2 Some typical questions for DNA evidence
Acting as a lawyer in this case, consider questions that you will ask the DNA expert in court. You need to be able to justify why these should be asked.
DNA experts might be able to provide some information in relation to three key questions that both the prosecution and defence lawyers would want to address. These are:
- Is there DNA on the bat? Is the DNA present from a single source or multiple sources and how could multiple sources be accounted for?
- Whose DNA is on the bat and how sure are you of that result?
- What activities could result in a person’s DNA being transferred to the bat?
- You might think that a question such as ‘is the person’s whose DNA on the bat the perpetrator of the crime?’ might be asked, that is an offence level question and scientists would not, and should not, answer questions such as that.
It is likely that the digital forensic expert in the case would be questioned in a similar way though about digital evidence.
The trustworthiness of both the scientific tests and the expert’s interpretation of the scientific results will depend on how clearly the expert explains what the evidence means within the case context, what are the limitations of the evidence and, how this is communicated (more on this in Week 4). To answer the first question, the DNA expert would report the results of a statistical calculation which evaluates how likely it is that the DNA found on the bat belongs to the suspect, compared to how likely it is to belong to someone else in the ‘world’ who is unrelated to the suspect. This calculation is called a likelihood ratio. In trying to communicate this information to the court and jury the expert might say something like:
‘Analysis of the DNA recovered from the bat gave a full DNA profile that matched that of the suspect. If the DNA evidence did not come from the suspect then the DNA profile must match by chance. It is estimated that the chance of obtaining these matching profiles if the evidence came from a random person unrelated to the suspect is in the order of 1 in a billion (where a billion is a thousand million).’
Research has shown that, since the introduction of the use of likelihood ratio calculations in the late 1980s, both legal practitioners and juries have found it difficult to understand and this difficulty remains to this day (Morrison et al., 2025).