Statistics
The forensic application of statistical analysis is of high importance when determining the value of evidence. Statistical evaluation of evidence will highlight the strength or weakness of a piece of evidence in the context of a particular case. Forensic statistics is meant to provide an unbiased method by which a jury and/or judge can weigh the value of evidence and draw appropriate conclusions accordingly. Forensic statistical methods help scientists to understand the uncertainty and error that may occur in evidence processing and reporting, as well as to avoid bias. One way that bias could be prevented is by the proper reporting of evidence, either by careful phrasing when presenting evidence or by reporting evidence in terms of its likelihood of being present as evidence based on case specifics. Likelihood ratio reporting is the current preferred method of reporting results among forensic scientists since no consensus has been reached within the community on the proper terminology or phrasing to reduce bias and maintain consistency of reaching a final determination.
Refer to chapters 4 and 6 of the 2009 NAS1 report on forensic science for more information about uncertainty, error, sources of bias, and result reporting in forensic science.
References
1. Strengthening Forensic Science in the United States: A Path Forward; National Academies Press: Washington, D.C., 2009.