Bounded Watchdogging: Platform Governance, Risk Structures, and the Reconfiguration of Science Journalism

This article examines six cross-platform biomedical research-integrity controversies in mainland China to analyze how open-source investigation becomes integrated with journalistic verification under platform governance. Drawing on a research design that combines multiple qualitative methods such as digital trace ethnography, process tracing, and semi-structured interviews, the study identifies a four-stage process of science watchdogging: initial disclosure, verification and evidence accumulation, media intervention, and institutional response. This process shows how journalists use open-source information to conduct investigative science reporting. Building on these findings, the article further proposes a model of bounded watchdogging organized around trade-offs among speed, verifiability, and risk, showing how public-facing online leads and digital traces can be transformed into journalistic evidence. In this way, the findings not only offer journalists a practical verification pathway for producing science investigations, but also extend watchdog journalism theory beyond liberal-adversarial assumptions in the Chinese context.