Seminar

Attitudes toward immigration rank among the most politically consequential issues in contemporary Western democracies, yet the mechanisms through which media coverage shapes those attitudes remain incompletely understood. Existing research has documented a robust association between migration-related news coverage and people’s opinions, but by relying on aggregate national press data without distinguishing between media consumers and non-consumers, it implicitly conflates two theoretically distinct processes.

This paper introduces a two-channel framework separating the direct effect of media coverage on newspaper readers from an indirect effect operating through the local media information environment, whereby non-readers are reached via peer transmission rather than direct exposure. I argue that local news provides an ideal testing ground for this framework, as the geographic boundedness of local newspaper markets aligns the media environment with the social community through which indirect effects travel.

I test this framework by merging a content analysis of migration-related coverage across 120 local German newspapers with individual-level panel data from the German Socioeconomic Panel (SOEP). A panel fixed effects design with individual and wave fixed effects exploits within-person variation in attitudes over time, while a reader/non-reader interaction term operationalises the two-channel distinction. Preliminary results suggest that local migration coverage shapes attitudes among both readers and non-readers, though the relative magnitude of the two channels remains under investigation. By decomposing media effects into two theoretically distinct pathways, this paper advances our understanding of how news coverage shapes public opinion beyond the boundaries of the readership itself.