(Mainly) bayesian data analysis with a philosophical twist

False positives in DNA identification 5

In a few recent posts we discussed the error probability for false positives in DNA evidence evaluation. Now, let’s think about the error probability for false negatives and its impact on the value of exculpatory DNA evidence. Consider the probability of no match being reported if an error has been made, analogus to k in the previous calculation... Read more

False positives in DNA identification 4

Last time we derived one formula for the impact of false positive probability on the value of DNA evidence. Today, we’ll look at a seemingly different formula to get a better perspective. Buckleton, Bright, & Taylor (2018) give a formula for the impact of errors on likelihood ratio different from the one we discussed before. Since the deriva... Read more

False positives in DNA identification 3

Hopefully, having convinced the reader that the false positive probability is non-negligible, I can now move to a probabilistic approach to the impact such probability can have on the value of evidence. Since I will use the likelihood ratio, a fairly standard measure of evidential strength, I first explain this notion, then explain one relevant ... Read more

False positives in DNA identification 2

Last week I briefly mentioned the phenomenon of false positives in DNA identification. Today, we’ll look at a few more examples, divided by the problem type: Cross-contamination of samples. In Dwayne Johnson (2003) samples were accidentally swapped. In Lukis Anderson (2012) the material has been carried over by the responding paramedic... Read more

False positives in DNA identification 1

One context in which probabilities are extensively used is the use of DNA evidence. In testifying about the DNA match at trial, experts often assess the probability that a random person, unrelated to the crime, would coincidentally match the crime stain profile (random match probability). RMP is often an impressively low number, say 1 in 100 ... Read more

Hidden subjectivity in DNA evidence evaluation: binning

While DNA evidence might seem the ideal of objectivity, in fact there is quite some wiggle room in DNA evidence interpretation. Here’s a fairly simple example of how the choice of hidden parameters might impact such evaluation. The method relies on regions of DNA sequences in which there are variable numbers of short tandem repeats (VNTRs). Any... Read more

Context importance in DNA evidence evaluation: an extreme example

In a 2005 killing of Anne Marie Foy, David Butler, a former taxi driver from Wavertree, Liverpool, was accused of murdering a local sex worker. An important role in the initial conviction and final acquittal was played by DNA evidence. The prosecution’s case was based on DNA trace evidence matching Mr Butler’s found on the victim’s nail. Now, ... Read more