2007 of six counts of second-degree murder, but the remains or DNA of 33 women were found on his farm. He claimed to have killed a total of 49 women. He is currently serving a life sentence.

Rossmo, now a professor at Texas State University, invented a technique of tracking crimes that is used around the world. He was the first Canadian police officer to get a PhD in criminology.

The missing women inquiry, headed by Oppal, is examining why Vancouver police and the RCMP failed to catch Pickton in the late 1990s and early 2000s, and why prosecutors declined to pursue an attempted murder charge against him after an attack on a sex worker in 1997.

A final report is due by June 30, 2012.

Ahh the English Montreal media: A day late and a dollar short.

Man accused in murder of Montreal woman faces new charges

THE GAZETTE MAY 4, 2010

MONTREAL – The man accused in the murder of Natasha Cournoyer has been charged with attempted murder in connection with an attack on a sex worker.

Claude Larouche, a 48-year-old carpenter, is charged with first-degree murder in the killing of Cournoyer, 37. Her body was found last October in thick bushes in Pointe aux Trembles.

The woman told police Larouche beat her in her Hochelaga-Maisonneuve home Oct. 17, 16 days after Cournoyer disappeared from the Laval office building where she worked.

Larouche told the sex worker he didn’t want sex, but wanted to talk.

After taking some drugs, he allegedly hit the woman in the face and tried to strangle her. Larouche took off after the woman knocked his glasses off. Police identified him through a DNA sample found on the glasses.

Larouche is no stranger to the criminal justice system, with charges dating back to 1984 in Chicoutimi, Trois Rivières, Joliette, Quebec City and Montreal.

He’s being detained and is due back in court May 25.

Tags:

Rocky Mount Missing Women: What we already knew.

Two disappointments cloaked as victories this week (the other I’ll get to shortly)…

The first is the discovery of remains in Edgecombe County last Saturday that have now been positively identified as those of Roberta Williams. I have avoided commenting on this recent newsoid for fear of flogging the Rocky Mount Missing Women story into the ground. My contempt for how authorities have mishandled these cases is hardly a secret, so let’s just spell it out:

Blatant racism… 11 black people are murdered or go missing in an area the size of a postage stamp and for nearly a decade no one manages to give a tinker’s cuss about the matter. Yes, deja shades of Robert Pickton and the Vancouver downtown Eastside murders all over again. It only took the Olympic games for B.C. to recover from that tragedy, so what do you think is in store for the tiny impoverished East Carolina region of Rocky Mount? I will tell you: the trauma of endless fear, self-loathing and humiliation.

It is no balm that Rocky Mount chief of police has finally… glacially… come forward and stated what has been obvious to my five-year-old child all along:  “It’s clear that we are dealing with a suspected serial killer.”.

Thank you chief, you can go back to whatever busy work has occupied you for the last decade (perhaps there’s an abandoned vehicle that needs towing?). This week NC Wanted anchor Gerald Owens finally grew a pair and boldly asked of the chief, “how many more victims are there?”. Thanks for showing up Gerald, where have you been? This isn’t about giving your Kodak image the perfect frame for tragedy: this is a real story, with real families that are suffering: you should have been in the game years ago.

While we all sit and wait for this to play out (ya, as if it’s some kind of parlor game), the prime suspect, Antwan Pittman has been sitting in jail for 8 months. What are authorities waiting for? For a gun to literally smoke? Meanwhile victims’ families continue to be traumatized daily by the mistakes and missteps of an uncaring and insensitive media and justice system.

Let’s not forget that in the midst of this madness Newsweek got it right 5 months ago:

“For the families who just want to locate their daughters or bring closure to their murders, the investigation has been a long, drawn-out process. Tucker speaks about her daughter in the past tense, quickly catches herself, and shifts to the present tense, emphasizing her commitment to finding her daughter. “As far as the investigation goes, I just hope they continue to do the best they can to put closure to the missing girls and the girls that have been found,” Tucker says. “Whatever it is, we are here waiting.”

“Regardless of drug addiction or other problems, that still doesn’t give a person the right to kill another,” says Knight. “If we can give a terrorist a day in court, we can get these women justice.””

