181 . U LERY , H ICKLIN , B USCAGLIA & R OBERTS , supra note 176, at 7736. Once again, the authors arrive at a lower error rate (7.5%) due to their inclusion of 1,856 “ inconclusive ” conclusions in the denominator. When the inconclusive conclusions (which represented nearly 1/3 of the mated samples) are set aside, the false negative error rate is 450 / 4,113 = 10.94%. 182 . United States v. Love, No. 10cr2418 – MMM, 2011 WL 2173644, at *5 (S.D. Cal. June 1, 2011) ( “[A] false positive rate of 0.1% [is] quite low. ” ).
49:1369] FORENSICS OR FAUXRENSICS? 1411 being tested. We know that fingerprint examiners respond differently when they know that they are in a testing situation. 183 The representativeness of the samples used is also questionable. 184 Finally, this study fails the disinterested researcher requirement for Type II proficiency tests because it was paid for by the FBI, and two of the four authors work for the FBI. 185 The authors 183. Langenburg, supra note 168, at 242 (referring to a “ bias loop ” that arises when examiners know their work will be checked by verifiers who also know that they are merely verifying another examiner ’ s decision). Others take issue with the suggestion raised in Ralph Norman Haber & Lyn Haber ’s article that examiners who know they are being tested will “ perform better than when the tests are not announced and cannot be differentiated from routine work. ” Haber & Haber, supra note 175, at 386. R. Austin Hicklin et al. respond as follows: While participants in tests may indeed have different performance than in routine work, it is not reasonable to conclude that the results are necessarily better in the tests: a few examiners who are not taking the test seriously could have notably affected the results of a study, especially with respect to rare events. For example, we do not know if the examiner who made two erroneous individualizations was acting as s/he would have in routine work, or was just tired and apathetic, given it was just a test. It seems likely that at least some of the participants took the test less seriously than casework, given the serious implications of actual casework, and the absence of any negative implications on an anonymous test. R. Austin Hicklin, Bradford T. Ulery, JoAnn Buscaglia & Maria Antonia Roberts, In Response to Haber and Haber, “Experimental R esults of Fingerprint Comparison Validity and Reliability: A Review and Critical Analysis , ” 54 S CI . & J UST . 390, 391 (2014). 184 . B RADFORD T. U LERY , R. A USTIN H ICKLIN , J O A NN B USCAGLIA & M ARIA A NTONIA R OBERTS , A S TUDY OF THE A CCURACY AND R ELIABILITY OF F ORENSIC L ATENT F INGERPRINT D ECISIONS A PPENDIX : S UPPORTING I NFORMATION 3 (2011), (cautioning that, “ the overall distribution of the fingerprint data cannot as a whole be considered as statistically representative of operational data, ” though they suggest that the prints used included a large proportion of poor quality prints).
You've reached the end of your free preview.
Want to read all 48 pages?