Code for: Building Non-Discriminatory Algorithms in Selected Data

Principal Investigator(s): View help for Principal Investigator(s) David Arnold, University of California-San Diego; Will Dobbie , Harvard University. Kennedy School of Government. Joan Shorenstein Center on the Press, Politics and Public Policy; Peter Hull, Brown University

Version: View help for Version V1

Name	File Type	Size	Last Modified
replication			12/11/2024 01:45:PM

Project Citation:

Arnold, David , Dobbie , Will , and Hull, Peter . Code for: Building Non-Discriminatory Algorithms in Selected Data. Nashville, TN: American Economic Association [publisher], 2025. Ann Arbor, MI: Inter-university Consortium for Political and Social Research [distributor], 2025-05-09. https://doi.org/10.3886/E209804V1

Project Description

Summary: We develop new quasi-experimental tools to understand algorithmic discrimination and build non-discriminatory algorithms when the outcome of interest is only selectively observed. We first show that algorithmic discrimination arises when the available algorithmic inputs are systematically different for individuals with the same objective potential outcomes. We then show how algorithmic discrimination can be eliminated by measuring and purging these conditional input disparities. Leveraging the quasi-random assignment of bail judges in New York City, we find that our new algorithms not only eliminate algorithmic discrimination but also generate more accurate predictions by correcting for the selective observability of misconduct outcomes

Scope of Project

Subject Terms: algorithmic discrmination; quasi-experimental; pretrial detention; non-discriminatory algorithms

JEL Classification:
      C26 Single Equation Models: Single Variables: Instrumental Variables (IV) Estimation
      J15 Economics of Minorities, Races, Indigenous Peoples, and Immigrants; Non-labor Discrimination
      K42 Illegal Behavior and the Enforcement of Law

Geographic Coverage: New York City

Time Period(s): 11/1/2008 – 11/1/2013

Universe:

Universe of arraignments in New York City between November 1, 2008 to November 1, 2013

Data Type(s): administrative records data

Methodology

Data Source: New York City Division of Criminal Justice

Unit(s) of Observation: arraignment-level

Geographic Unit: borough within New York City

Related Publications

Download this project

Published Versions

V1 [2025-05-09]

Export Metadata

Dublin Core

DDI 2.5

Report a Problem

Found a serious problem with the data, such as disclosure risk or copyrighted content? Let us know.

This material is distributed exactly as it arrived from the data depositor. ICPSR has not checked or processed this material. Users should consult the investigator(s) if further information is desired.