Another example of flat out bad forensic science: Bad DNA results

In prior posts we discussed our forensic science wall of shame.

  1. Urine testing ain’t worth piss
  2. Government run Crime Labs are never wrong

Now, we have to add this one to the ranks…

This article is is so remarkable that it earns its own post.


While the content is sad, but it is not totally abnormal unforunately as the above shows, but what makes it worthy of examining are the comments left behind by readers.  Do people really care?

Newspaper article reveals bad DNA testing at state lab. Meet Shelly Schultz who performed something vaguely resembling DNA testing in the Biology lab at the OSBI Forensic Science Center according to the Newspaper resport

First the body of the article and the factual assertions by the Newspaper and other media:

An Oklahoma State Bureau of Investigation DNA analyst was fired last month after a review of her case work revealed instances of sample contamination, handling errors and documentation mistakes, an investigation by The Oklahoman has revealed.

Analyst Shelly Schultz, 32, of Oklahoma City, was given an opportunity to improve her performance, but was fired after additional problems were noted — including accidentally switching samples during a proficiency test overseen by quality control officials.

“The OSBI believes that ultimately no erroneous results involving Ms. Schultz’s work were left uncorrected by amended reports issued in actual prosecutions or investigations or further testing or retesting of evidence,” Bunn stated. “Consequently, no erroneous results were ever relied upon or used in an actual prosecution or investigation.”

Schultz, a five-year OSBI employee, was notified in a June 18 letter that she was being terminated effective June 30.

…”I multiplied the pressure I felt, putting even more pressure on myself to perform and I buckled under the pressure I put on myself,” Schultz wrote.

Lack of notification


Concern has surfaced over OSBI officials’ failure to immediately notify prosecutors and defense attorneys that there was a problem.

…Prater said local prosecutors are “hypersensitive” concerning any problems regarding the credibility of tests performed by forensics specialists because of all the problems that ensued several years ago after former Oklahoma City police chemist Joyce Gilchrist’s credibility came into question.


Previous problems


However, documents obtained by The Oklahoman through an Open Records request show OSBI officials had known for at least nine months that there were serious problems with some of Schultz’s work.

Schultz was suspended without pay in early October for poor work performance and was involuntarily demoted Feb. 17 because of continuing problems.

Records show Schultz was hired by the agency as a criminalist (evidence analyst) in February 2005 at a salary of $34,900 a year. She obtained additional training and rapidly gained promotions and pay increases, rising to the level of Criminalist III with an annual salary of $62,707 by March 1, 2008.

In late 2008, however, supervisors began criticizing her for her slow work, writing that she was falling far short of the agency’s case work standard of completing 3 to 4.75 cases per month. After a period of mentorship, Schultz was returned to case work at the end of March 2008 but only completed 13 cases during the next 11 months, causing her to receive more criticism, records reveal.

Samples tested


During May 2009, “the potential contamination of a case sample and the possible reporting of the DNA profile regardless of the contamination led to an investigation by Erin Henry, OSBI quality manager, and me,” supervisor Ryan Porter reported in September. “It was also discovered that the contamination also affected another case that was batched together with the previously mentioned case.”

That prompted a case file review, during which “additional instances of contamination, sample handling errors and documentation errors” were observed in Schultz’s past work, Porter reported.

Porter also stated “circumstances surrounding the contamination and the possible reporting of the profile indicated that she may be trying to cover up the contamination.”

Henry told The Oklahoman a subsequent inquiry revealed there had been no attempted cover up, but just a miscommunication concerning which supervisors needed to be notified about the contamination.

Schultz had informed her biology unit technical supervisor, but another supervisor should also have been informed, Henry said.

Schultz was pulled off new DNA case work in June 2009 because of the problems and required to take a proficiency test, records show.

During the proficiency test, she accidentally switched samples being tested.

Laying blame


In a memorandum to her supervisor, Schultz blamed the switch at least partially on having her routine for processing case samples broken by having been removed from case work.

