If the article in the link above is accurate, the problem is that reports were issued in which associations were made between DNA on items of evidence and DNA of individuals with no statistic to indicate the significance of that association.
This is like receiving a voicemail that a family member was in a car accident and is in the hospital-you automatically assume the worst when in fact there are a range of things that could have happened.
Jurors see DNA as powerful evidence, but need to have all of the facts presented to them. The very strength of DNA evidence when compared to other associative evidence such as fingerprints, tire tracks, and fiber analysis is that a numerical ‘certainty’ can be used with DNA.
A common metric used in expressing the weight of DNA evidence is a random match probability. For an example, a DNA conclusion may read:
‘The DNA profile on the handle of the knife is from a male contributor and matches the DNA profile determined for Mehul B. Anjaria. This profile is expected to occur at random in 1 in 34 quadrillion unrelated individuals.’
Given that there are approximately 7 billion people on the planet and a quadrillion is a ‘million billion’, the power of the DNA association in this conclusion is quite high!
However, with the incredible sensitivity of DNA testing, incomplete and mixed DNA profiles often are detected and are more difficult to interpret. Sometimes the random match probability is not as impressive as 1 in 34 quadrillion and can be as weak as 1 in 2 ! This article highlights the issue with weak DNA associations: Weak DNA evidence could undermine justice, experts say
‘4.1. The laboratory must perform statistical analysis in support of any inclusion that is determined to be relevant in the context of a case, irrespective of the number of alleles detected and the quantitative value of the statistical analysis.’
Part of the review process of DNA casework at the Columbus PD DNA lab will likely be to assess the judgment calls made regarding relevancy in the context of the case. For example, if the DNA profile determined for a vaginal swab matches the known DNA profile of the woman from whom it was collected, certainly a statistic is not needed. It will also be important to review the laboratory’s protocols for reporting statistics as they may actually be stricter than the FBI’s guidelines for interpretation.