A UBC study says some forensic techniques treated as reliable in criminal cases can produce inaccurate results.
The Story
A study from the University of British Columbia says several forensic methods relied on in criminal investigations and presented in court may be less reliable than commonly assumed.
The study author, Sara Gordon, an associate professor at UBC’s Peter A. Allard School of Law, said long-established courtroom use should not be taken as proof of reliability when a person’s liberty is at stake in a criminal trial.
Gordon said ballistics evidence—matching bullets and cartridge cases to a particular firearm—can be wrong. In the study’s test, investigators misidentified bullets 37 per cent of the time.
The research also points to issues with bloodstain pattern analysis. The study says Canadian courts treat the approach as solid proof, while practitioners may use concepts such as fluid dynamics without necessarily having physics training.
According to UBC, one study found bloodstain pattern analysts correctly classified patterns only 70 per cent of the time on solid surfaces. On fabrics, the figure dropped to 62 per cent.
The study further says drug-detection or sniffer dogs can generate inaccurate results, citing research conducted in Australia that found nearly 75 per cent of police dog alerts were unfounded.
Gordon said clearer standards for training, certification, and performance monitoring are needed, and the researchers call on judges to ensure expert evidence used in court is supported by science.
← More stories