Series of LAPD Statistical Snafus Dismays Observers

By JIM NEWTON

July 24, 1996 12 AM PT

TIMES STAFF WRITER

In days gone by, the Los Angeles Police Department pioneered the use of statistics in police work.

Hand tallies of arrests, major crimes and traffic citations were delivered to the chief every Wednesday, a tradition started by Police Chief William H. Parker, who ran the LAPD from 1950 to 1966. Crimes were plotted on pin maps. Response time was clocked religiously. Strategies were gauged by their effect on the statistical bottom line.

And woe to the aide who got a number wrong. Veterans recall the “DS” memos that would fly through the department, sailing from the chief’s office to the unfortunate underling who goofed. The “D” stood for dumb; the “S” was even less flattering.

So it is with wonder and dismay that some LAPD insiders and others take stock of today’s Police Department and its inability to add up and interpret a column of numbers. A series of recent counting slip-ups has raised questions about the reliability of Los Angeles’ most important and controversial agency, one whose performance is measured largely by statistics.

Without good numbers, many observers say, it may be all but impossible to monitor the Police Department and hold its leaders accountable for progress in everything from how many officers are on the beat to how many arrests are on the books.

“The word ‘accountability’ contains the word ‘count,’ ” said Merrick J. Bobb, a Los Angeles lawyer who monitors reform at the LAPD and the Los Angeles County Sheriff’s Department. “If you can’t count right, how can you hold someone accountable?”

Or, as former Chief Daryl F. Gates put it: “Once you give out bad data, your credibility is shot.”

In the past few months, the LAPD miscounted arrests time after time, then amazed police commissioners and many department insiders with its convoluted attempts to explain away the issue. The department produced one set of statistics purporting to detail civilian complaints against cops, then a second set of numbers, then a third. When asked for copies of one type of police misconduct files, the department identified 78 of them, then counted again and came up with 148. Subsequent checks have produced still more.

And in the space of a few months, department officials told one questioner that the LAPD’s Spanish-language curriculum consisted of 85 hours of training, then told another the number was 83 hours and then informed a third that the real number was 89.

“Well,” Police Commissioner Edith Perez finally exclaimed, “which is it?”

Some LAPD leaders recognize their problems with statistics, and the department has wrestled with possible solutions. Police Chief Willie L. Williams says an automated payroll system would help, as would an integrated departmentwide computer system. The LAPD’s Office of Operations has proposed adding staff to its Information Resources Division, and the department has expanded its Crime Analysis Unit.

Those things take money, however, and both Mayor Richard Riordan and Williams are primarily focused on increasing the number of officers on the street. So for now, the troubles continue. And so do the department’s credibility problems.

The most striking example of that--and the issue that has raised the most serious questions lately about the LAPD’s competence--is the department’s inability to say with clarity how many people it arrests every year and to provide consistent explanations of why that number is going up or down.

That saga began in March, when The Times reported that arrests had dropped off from 289,000 in fiscal year 1990 to 189,000 in fiscal year 1995. Traffic citations, field interviews and case clearances also had plummeted, according to statistics compiled by the LAPD as part of its annual budget request.

Although the Police Department produced those numbers, LAPD officials were stung by their disclosure, and the anxiety of top officials was further exacerbated when Riordan demanded explanations. The mayor said he wanted a response within 90 days, and the LAPD used that whole time preparing a detailed report, illustrated with cartoon characters, charts and graphs. The final product was sent to the Police Commission over Williams’ signature.

But for all its bells and whistles, the new report distinguished itself more for its questionable reasoning and factual errors. It produced a new year-by-year tally of arrests, one that inexplicably contradicted material the department supplied to the Police Commission’s special counsel just months earlier.

Based on those suspect numbers, the report argued in one breath that it was good that arrests were declining because that was the fallout from a shift to community-based policing, then simultaneously reasoned that a recent increase in arrests also was good because it was evidence that officers were becoming more productive.

The report said the department was handling its statistical analysis well, then asked for more staff to help with statistical analysis. And it argued that officers should be praised for their productivity at a time of increasing calls for service, only to make clear a few pages later that calls for service were down, not up.

The report even coined a new LAPD euphemism: In describing various sets of statistics produced by different parts of the department, the report called some “more accurate” than others.

Those and other issues troubled members of the Police Commission. Acting President Art Mattox complained that the report “raises more questions than it answers.” And others voiced concerns about the department’s compilation of the statistics, as well as its analysis of them.

“If we’re getting different answers when we go to different sources, that’s quite troublesome,” said Police Commissioner Raymond C. Fisher. “It’s not only embarrassing. It’s a problem.”

Perez agreed.

“How are you going to run an organization when you don’t know what’s going on with any precision?” she asked. “You base decisions on the information you gather. The decisions you’re making have an impact on society as a whole. And those decisions are not going to be very good if the underlying information isn’t reliable.”

Without reliable information in other reports, Perez said she is left to question the LAPD’s oft-cited crime reductions. “If this department consistently cannot count,” she asked, “how do we know that the crime statistics are accurate?”

Nor are the problems limited to material that the department releases for public consumption. When a pair of blue-ribbon consulting groups, Blue Marble Partners and Decision Management Associates, set out to study the LAPD’s inner workings, they ran into statistical black holes again and again. The department didn’t keep some statistics, others were off, and in some cases, different parts of the LAPD produced contradictory numbers.

