Comparing Conventional and Machine-Learning Approaches to Risk Assessment in Domestic Abuse Cases

67 Pages Posted: 8 Mar 2020 Last revised: 2 Oct 2020

See all articles by Jeffrey Grogger

Jeffrey Grogger

University of Chicago - Harris School of Public Policy; National Bureau of Economic Research (NBER)

Ria Ivandic

London School of Economics & Political Science (LSE) - Centre for Economic Performance (CEP); King’s College London - Department of Political Economy

Tom Kirchmaier

London School of Economics - Centre for Economic Performance

Multiple version iconThere are 3 versions of this paper

Date Written: February 5, 2020

Abstract

We compare predictions from a conventional protocol-based approach to risk assessment with those based on a machine-learning approach. We first show that the conventional predictions are less accurate than, and have similar rates of negative prediction error as, a simple Bayes classifier that makes use only of the base failure rate. A random forest based on the underlying risk assessment questionnaire does better under the assumption that negative prediction errors are more costly than positive prediction errors. A random forest based on two-year criminal histories does better still. Indeed, adding the protocol-based features to the criminal histories adds almost nothing to the predictive adequacy of the model. We suggest using the predictions based on criminal histories to prioritize incoming calls for service, and devising a more sensitive instrument to distinguish true from false positives that result from this initial screening.

Keywords: Domestic abuse, Risk assessment, Machine learning

JEL Classification: K42

Suggested Citation

Grogger, Jeffrey T. and Ivandic, Ria and Kirchmaier, Tom, Comparing Conventional and Machine-Learning Approaches to Risk Assessment in Domestic Abuse Cases (February 5, 2020). Available at SSRN: https://ssrn.com/abstract=3532560 or http://dx.doi.org/10.2139/ssrn.3532560

Jeffrey T. Grogger

University of Chicago - Harris School of Public Policy ( email )

1155 East 60th Street
Chicago, IL 60637
United States

National Bureau of Economic Research (NBER)

1050 Massachusetts Avenue
Cambridge, MA 02138
United States

Ria Ivandic

London School of Economics & Political Science (LSE) - Centre for Economic Performance (CEP) ( email )

Houghton Street
London WC2A 2AE
United Kingdom

King’s College London - Department of Political Economy ( email )

Strand Building
London
United Kingdom

Tom Kirchmaier (Contact Author)

London School of Economics - Centre for Economic Performance ( email )

Houghton Street
London, WC2A 2AE
United Kingdom
+44 207 955 6854 (Phone)

HOME PAGE: http://sites.google.com/site/tomkirchmaier/home

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
152
Abstract Views
906
Rank
267,333
PlumX Metrics