Ten years after federal officials began compiling them, national hate crime statistics are plagued with inaccuracy due to shoddiness in voluntary reporting and other errors.
There isn't much question about the race hate that fueled the murder of Sasezley Richardson, a 19-year-old black teenager shot dead as he strolled back from a mall in Elkhart, Ind., with diapers for a friend's baby.
One of the shaven-headed suspects told police he was a member of the violent, white supremacist Aryan Brotherhood, while the other reportedly said he wanted to kill a black person in order to get in.
Police called it a hate crime from the start. Prosecutors agreed, saying Richardson was slain for no other reason than that he was African-American. Neighbors, white and black alike, held two racial unity rallies in Elkhart and raised money to help pay for Richardson's funeral.
But today, Sasezley Richardson isn't even a statistic.
When journalists, law enforcement officials, scholars or others pick up a copy of the FBI's "Hate Crime Statistics: 1999," report, containing the latest available statistics, they won't find anything representing the death of a young man who was trying to put his life together when he was shot on Nov. 17, 1999.
Instead, in the column representing Elkhart, they will find nothing. No hate crimes. Zero.
One decade after the FBI began collecting state hate crime statistics and publishing them under the federal Hate Crime Statistics Act, the national effort to document hate-motivated crime is in shambles. A recent Justice Department study concludes that "the full picture of hate crimes ... has not yet been captured through official data."
And a just-completed survey of the 50 states and the District of Columbia by the Intelligence Report illustrates how the system, already hobbled by the voluntary nature of reporting, is riddled with errors, failures to pass along information, misunderstanding of what constitutes a hate crime and even outright falsification of data.
While the published hate crime totals have been running recently at some 8,000 cases a year, the real figure is probably closer to 50,000.
"The overall numbers are worthless," says Donald P. Green, a Yale University political scientist and hate crime expert. "The entire reporting system," adds Jack Levin, another hate crime expert and director of the Brudnick Center on Violence and Conflict at Northeastern University in Boston, "is plagued with errors."
Doomed From the Start
The Hate Crime Statistics Act of 1990 was passed by Congress in the wake of an outbreak of anti-gay violence in the late 1980s. It was groundbreaking, the first federal civil rights law to include sexual orientation as a class (conservatives did insert language announcing the statute was not meant "to promote or encourage homosexuality").
And although it initially only required the Justice Department to compile and publish data for five years, that was later extended indefinitely.
"The faster we can find out about these hideous crimes," then-President George Bush said in a hopeful speech at the April 23, 1990, signing ceremony for the law, "the faster we can track down the bigots who commit them."
Bush understood that in order to deal effectively with the problem, policymakers needed to comprehend its shape and size. "Think of the FBI statistics that showed murders committed by teenagers increasing dramatically from 1986 through the early 1990s," explains Levin, the hate crime expert.
"Armed with this information, we were able to push for more after-school programs, community centers, mentors and tutors in local schools. These statistics are not just an academic exercise. They are an extremely important means whereby policy is affected."
But the law was doomed from the start. Like reporting under other national crime statutes, reporting under the federal hate crime act is voluntary. But unlike data collection under other statutes, collecting hate crime statistics has been controversial — so much so that over one-third of police jurisdictions have opted not to participate in the effort.
In some jurisdictions that have chosen to participate at the official level, opposition or indifference among personnel responsible for gathering the figures has compromised the effort and has discouraged already reluctant victims to come forward.
And because hate crime categories are relatively new and vary among jurisdictions, even conscientious officials have had problems reporting accurately.
Given these difficulties, law enforcement officers need training — both to overcome their resistance and to provide uniform reporting standards. Most have never gotten it.
The gaping holes in the reporting system have been no secret. Alabama, for instance, has not reported a hate crime for years — including 1999, the year when Billy Jack Gaither was savagely beaten to death and his body set afire in a notorious anti-gay murder in the town of Sylacauga.
Quite apart from the obvious bias involved in the Gaither murder, it was hardly plausible that Alabama had zero hate crimes in the same year that California recorded 2,295 such offenses.
