Does Accountability Reporting Undermine Trust in the Media?

The contemporary crisis of democracy is deeply intertwined with a crisis of communication, particularly around journalism’s capacity to hold state power accountable while maintaining public trust. While political communication research has extensively documented how news exposure shapes citizens’ attitudes toward political institutions, far less attention has been paid to how coverage of institutional wrongdoing feeds back into perceptions of journalism itself. This paper addresses this gap by examining how media coverage of police misconduct shapes not only attitudes toward the police, but also evaluations of the media’s watchdog role, credibility, and democratic function. Theoretically, we integrate research on media trust, accountability journalism, and affective polarization to argue that reporting on state violence constitutes a critical stress test for democratic communication. Coverage of police misconduct places journalism at the intersection of competing normative expectations: exposing abuse of power, maintaining factual balance, and avoiding perceptions of ideological bias. We argue that how media navigate this tension—through framing choices, evidentiary balance, and sourcing—can either reinforce journalism’s democratic legitimacy or further erode trust in media institutions, especially among audiences polarized over law enforcement. Empirically, we implement a factorial vignette experiment embedded in a nationally representative French survey. Respondents are randomly exposed to a simulated news article reporting on police misconduct across three institutional contexts: a routine traffic stop, policing during a public demonstration, and the handling of a judicial complaint involving sexual violence. The design varies along two dimensions: (1) the severity of the misconduct (verbal abuse, physical violence with injury, and severe violence resulting in life-threatening harm), and (2) the balance of factual framing (anti-police, neutral, or pro-police). The experiment also incorporates latent treatments capturing sourcing cues and the alignment between respondents’ prior attitudes toward the police and the framing of the article. We measure perceived message credibility, normative evaluations of the media’s accountability role toward state institutions, and trust in the police. This design allows us to assess how journalistic balance shapes perceptions of journalism itself, moving beyond traditional media-effects models and highlighting the reflexive relationship between journalism and democratic legitimacy. Our findings will have direct implications for debates on democratic erosion and renewal, by speaking to concerns about affective polarization and the erosion of shared standards for credible journalism, showing how identical facts can differentially shape trust in the media depending on framing and prior attitudes. They will also inform discussions on accountability journalism by identifying the conditions under which reporting on state misconduct strengthens—or undermines—confidence in journalism’s democratic role. By providing comparative evidence from France, this study extends a largely U.S.-centric literature on media trust and political communication, and by foregrounding how citizens evaluate journalism itself when democracy is “under stress,” it advances research on media trust, accountability, and democratic resilience by showing how journalism’s perceived legitimacy as a democratic counterweight to state power can be reinforced—or undermined—through media coverage of institutional wrongdoing.