Tags:

NamUs Missing Person Database Goes Unused by 93 Percent of Law Enforcement

Is anyone surprised by this news?  No. Because we still have a police culture so set in its ways that they’d prefer to rely on memory, scratch pads and file boxes to solve problems when more than adequate tools are practically begging for utilization. Tools that could save lives:

PC News by David Murphy

Since 2009, families and medical examiners have had access to a free online database that’s designed to assist in the identification of more than 40,000 sets of unidentified remains across the country. Dubbed “NamUs,” short for the National Missing and Unidentified Persons System, the program allows both parties to enter identifying characteristics of a missing person or unidentified body in the hopes that this information exchange will help match a face to a fate.

It’s a grim consolation for those whose friends or families have been affected by violence or accidents. Nevertheless, the Associated Press reports that the free service has helped solved 16 cases since the cross-matching feature went live in July of last year. The numbers don’t end there: the service is home to around 6,200 unidentified sets of remains, 2,800 missing people, and–according to The Crime Report–has been accessed (on the missing persons front) by more than 185,000 people as of January 2009.

What’s the problem? According to the AP, only 1,100 of the nation’s 17,000 law enforcement agencies, or 6.5 percent, are registered with the service. That’s partly a publicity issue, as numerous law enforcement agencies simply don’t know the service exists. Others are more leery about using limited resources to participate in the service.

That doesn’t sit well with Janice Smolinski, sponsor of the “Billy’s Law” bill that aims to encourage wider use of the NamUs system. If passed–it’s already received House approval and remains pending in the Senate–the bill would generate $10 million in annual grants for law enforcement agencies to both train new users and help them resource the data entry process of adding new details to the system. The bill would also allow for an annual grant of $2.4 million to keep NamUS, as a whole, up-and-running.

As for how the system actually works, NamUs profiles are rated based on a one-to-five star system. A one-star profile contains scant details about a person: perhaps a name, or the location where they disappeared, but that’s it. A five-star profile is the whole kit-and-caboodle, with a full swath of details and identifying characteristics, as well as a picture or rendering of a person’s likely image.

According to The Crime Report, there’s currently no mandate that forces law enforcement to database details about a 21-or-over missing adult. Billy’s Law won’t change that aspect of the system, but it will allow the database to link up with the National Crime Information Center Missing and Unidentified Person File database in hopes that this could increase the detail of NamUS profiles (or, conversely, fill out the system with more.) Similarly, law enforcement will be required to submit missing persons reports for children (21-and-under) to the NamUs database.

For Smolinski, the legislative victory would be bittersweet. She remains confident that the NamUs database will give her the details she needs to close her own case–that of her son, Billy, who went missing in Connecticut in 2004.

Tags:

A suspect in the Louise Chaput cold case

Police have a suspect in the case of Louise Chaput, the Sherbrooke social worker who disappeared and was found murdered in the White Mountains of New Hampshire in 2001. NH police aren’t saying much other than that the suspect is male and lived in the NH region at that time.

There is DNA evidence from the crime scene that could link the suspect.

– TVA film footage here.

– Details on Chaput from the NH cold case website here.

Tags:

Well looky here: The NYPD’s crime data is in question:

Why am I not surprised. Bratton and Giuliani were crooks. I never believed in “broken-windows” theory, despite the fawnings it received in faux-social theory best sellers. Compstat was a pop fad, and the validity of the uniform crime index is a joke. Crime ebbs and flows, and law enforcement can do little to influence or deter it. But that never stopped Bloomberg or his predecessors from taken credit for its decline:

More than a hundred retired New York Police Department captains and higher-ranking officers said in a survey that the intense pressure to produce annual crime reductions led some supervisors and precinct commanders to manipulate crime statistics, according to two criminologists studying the department.

James Estrin/The New York Times

The retired members of the force reported that they were aware over the years of instances of “ethically inappropriate” changes to complaints of crimes in the seven categories measured by the department’s signature CompStat program, according to a summary of the results of the survey and interviews with the researchers who conducted it.

