Proficiency testing of fingerprint examiners with Bayesian Item Response Theory1

In recent years, the forensic community has pushed to increase the scienti c basis of forensic evidence, which has included pro ciency testing for ngerprint analysts. We use pro ciency testing data collected by Collaborative Testing Services in which 431 ngerprint analysts were asked to identify the source of latent prints. The data were analyzed using a Rasch model with a Bayesian estimation approach. Although these data provide valuable information about the relative ability of the examiners and the relative di culty of the questions, it does not necessarily extrapolate onto general performance of examiners or di culty in casework, which we show through sensitivity analysis and simulation. We show that a Bayesian Item Response Theory analysis provides a deeper understanding of analysts’ ability and question di culty than other forms of analysis. A large-scale adoption of Item Response Theory in this area would provide both more precise estimates of pro ciency and quantitative evidence for the relative di culty of di erent questions.