It's COMPASlicated: The Messy Relationship between RAI Datasets and Algorithmic Fairness Benchmarks

TL;DR

This paper critically examines the use of COMPAS and similar RAI datasets in algorithmic fairness research, highlighting biases, limitations in real-world applicability, and the need for normative context in benchmarking criminal justice algorithms.

Contribution

It reveals the measurement biases in pretrial RAI datasets and argues that their use as benchmarks is problematic without considering justice and normative factors.

Findings

RAI datasets contain measurement biases and errors.

Algorithmic fairness on RAI datasets has limited real-world implications.

Benchmarking in criminal justice requires normative and contextual considerations.

Abstract

Risk assessment instrument (RAI) datasets, particularly ProPublica's COMPAS dataset, are commonly used in algorithmic fairness papers due to benchmarking practices of comparing algorithms on datasets used in prior work. In many cases, this data is used as a benchmark to demonstrate good performance without accounting for the complexities of criminal justice (CJ) processes. However, we show that pretrial RAI datasets can contain numerous measurement biases and errors, and due to disparities in discretion and deployment, algorithmic fairness applied to RAI datasets is limited in making claims about real-world outcomes. These reasons make the datasets a poor fit for benchmarking under assumptions of ground truth and real-world impact. Furthermore, conventional practices of simply replicating previous data experiments may implicitly inherit or edify normative positions without explicitly interrogating value-laden assumptions. Without context of how interdisciplinary fields have engaged in CJ research and context of how RAIs operate upstream and downstream, algorithmic fairness practices are misaligned for meaningful contribution in the context of CJ, and would benefit from transparent engagement with normative considerations and values related to fairness, justice, and equality. These factors prompt questions about whether benchmarks for intrinsically socio-technical systems like the CJ system can exist in a beneficial and ethical way.