Forensic judgments: Validity, reliability, and bias

Contains fulltext : 233894.pdf (Publisher’s version ) (Open Access)Forensic examiners provide answer to questions such as: "Is this bullet fired with this firearm?" or "Is this person the source of this fingerprint?". To provide an answer to these questions, the examiners evaluate the evidential material. Based on their observations, examiners provide a judgment about the source of the evidential material and how much support the evidence provides for this judgment (the evidential strength). The judgments of examiners are accepted in court on the merit of the examiners' training and experience. The court uses these judgments to decide about a person's innocence or guilt. To ensure that a suspect is not unjustly convicted or acquitted because of misrepresented forensic evidence, it is imperative that the reliability and validity of the examiners' judgments are known and as high possible, and that these judgments are unbiased. This dissertation shows that while the support for the reliability and validity of the examiners' source-judgments was quite high, they are not perfect. Examiners are not infallible and have difficulty providing calibrated evidential strength judgments. Furthermore, examiners are prone to bias by irrelevant information. Additional training, personal feedback, and support by the outcomes of computer-based methods could lead to better judgments. Adjusting standard operating procedures based on behavioral insights can minimize the occurrence of bias.Radboud University, 22 juni 2021Promotores : Witteman, C.L.M., Berger, C.E.H. Co-promotor : Stoel, R.D.301 p