Arrest Statistics Have Limited Accuracy

"Arrest statistics also have limitations for measuring the volume of arrests for a particular offense. Under the UCR Program, the FBI requires law enforcement agencies to classify an arrest by the most serious offense charged in that arrest. For example, the arrest of a youth charged with aggravated assault and possession of a controlled substance would be reported to the FBI as an arrest for aggravated assault. Therefore, when arrest statistics show that law enforcement agencies made an estimated 193,900 arrests of young people for drug abuse violations in 2004, it means that a drug abuse violation was the most serious charge in these 193,900 arrests. An unknown number of additional arrests in 2004 included a drug charge as a lesser offense."

Source: 

Snyder, Howard N. Juvenile Arrests 2004. Washington, DC: US Department of Justice, Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention, December 2006.