On paper, the number of burglaries at Harvard’s Cambridge campus plummeted 91 percent last year, but the decline had more to do with policy in Washington than safety in Cambridge.

A national change in the definition of burglary used for campus crime reporting led the Harvard University Police Department to narrow its categorization of this type of crime. The resulting drop, from 230 reported instances of on-campus burglary in 2008 to 21 in 2009, put the statistic in line with the rest of the Ivy League.

Now, HUPD, as well as campus police departments across the country, must find evidence of trespassing in order to categorize a crime as a burglary, a standard akin to the Federal Bureau of Investigation’s Uniform Crime Reporting standards. Otherwise, the crime would be categorized as a larceny.

Larceny, unlike burglary, is not required to be reported under the Clery Act, a law that mandates yearly disclosures of campus crime statistics.

Prior to the change, universities were instructed to default to the burglary classification, unless evidence suggested an incident was a larceny.

Categorizing thefts as burglaries or larcenies has long been contentious, for HUPD as well as for other university police departments.

“There’s been a lot of controversy in the higher education and law enforcement fields in how to report burglary,” said S. Daniel Carter, director of public policy at Security on Campus, an organization that advocates for transparency in campus crime reporting.

At Harvard, that controversy dates back at least to 2000, when then-HUPD associate director Kevin W. Regan wrote a letter to the Department of Education seeking clarification on the definition of burglary and comparing Harvard’s dramatically higher burglary numbers to those of other Ivy League schools.

“As you can easily see, it appears that our report has us out on an island in comparison to other Universities,” Regan, who is now HUPD deputy chief, wrote.

Despite Regan’s letter, the controversy remained unresolved, with Harvard adhering to a wider definition of burglary than many other schools.

“There’s absolutely no question that there’s been inconsistent applications of Clery definitions and guidelines by schools across the country,” Carter said.

The classification of thefts was among those discrepancies, with “a definite preference to default to the less inclusive definition,” which was larceny, according to Carter.

“It was technically a violation of the act,” he said.

In 2005, the Department of Education issued a handbook that explained crime reporting requirements for universities. In its definition of burglary, the handbook stated that if the police do not have evidence that suggests a larceny took place, the crime should be classified as a burglary.

Since then, Harvard’s burglary incidence outpaced its peers. For example, in 2007, Harvard reported 259 incidents of burglary on its Cambridge campus, while Columbia’s Morningside campus had 112 reports and the University of Pennsylvania’s Main Campus had 41.

HUPD approached both Security on Campus and Harvard’s Office of Federal Relations several years ago to lobby for the inclusion of larcenies on Clery Act reports “as a more transparent way to accurately portray property crime on campus,” according to HUPD spokesperson Steven G. Catalano.

But Suzanne Day, Harvard’s director of federal relations, wrote in an e-mail that “the crime definitions were not a particular issue for us in Washington.”

Though HUPD advocated for the inclusion of larcenies, the federal government amended the Clery Act to the opposite effect, requiring evidence of unlawful entry for the categorization of burglary.

Despite its advocacy, HUPD complied with the change. The number of on-campus burglaries—which, in 2009, would have been 245 under the previous definition—was reported as 21, in compliance with the new criteria.

The sharp decline provides a narrower portrayal of property crime at Harvard than the previous standard, but Carter, while calling the change ”somewhat troubling,” said that it allows for campus crime definitions to conform to ones used by local police departments.

“I think it’s unfortunate that the numbers are not as inclusive, but it is consistent with the national definition,” Carter said. “So it’s frustrating to us because it’s less inclusive, but at the same time, we recognize that it’s consistent with what every city and town is reporting.”

—Staff writer Sofia E. Groopman can be reached at segroopm@fas.harvard.edu.

—Staff writer Naveen N. Srivatsa can be reached at srivatsa@fas.harvard.edu.