The totals for those seven so-called major index crimes are provided to the F.B.I., whose reports on crime trends have been used by Mayor Michael R. Bloomberg and his predecessor, Rudolph W. Giuliani, to favorably compare New York to other cities and to portray it as a profoundly safer place, an assessment that the summary does not contradict.

In interviews with the criminologists, other retired senior officers cited examples of what the researchers believe was a periodic practice among some precinct commanders and supervisors: checking eBay, other Web sites, catalogs or other sources to find prices for items that had been reported stolen that were lower than the value provided by the crime victim. They would then use the lower values to reduce reported grand larcenies — felony thefts valued at more than $1,000, which are recorded as index crimes under CompStat — to misdemeanors, which are not, the researchers said.

Others also said that precinct commanders or aides they dispatched sometimes went to crime scenes to persuade victims not to file complaints or to urge them to change their accounts in ways that could result in the downgrading of offenses to lesser crimes, the researchers said.

“Those people in the CompStat era felt enormous pressure to downgrade index crime, which determines the crime rate, and at the same time they felt less pressure to maintain the integrity of the crime statistics,” said John A. Eterno, one of the researchers and a retired New York City police captain.

His colleague, Eli B. Silverman, added, “As one person said, the system provides an incentive for pushing the envelope.”

The Police Department disputed the survey’s findings, questioned its methodology and pointed to other reviews of the CompStat process that it said supported its position.

The survey, which involved an anonymous questionnaire, was done in coordination with the union representing most of the senior officers in the department. The questionnaires were sent to 1,200 retired captains and more-senior officers; 491 responded, including 323 who retired from the department after 1995, the first full year that the agency, then under William J. Bratton, used CompStat. It is based on the scrupulous tracking of crime complaints and a mix of mapping crime trends, identifying trouble spots and holding precinct commanders directly responsible for attacking those problems.

The survey has its limitations. It is unclear exactly when the retired senior officers left the department, making it impossible to say whether any alleged manipulations came early on or had developed over years and across more than one mayoral administration. The CompStat approach has been widely replicated across the country and has been credited with improving police work in many cities.

Also, the questionnaires did not set out to measure the frequency of any manipulation. None of the respondents were asked to identify specific acts of misconduct, and none admitted to having done it themselves. In addition, it was unclear whether the officials who said they were aware of unethical conduct had firsthand knowledge.

But the survey asked provocative questions and clearly elicited disturbing answers. The retired members of the force were asked whether they were aware of changes to crime reports. Of the 160 who indicated that they were, more than three-quarters said the changes were unethical.

Paul J. Browne, the Police Department’s chief spokesman, who was provided a copy of the survey’s summary Thursday, said that two other significant, independent and more comprehensive studies had been done in recent years analyzing the integrity of the city’s crime statistics — one in 2006 by a New York University professor and another by the state comptroller’s office — and that he had found them to be reliable and sound.

The report by the N.Y.U. professor, Dennis C. Smith, contained this assessment: “We conclude, as did the state comptroller five years ago, that the city and department officials, and the public can be reasonably assured that the N.Y.P.D. data are accurate, complete and reliable.”

The researchers in the new survey emphasized that the responses — the questionnaires were mailed in September 2008 and returned in early 2009 — showed that most of the senior officers believed that CompStat had been a valuable management innovation. And even few department critics would seriously dispute that the city is much safer than it was in the early 1990s, with murders cut by nearly 80 percent and with neighborhoods, from the notoriously violent to the largely affluent, transformed.

The CompStat system was put in place by Mr. Bratton, Mr. Giuliani’s first of three police commissioners. Versions of the system have been franchised to hundreds of police departments. It was adopted, and in some cases modified, by Mr. Bratton’s successors under Mr. Giuliani, Howard Safir and Bernard B. Kerik, and by Mr. Bloomberg’s commissioner, Raymond W. Kelly.

But as the city annually reported reductions in crime, skepticism emerged in certain quarters — several police unions other than the one that assisted with this survey, elected officials, residents in some neighborhoods — about whether the department’s books were being “cooked.”

Concerns over crime statistics are not unique to New York. Police departments have faced accusations of tampering in Atlanta, Baltimore, Dallas, New Orleans and Washington.

