Legal cases (both criminal and civil) often require assessment of numerical evidence, for example DNA, chemical measurements, medical data and images. There are important research questions about how such numerical evidence should be incorporated within legal decision-making processes both before and during any legal action to ensure the fair administration of justice. One strand of research in this area at Edinburgh is the development of statistical models to assess the weight of support given by individual types of forensic evidence to competing propositions in a criminal case (e.g. one from the prosecution and one from the defence). Applications worked on have included chemical measurements on glass fragments, drug traces on banknotes and gait analysis.

Another area of research is the use of graphical approaches (such as Bayesian Networks and Chain Event Graphs) for combining multiple pieces of evidence in the legal decision-making process. As this is a very interdisciplinary area, requiring input from statisticians, forensic scientists and lawyers, exchange is crucial. Researchers at Edinburgh have therefore led and contributed to projects designed to bring together the different communities such as the Royal Statistical Society Statistics and the Law Section and several primers on statistics produced for lawyers and forensic scientists.