Page 25 - Delaware Lawyer - Spring 2022

the barrel of the firearm with those on the bullet or casing. Testimony from firearm testing experts that “the chances that the bullet was not fired from the defendant’s gun were 275 billion to one” is not unusual. The 2009 NAS report speaks specifically to such conclusions and states, “even with more training and experience using newer techniques, the decision of the toolmark examiner remains a subjective decision based on unar- ticulated standards and no statistical foundation for estimation of error rates.”14 Following up on the 2009 study, reporters in the 2016 PCAST study found that, “firearms analysis currently falls short of the criteria for foundational validity, because there is only a single appropriately designed study to measure validity and esti- mate reliability. The scientific criteria for foundational validity require more
than one such study, to demonstrate reproducibility.”15 At the very least, exaggerated reporting of statistical probabilities will greatly impress ju- rors, leading them to draw inaccurate conclusions.
Additional Forensic Disciplines Needing Re-examination
Many other types of forensic evi- dence that are routinely presented to juries have been questioned by recent studies.16 These include bite marks, tire and footwear impressions, bloodstain pattern analysis,17 fiber comparisons, paint or other coatings, and fire or explosives analyses.
Confirmation or Other Biases
Overlying these problems is the rev- elation that psychological biases may influence the results of scientific test- ing. Scientific analyses are assumed to be objective, as they ought to be.
Several studies, however, cast doubt on this assumption.18 Contextual influences such as expectations or hopes, groupthink, a range of biases, and peer or other types of pressure may subtly — or not so subtly — impact test results and conclusions. For example, consider the routine situation in which a police laborator y technician is given a specimen to test in order to link a suspect to a crime. When the technician is told that the specimen is from the prime suspect, or a suspect who confessed, this may impact the forensic analysis. The risk is greater when dealing with a par- ticularly violent or heinous crime where there is much pressure on law enforcement to solve the case.
Race-based biases bring another risk to the reliability of certain forensic conclusions.19 Researchers in a con- trolled study have expressed concerns,
SPRING 2022 DELAWARE LAWYER 23