Mr. Kelly, for his part, has said that he instituted a rigorous auditing system to maintain the integrity of the crime reporting operation. And Mr. Browne said Friday that every precinct’s books were audited twice a year, “and where errors are discovered, they are corrected and reflected in revised crime statistics.” He added, “In cases where it is determined that the errors were the result of intentional manipulation, the personnel responsible are disciplined.”

Mr. Browne said that Mr. Kelly had meted out discipline in 11 cases, 4 involving precinct commanders. One of them, he said, was demoted and three others lost their commands. Last week the department confirmed in an article in The Daily News that it was investigating whether the commanding officer in the 81st Precinct in Brooklyn downgraded crimes or refused to take complaints from complainants to artificially reduce serious-crime statistics.

Mr. Browne criticized numerous aspects of the survey, suggesting, for instance, that many of the respondents might simply have been repeating what they had heard or learned from news reports about the “relatively rare instances that gained notoriety.”

“The survey’s biggest flaw is that a hundred respondents may be recalling the same lone incident everyone was talking about when they said they knew of instances when crime reports were manipulated,” he said. “Further, anonymously supplied answers are problematic because it’s hard to assess whether they originate from retirees who felt they were unfairly denied promotion or have some other ax to grind.”

Mr. Browne said that only 37 of the 323 retired senior officers surveyed had served as precinct commanders, arguing that only they would have firsthand CompStat experience. But the researchers said the survey included responses from aides to precinct commanders and higher-ranking officers who oversaw the work of the commanders.

Professor Eterno said the suggestion that 100 former officials might be talking about the same incident was “ludicrous,” and Professor Silverman said the department’s criticism of the use of an anonymous survey indicated a limited understanding of social science methodology.

The seven-page summary of the survey certainly indicates that many of the retired officers believe the system has gone significantly wrong.

Indeed, the researchers said the responses supported longstanding concerns voiced by some critics about the potential problems inherent in CompStat. The former officers indicate that it was the intense pressure brought to bear on the commanders of the city’s 76 precincts in twice-weekly CompStat meetings — where they are grilled, and sometimes humiliated, before their peers and subordinates, and where careers and promotions can be made or lost — that drove some to make “unethical” and “highly unethical” alterations to crime reports.

Mr. Browne said that when Mr. Kelly took over the department in 2002, he barred spectators from CompStat meetings in light of complaints from some commanders that they had been ridiculed in the forum in front of outsiders. He said Mr. Kelly believed that the presence of outsiders “demeaned the process and was unprofessional.”

The two researchers are writing a book scheduled for publication this summer based in part on the survey; it is tentatively titled “Unveiling CompStat: The Naked Truth.” They provided a copy of the summary and the survey questions to The New York Times. They declined, however, to provide a full report until the head of the union with which they worked had shared it with the Police Department.

When Professor Eterno retired as a captain from the Police Department in 2004, he was working in its crime analysis and program planning section. He is now the director of graduate criminal justice studies at Molloy College on Long Island, which financed the study. Professor Silverman wrote a book about CompStat in 1999 before retiring from theJohn Jay College of Criminal Justice in 2003.

Roy T. Richter, president of the Captains Endowment Association, which represents the retired officials, said the challenges that his retired members had faced — and his active members still face — were significant, as crime continues to decline and precinct commanders must continue to beat their previous year’s performance despite a force with thousands fewer officers.

He called the survey results “troubling,” and said that while CompStat can be an effective tool, to the extent that it is “used as a sword to subject a commander to humiliation before his peers, I don’t think it’s an effective management tool.”

More than a year before Professor Smith of N.Y.U. published his study praising CompStat in 2006, a city commission created to monitor the Police Department’s effort to fight corruption sought to examine the integrity of the department’s statistics. But while the department cooperated with the professor, it refused to comply with the commission.

And despite the efforts of its chairman, Mark F. Pomerantz, a respected former federal prosecutor, the commission could not win subpoena power, and it was not able to examine allegations that crime complaints were downgraded.

The department had argued that those allegations did not fall under the panel’s mandate because the matters did not constitute corruption.

Tags: