Findings from the DNA Field Experiment

In June 2008, NIJ released the results of a five-city field study that looked at the effectiveness of performing DNA analysis on biological evidence collected from property crime scenes.

Major findings of the study, averaged across the five jurisdictions, were that when DNA evidence was collected at property crime scenes:

• Suspect identifications and arrests doubled.
  Twice as many property crime suspects were identified and arrested when DNA evidence was collected (in addition to fingerprint evidence) compared to a traditional property crime investigation.
• Prosecutions doubled.
  More than twice as many cases were accepted for prosecution when DNA evidence was processed than when it was not.
• The suspects arrested through DNA identifications were more dangerous.
  DNA arrestees had double the number of prior arrests and double the prior convictions as those arrested through traditional investigations.
• DNA was twice as effective in identifying suspects as fingerprints.
  In cases where both fingerprint and biological evidence were collected, more suspects were identified via the FBI's Combined DNA Index System (CODIS) than were identified via the FBI's Automated Fingerprint Identification (AFIS) system.

Due to differences in policies, procedures and experiment implementation among the five jurisdictions, outcomes varied, sometimes substantially. However, an average of the five sites produced additional findings, including:

• DNA samples collected by patrol officers were no less likely to yield good evidence than those collected by forensic technicians.
• Among the cases in which biological evidence was collected, fingerprint evidence was collected in one-third of the cases.
  • Comparing all cases:
    • DNA was five times as likely to ID a suspect
    • DNA was nine times as likely to yield an arrest
  • Comparing cases with both types of evidence collected:
    • DNA was twice as effective as fingerprints in identifying suspects
    • DNA yielded three times as many arrests
• The average case cost to process DNA from the time the evidence was sent to the lab until an arrest was $1,397; there was little variation in costs, except that outsourcing (sending samples to a private laboratory) was more expensive.
• The additional cost of a new suspect identification was $4,514 (see full report for explanation and context [pdf, 164 pages]).
• The additional cost of a new arrest was $14,178 (see full report for explanation and context [pdf, 164 pages]).
• The additional cost of a new case that was accepted for prosecution was $6,913 (see full report for explanation and context [pdf, 164 pages]).
• The larger the state DNA database (percentage of state population), the more likely that an identification was made; this trend was revealed when comparing across sites and over the course of the study as California rapidly expanded its database.