Detecting Statistically Significant Fairness Violations in Recidivism Forecasting Algorithms
This research introduces a rigorous statistical framework using k-fold cross-validation and hypothesis testing to distinguish true algorithmic bias from samp...