Uploaded November 2025 | Updated September 2026, 10 hours ago
Recorded as part of the CFvW-Colloquium on November 19, 2025
Intimate Partner Violence Risk Assessment Tools: Remarks on Mechanisms,
Values, and Impact - Raysa Benatti (Universität Tübingen)
Intimate partner violence (IPV) is one of the more complex societal issues to become a target of algorithmic solutions in the past decades. In several nations, increased public awareness and political articulation of questions of gender justice has led to the implementation of institutional and legal measures to address gender-based violence. In the context of the early-2000s movement toward evidence-based policy, social workers, health professionals and researchers turned to risk assessment tools to manage the occurrence of IPV. The literature explores a range of approaches, from more traditional clinical judgment to actuarial systems, which claim to provide more accurate predictions. It remains debatable whether actuarial systems for predicting IPV are superior to clinical judgments, since their impacts have yet to be properly assessed. In this talk, we (1) situate some of the IPV risk assessment tools in the clinical-actuarial spectrum, stressing the debate on the significance of the distinction; (2) outline some values at stake in their development and use; and (3) discuss impact assessment and implications of augmenting such tools with methodologies adopted from machine learning.
(Joint work with Konstantin Genin, University of Utah.)
Recorded as part of the CFvW-Colloquium on November 19, 2025
Intimate Partner Violence Risk Assessment Tools: Remarks on Mechanisms,
Values, and Impact - Raysa Benatti (Universität Tübingen)
Intimate partner violence (IPV) is one of the more complex societal issues to become a target of algorithmic solutions in the past decades. In several nations, increased public awareness and political articulation of questions of gender justice has led to the implementation of institutional and legal measures to address gender-based violence. In the context of the early-2000s movement toward evidence-based policy, social workers, health professionals and researchers turned to risk assessment tools to manage the occurrence of IPV. The literature explores a range of approaches, from more traditional clinical judgment to actuarial systems, which claim to provide more accurate predictions. It remains debatable whether actuarial systems for predicting IPV are superior to clinical judgments, since their impacts have yet to be properly assessed. In this talk, we (1) situate some of the IPV risk assessment tools in the clinical-actuarial spectrum, stressing the debate on the significance of the distinction; (2) outline some values at stake in their development and use; and (3) discuss impact assessment and implications of augmenting such tools with methodologies adopted from machine learning.
(Joint work with Konstantin Genin, University of Utah.)
