A Transdisciplinary Framework for Evidence-Based Governance

Abstract

Evidence-based governance has become a pervasive aspiration across public administration, regulatory policy, and socio-technical oversight, yet its operational meaning remains fragmented across disciplines and implementation contexts. This paper develops a transdisciplinary framework for evidence-based governance that treats evidence not as a static input to decision-making but as a continuously managed socio-technical infrastructure. Drawing on systems engineering, public policy, science and technology studies, organizational theory, and responsible artificial intelligence, we conceptualize evidence production and use as an end-to-end lifecycle spanning problem framing, data generation, model construction, evaluation, deployment, and iterative revision under conditions of uncertainty, contestation, and asymmetric power. The proposed framework foregrounds structural trade-offs among legitimacy and efficiency, robustness and adaptability, transparency and privacy, fairness and feasibility, and central coordination and local autonomy. We argue that evidence-based governance requires architectural commitments—standards, institutions, accountability mechanisms, and participatory channels—that stabilize epistemic quality while remaining resilient to shifting objectives, adversarial behavior, and distributional impacts. We elaborate design principles for evidence infrastructures, discuss governance patterns for complex systems, and offer case illustrations from public health, climate risk governance, algorithmic regulation, and urban service delivery. The paper concludes with an agenda for sustainable evidence ecosystems that integrate continuous evaluation, democratic contestability, and operational robustness, enabling governance systems to learn while preserving public trust.