Then, in September 2000, a virtually unnoticed academic study funded by the Justice Department found a "major information gap" in hate crime reporting. Based on a survey of 2,657 law enforcement agencies, the study estimated that some 37% of agencies that did not submit reports nevertheless had at least one hate crime. In addition, about 31% of the agencies with reports of zero hate crimes did, in fact, have at least one.
The study's co-authors — the Center for Criminal Justice Policy Research at Northeastern University and the Justice Research and Statistics Association in Washington, D.C. — estimated that almost 6,000 law enforcement agencies likely experienced at least one hate crime that went unreported.
The published numbers, in other words, were grossly off.
The Case of the False Zeroes
The Intelligence Report sought to uncover some of the concrete reasons for the failure of law enforcement agencies to report or for their errors in reporting.
To this end, the magazine spoke to law enforcement officials in all 50 states and the District of Columbia and asked them about specific incidents that were reported in the media but failed to show up in the FBI hate crime statistics.
Most of the incidents selected involved graffiti and similar vandalism. These incidents may be minor crimes, but they often serve as early warning signs of more serious trouble. In addition, they are difficult to mistake for non-hate crimes because they usually involve racial epithets or symbols.
What the magazine found was an array of problems, ranging from bureaucratic error to intentional omission.
Most troubling, perhaps, are the cases of false zeroes.
At least seven states, the Intelligence Report found, represented to the FBI that certain police departments had reported no hate crimes — when, in fact, those departments had chosen not to file reports at all.
In other words, instead of putting a department into the non-reporting column, state agencies charged with reporting to the federal government simply categorized those departments as having no hate crimes.
That assumption, of course, is wrong almost as often as it is right — and it has the effect of falsifying data, making states that submit false zeroes look good at the expense of those that do not. The dimensions of this problem may be enormous; fully 83% of jurisdictions that reported in 1999 said they had no hate crimes.
Wyoming may be typical of the "false zero" states. According to the FBI's "Hate Crime Statistics: 1999" compilation, 55 of 57 law enforcement agencies in the state reported no hate crimes at all — a record that other states might envy.
But things may not be as rosy in Wyoming — nicknamed the "Equality State" — as the figures suggest. Richard Russell, the manager of Wyoming's Uniform Crime Reporting (UCR) system, is the official responsible for compiling hate crime figures from Wyoming jurisdictions and then forwarding them to the FBI.
Russell concedes that he sometimes fills in an unverified zero when local police departments do not send in their reports — something that happens often, he says.
A 'Flagrant Disregard for Common Sense'
Russell is not alone. Officials in Georgia, Minnesota, North Carolina, Oregon, Utah and West Virginia say they, too, insert zeroes for non-reporting agencies.
In Georgia, Willene White-Smith of the state UCR office says there are "two types of zero" — zeroes reported by agencies and others, where state officials "just assume you had zero." This practice was documented in several states:
In Minnesota, Cathy Leatherman of the Department of Public Safety says, "If we don't get a report from an agency, that is a zero."
In Oregon, Susan Hardy of the Oregon State Police says, "If they don't send us a form, that's all there is to it. We put down a zero."
In Utah, Ogden Records Supervisor Lupe Huntley said that her city did not submit a hate crime report, but that state officials "just put down zero."
In West Virginia, state police Sgt. S. Gayle Midkiff, in charge of UCR reporting, says officials "assume it's a zero" if agencies don't report.
North Carolina is one state where the false zeroes actually got some public attention. In May of this year, an investigation by Charlotte's WSOC-TV found that state officials had filled in "hundreds" of zeroes for non-reporting agencies, despite the fact that at least some of them definitely had hate crimes in 1999.
Congressman Mel Watt, an attorney trained at Yale Law School, told the television station that the data manipulation represented a "flagrant disregard for common sense" that could have an important impact on public policy. He suggested that congressional hearings might be in order, although none have yet been called.
Reacting to the news stories, the FBI sent letters to North Carolina officials directing them to halt the practice.
Indeed, the FBI concedes false-zero reporting is widespread.
"We are aware that states do report zeroes for places that are non-reporting," Mary Victoria Pyne, who heads up the FBI office that compiles annual hate crime statistics reports, told the Intelligence Report.
