Some DAs, some where in the US must have an undergraduate science degree. Please explain these concepts to the @NDAA president Michael Ramos and NY ADA Melissa Mourges and while at it, to some appellate court Justices in Texas soon to rule on the bitemark aided conviction and upcoming September 2017 execution of Kosoul Chanthakoummane.

@gwynethnotpaltrow talks about the promotion and experimental framework of”Accurate” forensic method of hair chemistry. From her many learned statements, the first is that the tech is not commonly available anywhere in the crime lab environment. Other pertinent observations are spot on.
April 7, 2017 at 2:13 pm
Note that none of this is peer-reviewed yet. Note that “80% accuracy” in sex determination is from a sample size of 10 men and 30 women. I wonder what accuracy they would have gotten by measuring hair length.

I’d love to read the actual details of their technique – but they aren’t available as far as I can tell. They have a n of 40 (20 Jordanian women, 20 US men and women), they measure 15 different compound-specific carbon isotope values, and they look at >>5 parameters (“extensive” questionnaires about diet, hair care and health). There’s no indication how many of the individuals may have been related. You don’t need much statistics background to tell you how many high correlations you can come up with if you measure lots of things on a few individuals.

I find it surprising that they say LC-IRMS is a technique that many technicians can do, and is being more widely used in forensic laboratories. I don’t know of a single forensic laboratory that does this technique. IRMS (Isotope Ratio Mass Spectrometry) is widely used in academic laboratories, and there is an organization called FIRMS that tries to provide some rudimentary QA/QC to members. However, _LC-IRMS_ is only done at less than ~20 laboratories in the world. Note that the lab that produced these results is not accredited by FIRMS or any other organization.

There’s no confirmation from other researchers, because they haven’t made any details of their research public. Most researchers I know prefer to do press releases when the peer-reviewed paper comes out, so that interested people can find out more – not just when somebody gives a conference talk.

You can try to contribute to their research by registering to provide hair samples at http://glen-jackson.eberly.wvu.edu/nij-hair-project. I registered a few months ago, but haven’t heard anything back yet.

Thanks for your terrific blog!