Using machine learning to assess rape reports: “Signaling” words about victims' credibility that predict investigative and prosecutorial outcomes

• NCJ Number: 307441
• Journal: Journal of Criminal Justice Volume: 88 Issue: 102107 Dated: September 2023
• Author(s): Rachel E. Lovell; Joanna Klingenstein; Jiaxin Du; Laura Overman; Danielle Sabo; Xinyue Ye; Daniel J. Flannery
• Date Published: September 2023
• Grant Number(s): 2018-VA-CX-0002
• Agencies: NIJ-Sponsored
• Publication Type: Research (Applied/Empirical)

Annotation

 In the second of two articles from a larger study whose aim was to teach a computer to detect innuendo (or signaling) about a victim's credibility in incident reports of rape, the authors explored whether words expressed or not expressed, intentionally or not, influenced case progression and outcomes. As hypothesized, predictive phrases were different in cases that stalled earlier.

Abstract

 In the second of two articles from a larger study whose aim was to teach a computer to detect innuendo (or signaling) about a victim's credibility in incident reports of rape, the authors explored if the words expressed or not expressed, intentionally or not, influenced case progression and outcomes. As hypothesized, predictive phrases were different in cases that stalled earlier. Cases not recommended for prosecution lacked detail and more heavily mentioned: (in)actions of victims, actions that stall cases, and procedural words. Reports where victims were not believed or unfounded were similarly vague, procedural, and terse. Cases recommended for prosecution predictively mentioned suspects and the rape statute. The authors taught a computer to detect signaling via words that were predictive of case progression and outcomes. Negative signals about a victim's credibility often presented as unqualified statements of “fact” or observations or procedural words, indicating a focus on the process vs. victim or suspect. Implications and recommendations are provided, including how unqualified doubts about victims' credibility have substantial public safety consequences. The authors employed machine learning, specifically text classification, to identify predictive phrases. Sample consisted of 5638 incident reports of rape with a sexual assault kit from a U.S., urban Midwestern jurisdiction. (Published Abstract Provided)