Investigating Human + Machine Complementarity for Recidivism Predictions
This work addresses the problem of enhancing fairness and accuracy in recidivism predictions for criminal justice stakeholders, but it is incremental as it builds on prior research without achieving major gains.
The study investigated whether combining human and algorithmic predictions could improve recidivism risk assessment, finding that while human and COMPAS decisions differed, this did not lead to significant improvements in prediction accuracy.
When might human input help (or not) when assessing risk in fairness domains? Dressel and Farid (2018) asked Mechanical Turk workers to evaluate a subset of defendants in the ProPublica COMPAS data for risk of recidivism, and concluded that COMPAS predictions were no more accurate or fair than predictions made by humans. We delve deeper into this claim to explore differences in human and algorithmic decision making. We construct a Human Risk Score based on the predictions made by multiple Turk workers, characterize the features that determine agreement and disagreement between COMPAS and Human Scores, and construct hybrid Human+Machine models to predict recidivism. Our key finding is that on this data set, Human and COMPAS decision making differed, but not in ways that could be leveraged to significantly improve ground-truth prediction. We present the results of our analyses and suggestions for data collection best practices to leverage complementary strengths of human and machines in the fairness domain.