Forensic science evidence: Naive estimates of false positive error rates and reliability

Forensic Sci Int. 2019 Sep:302:109877. doi: 10.1016/j.forsciint.2019.109877. Epub 2019 Jul 26.

Abstract

We do not know how often false positive reports are made in a range of forensic science disciplines. In the absence of this information it is important to understand the naive beliefs held by potential jurors about forensic science evidence reliability. It is these beliefs that will shape evaluations at trial. This descriptive study adds to our knowledge about naive beliefs by: (1) measuring jury-eligible (lay) perceptions of reliability for the largest range of forensic science disciplines to date, over three waves of data collection between 2011 and 2016 (n=674); (2) calibrating reliability ratings with false positive report estimates; and (3) comparing lay reliability estimates with those of an opportunity sample of forensic practitioners (n=53). Overall the data suggest that both jury-eligible participants and practitioners consider forensic evidence highly reliable. When compared to best or plausible estimates of reliability and error in the forensic sciences these views appear to overestimate reliability and underestimate the frequency of false positive errors. This result highlights the importance of collecting and disseminating empirically derived estimates of false positive error rates to ensure that practitioners and potential jurors have a realistic impression of the value of forensic science evidence.

Keywords: Accuracy; Error rate; Forensic evidence; Forensic science; Reliability.

MeSH terms

  • Adolescent
  • Adult
  • Aged
  • Aged, 80 and over
  • False Positive Reactions
  • Female
  • Forensic Sciences*
  • Humans
  • Male
  • Middle Aged
  • Public Opinion
  • Reproducibility of Results
  • Surveys and Questionnaires
  • Young Adult