Saturday, November 10, 2012
Joseph L. Gastwirth (George Washington University - Columbian College of Arts and Sciences) has posted The Need to Carefully Interpret the Statistics Reporting the Accuracy of a Narcotics Detection Dog: Application to South Dakota v. Nguyen, State of Florida v. Harris and Similar Cases on SSRN. Here is the abstract:
Before carrying out a warrantless search of a person or their property, police officers need to have sufficient information concerning involvement in possible criminal activity to meet the “probable cause” criteria. In U.S. v. Place, the Court held that dog sniffs of vehicles, stopped for lawful purposes, were not a search. Courts have allowed police to search a vehicle or item after a trained narcotics dog indicates that it contains contraband as the positive indication establishes “probable cause”. The criteria many courts use to assess the reliability of the dog, the fraction of positive identifications in which drugs were found, however, is the predictive value of a positive test (PVP). While related to the two proper measures of the accuracy of a screening technique, the PVP also depends on the prevalence of contraband in the places the dog has examined. It is will be seen that the same PVP can arise in situations where an accurate dog sniffs items with a low prevalence of contraband or when a much less reliable dog examines items with a high prevalence of drugs. It will be seen that it is mathematically impossible to estimate the two accuracy rates of a narcotics dog from the data typically submitted by the state to show the narcotics dog is reliable.The problem arises because one needs three equations to estimate the prevalence and the two accuracy rates but the data only provides two. These issues will be illustrated on data from actual cases. Rather than continuing to rely on an inappropriate measure of the accuracy of dog sniffs, courts should require more information concerning the accuracy of dogs in their training sessions and in the field as well as better information on the prevalence of drugs in commonly occurring settings, e.g. vehicles stopped for routine traffic violations or items examined after police have received a “tip”. Then the legal system would have sufficient information to estimate both measures of accuracy of a narcotics dog and its PVP, which would assist courts in determining whether the police had “probable cause”.