Bridging Public Engagement and Policy Mechanisms: How Citizen-Generated Data Shapes Environmental Policy Design

Citizen-generated data plays a pivotal role in improving environmental policy design, refining policy instrument selection, and enhancing governance precision. However, the specific mechanisms through which such data influences policy processes remain underexplored. This study utilizes the BERT language model to classify public sentiment, incorporating it as a labeled variable within the policy process. Using data from Fuzhou’s 12345 public service platform, the research investigates the role of citizen-generated data in shaping environmental policy, specifically focusing on black-odor water management. The findings show that government responses to public demands are influenced by institutional trust, willingness to participate, and individual normative adherence. For individuals with high institutional trust, governments tend to use nudging strategies for psychological engagement, while those with strong willingness to participate see expedited administrative interventions. Long-term complainants, particularly those motivated by self-interest, are more likely to receive informational responses over direct administrative enforcement. Additionally, the influence of citizen-generated data on policy design is mediated through three key mechanisms: the focal effect, inertia effect, and cost effect. This study underscores the need for a robust link between citizen-generated data and policy design, emphasizing the importance of emotional governance and AI-driven strategies to improve the effectiveness of environmental policymaking.