Comments on: “FBI overstated forensic hair matches in nearly all trials before 2000”

By: Chuck

Chuck — Wed, 22 Apr 2015 02:47:30 +0000

Wow… even after reading it, it just seems unfathomable. If winning is your goal in trial, you should be a defense attorney. Prosecutors’ job is to seek justice and truth, not to win.

http://www.americanbar.org/publications/criminal_justice_section_archive/crimjust_standards_pfunc_blkold.html

By: MattS

MattS — Mon, 20 Apr 2015 18:52:22 +0000

In reply to DEM. And as usual, the government is a day late and a dollar short.

By: DEM

DEM — Mon, 20 Apr 2015 13:56:58 +0000

It’s as if the government is trying to destroy any remaining faith in the fair administration of the law.

By: gasman

gasman — Mon, 20 Apr 2015 12:27:06 +0000

… ranking Democrat, Patrick J. Leahy (Vt.), urged the bureau to conduct “a root-cause analysis” Wonder if he meant to deliver the pun.

But really, this could be a science, if only scientific methods were applied. Just for starters, some considerations.
1) Establish sensitivity, specificity, and predictive values by running lots of knowns
2) Strict blinding of examiners. Can’t have investigators whispering in the examiner’s ear their a priori belief in the likelihood of a match
3) Control samples must be randomly interspersed with day to day work.
4) all positives must be reviewed by an independent lab. That sample must be transported with 10 other control sets in a blinded manner. A control set is a match or non-match set of hair samples each representing a full examination. The independent lab’s success in identifying the control samples is reported to the jury also.