Study: Hair analysis is an unreliable, flawed forensic technique

CHARLOTTE, N.C., April 21 (UPI) -- If Jim Norton had his way, hair analysis would be inadmissible in the court of law. The forensics technique is flawed and unreliable.

Through the years, Norton has offered expert testimony involving hair analysis evidence saying exactly that. Now, Norton has made the case in print.

"For 20 years, FBI forensic experts gave flawed testimony regarding microscopic hair analysis," Norton and his colleagues wrote in the abstract of a new paper on the subject -- published this week in the journal Significance.

Since 1989, 74 people have been sentenced to hard time for serious crimes as a result of hair analysis, only to later be exonerated by DNA analysis.

Forensic scientists use up to 20 different characteristics to identify one hair from another, but Norton and his colleagues say many of these characteristics are subjective. Give a hair to five forensic scientists, they say, and you might get five different descriptions.

Making matters worse, unlike fingerprint and DNA analysis, scientists are without population-based databases with which to compare hair samples with. Therefore, they're unable to properly estimate the probability of a hair having a specific hair characteristic.

"The magnitude of the injustice enabled by such flawed 'science' is reflected in the 1,056 years of unjust prison time served by the 74 exonerated defendants," Norton, who serves as the director of biostatistics at the James G. Cannon Research Center in North Carolina, said in a news release.

"Relatedly, similar flawed 'science' based upon bite mark evidence has led to 24 unjust convictions or indictments, later disproved by DNA evidence."