"We have cautioned and cautioned and cautioned against this. We have tried to make the case clearly that zero is data and when you report zero for non-reporting you are creating [false] data."
And then there is the strange case of Illinois. Every Illinois community listed in the FBI's 1999 report experienced at least one hate crime. As it turns out, that is partly because officials there rightly refuse to report a zero for an agency that does not file a report.
But the problem, says Tim Bray of the state police's research bureau, is that there are no forms for Illinois law enforcement departments to use to report a zero. That means that only those with hate crimes get reported, which is simply another way of skewing the data.
Just 'Drunk and Goofing Around'
In addition to the false zeroes phenomenon, there are several other factors that contribute to the systematic underreporting of hate crimes — from a lack of training in recognizing hate crimes, the false belief that relatively minor crimes need not be reported to the FBI, and an over-eagerness to write off the bias aspect of criminal incidents, to outright opposition to the very notion of hate crimes.
As a result, the Intelligence Report found, law enforcement jurisdictions in at least 10 states failed to report 1999 incidents that surely qualified as hate crimes.
After a building and car were spray-painted with neo-Nazi graffiti in Amenia, N.Y., officials did not report a hate crime because the incident was "not of a level that the FBI would get," according to Dutchess County Sheriff's Lt. Gary Basher.
In fact, the FBI records even minor hate crimes like vandalism. And the two suspects were actually charged with felony criminal mischief, plus two misdemeanors.
When anti-Semitic, anti-black and anti-Hawaiian slurs were painted on 14 student-owned cars at the University of Nevada at Las Vegas, the incident went unreported. "We thought it was just juveniles playing a practical joke," said university police Sgt. Don Drake, who maintains department records. "And [the graffiti] wasn't on a Jewish person's car."
In fact, the FBI manual for reporting hate crimes says that "even when offenders erroneously target the victims [by mistaking their heritage], their offenses are still considered 'suspected bias incidents.'" And at least one black victim said that his car was defaced with anti-black slurs.
In Port St. Lucie, Fla., a swastika and other graffiti scrawled on walls outside a black woman's apartment were not reported. An official in the department records division said such crimes are treated "as criminal mischief, not a hate crime."
In Oregon, Wis., a swastika spray-painted on a home was not reported. "The family wasn't Jewish and it was a bunch of 18- and 19-year-olds and they were drunk and goofing around," said Officer John Pierce, who handled the case.
And in Meridian, Miss., racist graffiti spray-painted on the home of a former major league baseball player — and signed "KKK" — was "not investigated as a hate crime" and so went unreported. The slurs were discovered by the black athlete's seventh-grade son.
Raising the Ante
Criminologists say that minimizing or even ignoring such crimes is a grave mistake.
"When police turn their backs on juveniles who spray-paint graffiti on walls and play their boom boxes too loudly, they take a risk of seeing teenage crime escalate to offenses like aggravated assault, rape and murder," says Jack Levin, the Northeastern expert.
"In the same way, when we regard hate crime incidents like vandalism and desecration of cemeteries as mere childish pranks, we inadvertently cause the perpetrators to raise the ante."
Despite the views of Levin and other hate crime experts, a number of officers and police chiefs express strong opposition to the very concept of hate crimes. For instance, in Texas, one of 45 states with hate crime penalty enhancement laws, San Augustine Police Chief Ken Delacerda seemed to dismiss the entire idea.
"I always had a problem with hate crime laws, anyhow," he told the Report. "I mean, you don't shoot people because you love them."
Overall, experts said, the answer to these sorts of problems is training. John Holland, a long-time law enforcement officer who led Federal Law Enforcement Training Center efforts to teach officers about hate crime until 2000, says that many are unsure what a hate crime is and how to report one.
He adds that training is rare in police academies and even in most police departments. "The difficulty is that most officers want to do the job but they need the training," Holland says.
Donald Green, the Yale University political scientist, says another key factor leading to underreporting is fear of negative publicity — publicity that can gravely damage a community's reputation.
"Especially as demographic change is moving into suburban and exurban areas, there is even less attention to the issue of reporting," Green says. "They don't want to do it. ... And that only exacerbates the problem. We need to address this embarrassment factor."
