Abstract

When confronted with a female victim of physical abuse, a forensic nurse must evaluate the pattern of injuries to determine whether the victim experienced sexual assault and/or attempted strangulation. The distinction has serious legal consequences if the evaluation fails to reveal signs of strangulation (attempted murder). Judges want data to justify a nurse's assessment: false positives and false negatives have critical consequences. We present data from two studies and describe two algorithms: one `black-box' and a statistical approach using clinical information. Communicating a likelihood ratio (LR) to non-statisticians is a persistent challenge due to frequent misinterpretations of LR as posterior odds. We present a novel graphical display that shows the estimated LR, its uncertainty, and its connection to posterior odds, to help laypeople to better understand the LR. The analysis here reinforces the cautions about ``black box' algorithms.