Document Type


Publication Date



This is the first study to explore the forensic science testimony by prosecution experts in the trials of innocent persons, all convicted of serious crimes, who were later exonerated by post-conviction DNA testing. Trial transcripts were sought for all 156 exonerees identified as having trial testimony by forensic analysts, of which 137 were located and reviewed. These trials most commonly included testimony concerning serological analysis and microscopic hair comparison, but some included bite mark, shoe print, soil, fiber, and fingerprint comparisons, and several included DNA testing. This study found that in the bulk of these trials of innocent defendants - 82 cases or 60% - forensic analysts called by the prosecution provided invalid testimony at trial - that is, testimony with conclusions misstating empirical data or wholly unsupported by empirical data. This was not the testimony of a mere handful of analysts: this set of trials included invalid testimony by 72 forensic analysts called by the prosecution and employed by 52 laboratories, practices, or hospitals from 25 states. Unfortunately, the adversarial process largely failed to police this invalid testimony. Defense counsel rarely cross-examined analysts concerning invalid testimony and rarely obtained experts of their own. In the few cases in which invalid forensic science was challenged, judges seldom provided relief. This evidence supports efforts to create scientific oversight mechanisms for reviewing forensic testimony and to develop clear scientific standards for written reports and testimony. The scientific community can through an official government entity promulgate standards to ensure the valid presentation of forensic science in criminal cases and thus the integrity and fairness of the criminal process.


The Appendix, trial materials, and other reports associated with exonerees' cases cited in this paper can all be found at the website:

Library of Congress Subject Headings

Expert evidence, Forensic sciences, Judicial error, Empirical

Included in

Law Commons