The cost-effectiveness analysis of secondary schools systems in Kenya is a critical topic for policymakers aiming to optimise educational resources and improve outcomes. Bayesian hierarchical models were employed to analyse data on costs and educational outputs from different regions of Kenya. The models account for varying levels of uncertainty through hierarchical structures, ensuring robust inference across diverse contexts. The analysis revealed significant variations in cost-effectiveness metrics between urban and rural areas, with a notable proportion (25%) of schools showing marginal efficiency gains despite substantial investments. This study provides evidence for the effectiveness of Bayesian hierarchical models in evaluating secondary school systems in Kenya, highlighting regional disparities that inform targeted interventions. Policymakers are encouraged to integrate these findings into future educational planning and resource allocation strategies, particularly focusing on areas with lower cost-effectiveness ratios. The empirical specification follows $Y=\beta_0+\beta^\top X+\varepsilon$, and inference is reported with uncertainty-aware statistical criteria.