Predicting Juror Predisposition Using Machine Learning: A Comparative Study of Human and Algorithmic Jury Selection

TL;DR

This study empirically compares human jury consultants and machine learning models in predicting juror predispositions, finding ML models significantly outperform humans under the same conditions and offering advantages in transparency and reproducibility.

Contribution

It provides the first rigorous empirical benchmark comparing human and algorithmic predictions in jury selection using controlled data and statistical evaluation.

Findings

ML models outperform jury consultants in prediction accuracy

Supervised ML models offer greater transparency and reproducibility

Provides an empirical benchmark for human versus machine judgment in legal contexts

Abstract

Prior studies on the effectiveness of professional jury consultants in predicting juror proclivities have yielded mixed results, and few have rigorously evaluated consultant performance against chance under controlled conditions. This study addresses that gap by empirically assessing whether jury consultants can reliably predict juror predispositions beyond chance levels and whether supervised machine-learning (ML) models can outperform consultant predictions. Using data from N mock jurors who completed pre-trial attitudinal questionnaires and rendered verdicts in a standardized wrongful-termination case, we compared predictions made by professional jury consultants with those generated by Random Forest (RF) and k-Nearest Neighbors (KNN) classifiers. Model and consultant predictions were evaluated on a held-out test set using paired statistical tests and nonparametric bootstrap procedures. We find that supervised ML models significantly outperform professional jury consultants under identical informational constraints, while offering greater transparency, replicability, and auditability. These results provide an empirical benchmark for evaluating human judgment in jury selection and inform ongoing debates about the role of data-driven decision support in legal contexts. To support reproducibility and auditability, all code and data will be made publicly available upon publication.