DNA test to assess criminal responsibility: a Bayesian approach

Judicial statistics is generally referred to as the detection of data on crimes and trial outcomes, with subsequent analysis of the data thus detected. Statistics applied to the evaluation of evidence is increasingly being recognized as an important part of the modern criminal justice system. This paper is about the use of DNA test for identification and judicial purposes—the greatest revolution in criminal investigation—, also in reference to a concrete case of statistics applied to forensic genetics. Science and law walk together in the process of determining the procedural truth, from the preliminary investigation phase until the final judgement. Statistical evidence is not always considered reliable in the criminal procedure, as it is often based on a subjective probability concept. On the other hand, a rational evidence that can claim criminal liability ‘‘beyond any reasonable doubt’’ is necessary. In this regard, in this paper we analyze how DNA test can be considered a scientific, reliable and valid evidence when contextualized and evaluated together with the other elements found during the criminal procedure. The scientificity of identifying the individual by typing the DNA profile detected on the crime scene is thus validated, supported by the statistical calculation of the rarity of the probability that the typed genetic profile could be casually attributed to another individual in the world population.