Disappearing Hate Crimes
The Department of Justice-funded survey unearthed a series of systemic problems. Led by Northeastern University professor Jack McDevitt and Joan Weiss of the Justice Research and Statistics Association, the study team found that time and again, senior officers said hate crimes had been reported to the state agencies or the FBI — and that these crimes were not reflected in the FBI reports.
"Very often," McDevitt and Weiss wrote, "these representatives were disturbed to find out that their jurisdiction was listed as not having reported or reporting zero information about hate crimes to the Uniform Crime Reports because they personally had been involved in the investigation of one or more incidents of bias crime" (emphasis in the original).
In one particularly egregious example, the study's authors wrote that "one capital city in the South reported on our survey that it had (and reported) 20 hate crimes; the official UCR reports indicate this city had zero incidents."
Recontacting surveyed departments to try to get to the bottom of these "procedural pitfalls," the researchers found three general themes.
First, a number of jurisdictions said their definition of hate crime differed from the generally more inclusive definition of the federal government, apparently leading to some confusion about what to report.
Second, several respondents said that "crime data in general was highly susceptible to political influences," either to make the jurisdiction appear safer to the public or to make it appear worse in order to win federal dollars.
Third, many respondents indicated that information about the bias motivation of crimes was being lost during the state data collection process. In many cases, crimes are coded simply on the basis of the underlying charge — assault, for example — and the bias motivation noted by an investigator in a police report is ignored.
For the five states with no hate crime laws, this problem may be particularly serious. In other cases, crimes are correctly coded at the local level, but then overlooked when the data is compiled by the state.
Garbage In, Garbage Out
The Intelligence Report found many examples of these and related kinds of erroneous reporting.
In Beaumont, Calif., police statistician Kari Mendoza said that a 1999 hate crime was apparently recorded in 2000. In Hamilton, Ohio, officials said they are far behind in reporting hate crime statistics because they only recently had an in-service to define hate crime.
In Iowa City, Patsy Porter of the University of Iowa police force said "an error" resulted in an incident — where racist graffiti was painted on a university fieldhouse — never making it into the FBI statistics. In Franklin, Mass., Lt. Steve Semerjian said that the department had "relied on" the district attorney to file hate crime reports. But he said that "something seems screwed up" as he recalled an incident that was not reflected in FBI statistics.
And in Elkhart, Ind., where 19-year-old Sasezley Richardson was murdered on his way home from the mall, Lt. John Ivory said that he had personally filled out a form to report the attack as a hate crime — but that the report was filed months past the deadline for 1999 incidents, possibly because of a computer changeover in late 1999. FBI officials said it would not be included in statistics for 2000.
"We are not doing a very good job as far as I'm concerned," Ivory said of police. "I don't think there has been training in our department on what constitutes a hate crime, so these types of crimes are not being reported or flagged."
Even the FBI, which generally has little responsibility for errors because all it does is compile state statistics, apparently makes its share of mistakes.
Officials in the state of Washington sent the Report a copy of the 1999 statistics they said were sent to the FBI. But a comparison of those numbers with the FBI's hate crime statistics shows that:
Three departments reporting hate crimes did not have their numbers included;
Three other agencies had different numbers than those in the FBI compilation; and
Sixteen jurisdictions reporting zero hate crimes to the state are not in the report at all.
In Alabama, a Law is Ignored
Some states do not participate, or barely participate, in the hate crime reporting, as is their legal right. Hawaii has never reported a hate crime, and it apparently doesn't intend to.
"We likely won't be participating in the FBI program," said Paul Peron, an official in the state attorney general's office who has been helping the state implement its hate crime penalty enhancement law.
He said the FBI's definition of a hate crime "is very broad and very subjective and it's hard to know what somebody had in their heart when they beat somebody up." As a result, Peron said, the state feels prosecutors, not police, should determine motive.
Alabama has submitted reports of a handful of hate crimes, but for years it reported none at all. In the latest, 1999 report, Alabama, like Hawaii, is not listed as reporting at all.
