Contribution statistics can make to "strengthening forensic science"

Abstract

In February 2009, the National Academy of Sciences (NAS) released the pre-print version of its report, "Strengthening Forensic Science in the United States: A Path Forward". While the report focused on forensic science in the United States, it did draw on inputs from other countries and much of the report is relevant to forensic science in other countries. The report makes thirteen detailed recommendations, several of which will require statistics and statisticians for their implementation. In this paper, the authors highlight some of these recommendations and their statistical needs, specifically, Recommendations 3 (on accuracy, reliability and validity in forensic science), 5 (on human bias and error), 6 (on measurement, validation, reliability and proficiency testing) and 8 (on quality control). Underlying some of these is the need for classification, which is also discussed. The discussion is illustrated with examples from the forensic science of bloodstain pattern analysis