The study found that property-crime cases with DNA evidence yielded twice as many suspects identified and arrested, as well as more than twice as many cases accepted for prosecution compared with cases without DNA evidence. It also found that DNA samples collected by patrol officers were just as likely to yield good DNA evidence as those collected by forensic technicians. Blood and saliva samples yielded significantly more usable CODIS (Combined DNA Index System) profiles compared with cell samples obtained from items potentially touched by a suspect. In addition, the study found that fewer good DNA samples were left by suspects at crime scenes where the stolen property was easily accessible (in unlocked or otherwise unsecured areas). The project assigned half of each site’s cases to a control group and half to cases in which biological material underwent DNA testing. The majority of crime scene samples were from residential burglaries, with the remainder involving commercial burglaries and automobile thefts. Participating sites collected potential sources of DNA evidence from up to 500 crime scenes between November 2005 and July 2007.