The impact of Schultz’s mistakes on specific cases is difficult to determine because references to specific cases were redacted from documents provided to The Oklahoman.

One memo indicated that at least two consumption samples were contaminated in work done by Schultz.

Consumption samples are evidence samples that are used up by the testing process, making retests impossible.

Andrea Solorzano, Schultz’s division supervisor, said no usable information was obtained from one of the test samples, so contamination of the sample won’t make a difference in that case.

In the other case, analysts were able to obtain a partial DNA profile from a second sample, so that evidence could be presented to a jury, she said.

When Schultz was demoted in February, she wrote that she took “full responsibility for my bad attitude” and promised to complete her next corrective action plan “with a good attitude and a smile on my face.”

OK.  That’s really, really bad.

Someone who obviously has no aptitude or facility in DNA testing is not only allowed to continue to do testing on case files, but also she was promoted, coached and allowed to continue to do case work!  It is a complete systemic failure and a complete abortion of a well-developed Quality Control and Quality Assurance program.

BUT DO PEOPLE CARE? Do normal folks stand up and voice their outrage? Here are some comments left by folks:

  • “Joe” wrote:  “While these chemist play with the lives of innocent people the guilty people remains free to continue their criminal activities.”
  • “Kathrine” wrote:  “I’m assuming you read that in both instances of contamination it didn’t adversely affect the results of the case. I’ll also assume that you aren’t suggesting that as an imperfect species we can function without error 100% of the time… What I won’t assume and what I can say with certainty is, the analyst written about in this article took her job very seriously and the last statement in this article was gravely taken out of context.”
  • “Iam Stinkerpants” wrote:  “Local Yokel DNA-Lab-Types just can’t be trusted. Tests should be run by an agency independent of the law enforcement collecting the samples.”
  • “Ed” wrote:  “Comment on OSBI incompetence? This box I’m typing in is truly a blank canvas and my keyboard is a palette containing every color in the world. I’m still laughing at the part where “agency officials are confident” her mistakes never left the building. I always like to hear about confidence levels at our state agencies.”
  • “James” wrote:  “Easy guys she could very easily be a scapegoat for the OSBI’s inability to solve any high profile cases. (Anadarko, Weleetka) Pretty sure there are other’s in the bureau who need their work examined as well”
  • “Phil” replied:  “More likely she is evidence of the problem, which as you say there are probably, almost certainly, others in need of audit. Not every crime can or will be solved and it is incredulous to think they can but to be involved in convicting someone of a crime they didn’t commit is worse than the crime for which they were convicted.”
  • “Cowboy” wrote:  “You would think that after the Gilchrist fiasco the crime labs throughout Oklahoma and the nation would have established a quality control system that safeguards every sample. Every sample should be treated with the utmost professionalism and only handed by people of integrity and ethics. Each sample should be double and occasionally triple checked for accuracy. There should be a law that states the first time your caught half assing your work you lose your certification. If they do not clean this mess up, scientific data will have as much credibility as witness testimony. Your talking about people’s liberty here, if they cannot land a conviction the legal and ethical way they are wasting the time and money of the taxpayers.”
  • “Phil” wrote:  “I’m always amazed when an article like this is produced as if it were a rarity; it isn’t. The problems with prosecutorial evidence are large and widespread and is not limited to forensic evidence but includes eyewitness identification, crime-scene investigation and mishandling of evidence before, during and after trial.  It’s also strange how an expert witness for the prosecution is usually considered more credible than an expert for the defense. I guess it comes from the blind being led by the blind. Too many people believe the state can do no wrong. In playing with people’s lives, there is no room for error.”

I have blogged before on the large scale lack of credentialed scientists in forensic science today.  This is yet again more evidence of why it is right to be a modern day skeptic.  Why it is right to be a modern day skeptic: The need to continue to challenge sacred cows: Paradigm Shift

[Bloger’s note:  Special thanks to Josh Lee, Esquire, who brought the story to all of our attention]

Leave a Reply

Your email address will not be published. Required fields are marked *