Assess students' ability to analyze algorithmic fairness trade-offs in the specific context of predictive policing by: explaining and contrasting key statistical fairness definitions (e.g., equalized odds, predictive parity, calibration), evaluating how these metrics conflict in practice, identifying societal impacts (e.g., disparate impact, feedback loops, community trust, legal considerations), and proposing technically and ethically grounded mitigation strategies (preprocessing, in-processing constraints, post-processing, policy interventions). Include interpretation of model outputs, dataset bias sources, case study critique, and justification for recommended trade-offs and evaluation criteria.