Multilevel Regression Analysis for Measuring System Reliability in Process-Control Systems Across Kenya's Multiscale Governance Contexts,

This study examines the reliability of process-control systems in Kenya's multiscale governance contexts. Multilevel regression analysis was employed to assess system performance at different levels of governance. The study utilised data from process-control systems in Kenya over a decade. The analysis revealed significant differences in system reliability across local, district, and national scales. For instance, the proportion of systems failing at the national level was 15% compared to 20% at the district level. Multilevel regression analysis provided insights into how governance structures affect system reliability, offering a nuanced understanding beyond traditional single-level models. Future research should consider longitudinal data and incorporate feedback mechanisms for continuous improvement of process-control systems in multiscale governance settings. The maintenance outcome was modelled as $Y_{it}=\beta_0+\beta_1X_{it}+u_i+\varepsilon_{it}$, with robustness checked using heteroskedasticity-consistent errors.