UConn Among America’s Most Crime-Rattled? Hardly

A June 2012 study by the Daily Beast found the University of Connecticut to be the 8th most crime-rattled in the country. But according to a rigorous analysis of the same data, UConn doesn’t even rank in the top 25.

The website NerdWallet.com studied Department of Education crime data reported under the Clery Act, spanning the years 2008 and 2010, and found that:

• UConn’s poor crime rating largely stemmed from three murders: a stabbing in 2008 that killed two and a 2009 brawl killing one. But these incidents, while terrible, are not always representative of the campus’ ongoing safety.
• Indeed, UConn’s per-capita rates of robbery, aggravated assault, burglary and vehicular theft were well below the national average.
• UConn’s 11 arson cases also contributed to the high ranking, but an in-depth analysis shows at least 4 of those to be dorm-room mischief rather than malicious or violent fires.
• Though its rate of per-capita aggravated assaults is above average, cases have declined sharply from 10 in 2008 and 12 in 2009 to a remarkable 2 cases in 2010.

“UConn does not deserve the designation as one of the 10 least safe schools,” says NerdWallet VP Anisha Sekar, who led the study. “Black swan events and pranks are of little use to current or prospective students looking to evaluate the environment.”

How does our methodology differ?

Our analysis differs from the previous analysis in the following ways:

1. We tally incidents. Whereas previous analyses assigned a severity to each crime, we compared the number of crimes to number of students.
2. We included only violent crimes, which we define as causing or threatening to cause physical harm to another. The two studies included:
   1. Previous analysis: Murder, negligent murder, robbery, aggravated assault, burglary, vehicular theft, and arson.
   2. Our analysis: Murder, negligent murder, forcible sexual offenses, simple and aggravated assault, robbery and intimidation.

 Why does our methodology differ?

1. We count, rather than weight, crimes. Our analysis sought to answer the question, “How many incidents of violent crimes occurred?”
   1. Such a question lends itself to an objective count, rather than a weighting system that is inherently subjective and vulnerable to human biases.
   2. A weighting system would also give outsize influence to a few tragic incidents that are nonetheless not representative of a college’s culture.
   3. We excluded burglary, vehicular theft and arson, because the data provided makes an analysis of violent crime easier than overall crime.
      1. Those categories, for example, exclude a laptop stolen from a student’s parked car, or a random pickpocketing.
      2. Yet they would include a fire set by a bored student, and would double-count a fire in which a student dies as both arson and murder.
      3. We included forcible sexual offenses, simple assault and intimidation because they encompass rape and other violent sexual crimes, physical attacks where the victim nonetheless suffers no major injuries, the credible threat of physical harm.