The forensic statistical analysis of incident series

Publication date

DOI

Document Type

Master Thesis

Collections

Open Access logo

License

CC-BY-NC-ND

Abstract

When a person is involved in a series of incidents, it may be used as evidence in court. An infamous case in the Netherlands is the one of the nurse Lucia de Berk, whose presence during a relatively large number of incidents in hospital wards raised suspicion. Other examples include persons being involved in a high number of fire incidents, or traffic accidents. In this research, we dive into four types of incident series and discuss the different forensic statistical methods that were proposed in the literature to quantify the strength of evidence. We will give comments and recommendations for further improvements of the proposed models. The aim is to write a recommendation to the Netherlands Forensic Institute on how to deal with a series of similar, possibly criminal, events when they are asked by parties such as the police or legal authorities to analyse the statistical evidence.

Keywords

Clusters of unusual events; incident rates; negative binomial distribution; serial crimes; naked statistical evidence; fraud; fire incidents; roster cases

Citation