If most attention to campus violence this year has focused on gun violence and murderous rampages, a report released on Thursday by the Federal Bureau of Investigation offers a more sober overview of the crimes that occur on schools, colleges and universities nationwide.
Simple assaults made up the vast majority of crimes against people at educational institutions from 2000 to 2004, according to the report, while the next most common offenses included property destruction, theft and burglary. In the five-year period covered by the statistics, 27 murders or cases of non-negligent manslaughter were recorded. This year's tally, by contrast, is historically anomalous: the 32 people killed at Virginia Tech last April already outnumber the report's total, and that doesn't count the several alleged murders logged on university campuses since then .
The report confirms what most law enforcement experts already agree on: By far, most reported victims of crime are acquaintances of their assailants, and most assaults involve "personal weapons" (hands, fists and feet) or knives rather than guns. But the study's authors warned against drawing broad conclusions from the data, which are not statistically representative of the population at large and rely on incomplete reporting.
The study  also doesn't shed light on several open questions about school and campus safety, such as whether school violence is increasing or decreasing; it notes competing studies that draw opposite conclusions.
For observers of security issues on college and university campuses, however, the most important caveat is the report's definition of "school" itself: the statistics encompass both K-12 and postsecondary crime reporting. Since the numbers can't currently be broken down further, the closest proxy is the report's summary of people arrested for crimes by age -- but even that metric isn't reliable since there is overlap between students who retake a grade and those who move on to college early.
The data come from a subset of the FBI's Uniform Crime Reporting program, the National Incident-Based Reporting System, which was designed in the 1980s to record specific information about individual crimes instead of top-level summaries. Since reporting is voluntary and local law enforcement agencies are still in the process of adopting the technology, the study represents crime data from only about 33 percent of police departments, covering 22 percent of the population as of 2004.
Most of the precincts reporting data are in rural areas, FBI officials said, and it wasn't clear how many cities were included. According to S. Daniel Carter, the senior vice president of Security on Campus, a nonprofit advocacy organization, some 500 colleges and universities submit crime data to the UCR program, but the number reporting to the more detailed NIBRS system, on which the report relies, would be even smaller. Billy Estok, a spokesman for the FBI's Criminal Justice Information Services Division, said that the data could include both campus police departments and the local agencies from surrounding college towns.
"Maybe there’s less ... institutional inertia with respect to making a change like that," Carter said, referring to upgrading reporting mechanisms for the NIBRS system. "In other words, it’s easier to change a smaller agency than, say, a major metropolitan area that’s got dozens of precincts spread out."
Assistant Director Thomas E. Bush III  implied at a press briefing on Thursday morning that the report was a work in progress, and he expected better results as more agencies start reporting incident-specific data to the FBI. He suggested that followup reports would be issued every other year or so.
With the available data, however, one thing was clear: "There is not a profile of who is going to be the school rampage shooter," said David Resch, chief of the FBI's Behavioral Analysis Unit in Quantico, Va.
The report also offers findings on drug and alcohol use as well as on the general age of offenders at schools and colleges:
- The use of alcohol immediately before or during a crime in progress has increased, according to the report's numbers (although that could reflect improved reporting by precincts), accounting for some 1,346 incidents in 2004 out of 129,745 crimes reported. Drug use was significantly more common, with 8,256 offenders apparently under the influence in that same year.
- The use of computers in conjunction with crimes -- to send threats or download pornography, for example -- declined over the five years covered by the study, which could reflect improved blocking mechanisms.
- Those arrested for crimes in school or college were 6 times more likely to be 18 or younger than 19 or older.