Evidence courts rely on in crime investigation may not be as accurate as we think: UBC study

Recent research from the University of British Columbia suggests that traditional crime investigation methods, such as bloodstain patterns, ballistics evidence, and sniffer-dog searches, may not be as reliable as courts and investigators assume. Associate Professor Sara Gordon, who led the study at the Peter A. Allard School of Law, stresses that the mere fact that a technique has been used in court for decades does not guarantee its reliability, especially when someone's freedom is on the line in a criminal trial.

One example of this potential unreliability is ballistics evidence, where the study found investigators misidentified bullets 37% of the time. Bloodstain pattern analysis is another method that may be overstated in its value; Gordon notes that many practitioners use fluid dynamics concepts without the necessary physics training. The study found that analysts correctly classified bloodstain patterns only 70% of the time on solid surfaces, and only 62% on fabrics.

The researchers also question the accuracy of drug-detection dogs, citing a study from Australia that found nearly 75% of police dog alerts were unfounded. Gordon emphasizes the need for clearer standards on training, certification, and performance monitoring for forensic evidence and calls on judges to ensure expert testimony in court is scientifically supported. She advocates for greater scientific literacy among lawyers and judges to better understand the limitations of these techniques.