But that is not for want of trying, says Carol Roberts, spokesperson for the Alabama Criminal Justice Information Center. Alabama, Roberts says, has a law requiring law enforcement agencies to report hate crimes to the center.
But they don't. "For five or six years, we have had this form available and no one has filled it out," Roberts says. "There is a reluctance on the part of law enforcement to determine the motive of the offender."
The anti-gay murder of Billy Jack Gaither was not reported, she says, because the Alabama hate crime law does not cover sexual orientation, and so officials decided not to submit it to the FBI.
State law, of course, has no direct bearing on which crimes need to be reported federally.
Kansas barely participates in the reporting program, with just the city of Wichita submitting data. But even that data is inaccurate, says Maryann Howerton, manager of the state's Crime Data Information Center.
"Quite honestly," Howerton said this September, "we have not finished keying in 1999, 2000 and 2001 data." The 1999 FBI hate crime statistics report was published last spring.
When police agencies do not consider hate crime reporting to be a priority — when hate crime victims rightly believe that their complaints will not be considered important — the Justice Department report concluded that the victims will be discouraged from coming forward.
Hate crime victims are often members of stigmatized minority groups who are distrustful of the police in the first place. Numerous studies have documented the fact that members of such groups are often reluctant to report hate crimes.
In a 1997 study funded by the National Institute of Mental Health, for instance, researchers found that only one-third of victims of anti-gay hate crimes reported the incident to police, compared to 57% of the victims of random crimes. The failure of police agencies to take hate crime reporting more seriously exacerbates the underreporting problem.
'Not Important Enough to Record'
Even though national hate crime reporting is voluntary, readers of "Hate Crime Statistics: 1999" are informed that "the 12,122 agencies that participated in the Hate Crime Data Collection Program in 1999 represented nearly 233 million United States inhabitants, or over 85 percent of the Nation's population."
But as both the Report's survey and Northeastern's academic study make clear, such an assertion is seriously misleading. Because of the false zeros and other problems, the report actually covers far fewer people and skews the data in other ways, too.
As Jack Levin points out, for example, Northern states generally do a better job of reporting than Southern ones. Because it's believed that black-on-white hate crimes are more prevalent in the North than the South, Levin thinks "hate crimes against whites may be overestimated" as a result.
"So it's not just sheer numbers that are in doubt, it's also the characteristics of victims and offenders," he says.
What is the real level of hate crime in America? About 8,000 are reported to the FBI annually, and the Northeastern study suggests that there may be at least 6,000 more.
If all states reported hate crimes at the same per capita rate as New Jersey did 1995-99 — a state with a relatively low overall crime rate, but a model hate crime reporting system — then approximately 25,000 hate crimes would have been recorded on an annual basis nationwide.
And, of course, none of this takes into account victims — like closeted gays afraid of being outed — who do not report attacks to police.
Some New Jersey experts estimate, for example, that only about half of all hate crimes there are reported to police — a figure that would swell the nationwide estimate of hate crimes to approximately 50,000, or about six times the number that has been reported in recent years.
The national hate crime statistics published annually by the FBI do have some utility. Levin, Green and other experts agree that certain jurisdictions — New York City, for instance — do a good job of reporting. So their numbers can be useful for scholars and others wanting to study changes within a jurisdiction.
But the numbers cannot tell you whether hate crimes have been going up or down nationally since the first statistics, covering 1991, were published. And they are nearly useless for cross-jurisdictional comparisons.
When criminologists and others wanting to learn from the outbreak of xenophobic hate crimes after the Sept. 11 terrorist attacks go to the statistics, they will find that the numbers tell them very little indeed.
And that bothers many serious scholars of hate crime. "We can't make any headway against this violence until we get high quality data," says Green, the Yale University professor. "They missed the rise of right-wing extremism in Germany in the early 1990s because they didn't know about these crimes."
This all comes as one more blow to Deborah Stout, the woman whose son, Sasezley Richardson, died when she agreed to take him off life support three days after he was gunned down in Elkhart.
"They don't treat my son's murder as coming from hate, even though the state of Indiana says it was, the prosecutors say it was, and one of the men already admitted it," Stout told the Intelligence Report.
"I guess his death is not important enough to be recorded."