•  
  •  

Home > Law > DePaul Law Review > Vol. 75 > Iss. 4

 

Abstract

Estimations of rates of errors committed by forensic fingerprint

examiners play an important role in the criminal legal system.

Judges and jurors rely on error rate estimations as primary

indications of the probative value of assertions by fingerprint

examiners that they can reliably identify the people who deposit

latent prints at crime scenes. Recognizing the need to qualify

fingerprint identification claims with error estimations, the forensic

community has sought to conduct large-scale controlled testing of

fingerprint examiners in order to offer error estimations. One type

of error estimation generated from controlled testing is a false

positive rate, which expresses the rate at which examiners offer

incorrect identification opinions for print comparison tasks that are

known to involve different-source comparisons and that should

trigger exclusion opinions from examiners. The most recent

controlled study, completed by the FBI in 2025, resulted in the

publication of a false positive error rate of 0.29%. In theory,

fingerprint examiners can accompany identification opinion

testimony in court with claims that the chance of a false positive

error is estimated to be less than one percent. But a fingerprint error

study can only result in reliable error estimations for use in the

criminal legal system if the testing environment in the study is

sufficiently similar to the casework comparison environment and if

aggregated rates of error for all types of comparison tasks can be

disaggregated to estimate rates of error for the particular type of

comparison task at issue in a case. Regarding the similarities

between the testing environment and the casework environment,

error studies must strive to be representative of casework in

important attributes, including the complexity of latent prints

encountered in casework, the particular comparison task involved

in casework comparisons, the profile and competency of examiners

participating in the controlled study, and the type of effort offered by

participants. If the testing environment in controlled error studies is

not rigorously representative of the casework comparison

environment, then error estimations generated from the studies will

not provide accurate insights into examiner errors in casework. And

even when a controlled fingerprint error study accomplishes

sufficient representativeness, the resulting error rates will not

provide meaningful estimations of error in a case in the criminal

legal system unless overall rates of error- aggregated to include

results from easier comparison tasks and more difficult ones- can be

disaggregated and estimated for the particular type of comparison

task in the case in court. In other words, when faced with a

challenging comparison task in casework that is likely to trigger

more errors by examiners, reporting overall rates of error from

controlled studies that include examiner performance on easy

comparison tasks will be misleading. Because the recent error

estimation study by the FBI fails to accomplish a representative

testing environment and does not present disaggregated rates of

examiner error, the resulting error rate estimations are not reliable

indicators of examiner error rates in real casework. Until the

forensic community can offer disaggregated error estimations

generated from representative testing environments, judges and

jurors will continue to lack the information necessary to assess the

value of fingerprint comparison evidence in the criminal legal

system.

Download

Included in

Law Commons

Share

COinS