“There was a lot of frustration on our part,” said Bill Allison, who managed that project for the two consultants. “They don’t have the systems to produce that kind of information.”

In the field, police officers have their own set of complaints about the department’s antiquated systems for collecting data.

For evidence of that, they say, look no further than how the department takes roll.

Every LAPD supervisor in every police station has a small, brown, hardcover LAPD time book. Every day, those supervisors flip open their books, take out their pencils and mark who is present, who is out and why. Periodically, an area timekeeper comes by and transfers those hand notations into a ledger, then transfers that information onto crew sheets. Those sheets are forwarded to the department’s Fiscal Operations Division, along with overtime slips. Every transfer is by hand, every step in the process time-consuming and fallible.

It can take officers a month--or more--to get a check for the overtime they worked.

“In most cases, it can take weeks for us to get paid [overtime], and sometimes, when it’s a departmentwide situation, it’s been months,” said Sgt. Jeri Weinstein, a 13-year LAPD veteran. “But if, by accident, they pay you too much, the bean counters are right there. There’s absolutely no delay at all.”

At Parker Center, department officials point to antiquated systems such as those and say it is no surprise that some mistakes get made. In fact, officials offer a mixture of explanations and excuses for the statistical errors.

In the case of counting the police misconduct files known as “miscellaneous memos,” documents that resolved internal police investigations for 1994 and 1995, LAPD leaders contend that there was a misunderstanding between Internal Affairs and the special counsel to the Police Commission. The slip-up was innocent, they say, and no effort was made to hide files.

Some of the LAPD’s data collection and analysis problems are byproducts of its woeful technological state, department officials and others agree. Many areas of the LAPD lack even basic computers, and though that has been partly remedied in recent years thanks to a private fund-raising campaign spearheaded by Riordan, the technological gap has created problems throughout the LAPD.

Also, the recent computer acquisitions have created some problems as they have solved others. Many of the computers don’t talk to each other, so aides in various offices enter their own data and don’t check each other’s work.

One result: Employees in different parts of the department compile their own statistics in part because they don’t know whether anyone else is doing it. Today, no fewer than six LAPD employees tabulate homicide statistics. That’s a waste of money and time, officials admit, and it creates multiple chances for internally inconsistent statistics.

If computers and human miscommunication are responsible for some of the counting snafus, others are harder to explain. Officials admit that they have no ready explanation for why the department has given out contradictory material on Spanish-language training at the Police Academy. (The actual training, according to the officer who runs it, takes 85 hours. At least, that’s close. “Eighty-five hours is pretty accurate,” said Sgt. Ruben Padilla.)

Similarly, when Williams announced to considerable fanfare last year that the LAPD had compiled a list of more than 100 potential problem police officers and was monitoring them, City Council members were pleased. But police union leaders were troubled. Who was on the list, they wanted to know. And what did it mean that department brass was “monitoring” them?

Faced with questions such as those, LAPD officials admitted the next day that no such list existed, that fewer than 40 officers were singled out for special attention, and that even they didn’t get much in the way of additional monitoring.

Meanwhile, some of the department’s counting woes may say more about asking cops to be accountants than anything else. Take the LAPD’s unfortunate experiment in facilities planning. Police officers helped draft a bond measure that voters approved in 1988, one that was to pay for construction of a host of new stations and other Police Department improvements.

The trouble with that? The spending plans forgot to take into account the full cost of land. Projects were delayed and inflation whittled away at the money. Finally, construction is underway and three new police stations will soon be in business, but the measure’s high expectations fizzled into a more modest game plan.

“Nobody on the Police Department,” said Cmdr. Carlo Cudio, “had any insight or knowledge or expertise in capital improvement projects.”

While that problem stands as an embarrassing reminder of past mistakes, even the department’s most stalwart defenders acknowledge that in at least one key area, the LAPD’s command of numbers remains distressingly thin. That area: the size and deployment of the police force.

To find out where his officers are deployed at any given moment, Williams, chief of the nation’s third largest and arguably most closely watched police department, has no computer buttons to punch, no printouts to consult, no automated payroll system to tell him who’s out there.

Instead, he has to have his staff start making calls. Commanding officers at the stations have to pass those requests on to subordinates, who can get some answers from station computers but have to turn to their little brown books for others. To get the overall list takes hours, with no real promise of accurate results.

How many police officers are at work right now in Los Angeles?

Williams doesn’t know. Nobody does.

(BEGIN TEXT OF INFOBOX / INFOGRAPHIC)

How Many Arrests?

The Los Angeles Police Department recently compiled statistics on arrests, but its numbers vary from one report to the next. In a report to the Police Commission and mayor, it said the most reliable statistics are those from its Information Resources Division. But that office produced one set of numbers for the commission’s special counsel and gave another set to the mayor and commission.

REPORT TO MAYOR AND POLICE COMMISSION

* 1990: 293,227

* 1991: 232,256

* 1992: 198,503

* 1993: 178,825

* 1994: 175,555

* 1995: 189,191

****

REPORT TO POLICE COMMISSION’S SPECIAL COUNSEL

* 1990: 312,870

* 1991: 252,026

* 1992: 216,449