Cavanah, Sarah2016-09-192016-09-192016-06https://hdl.handle.net/11299/182213University of Minnesota Ph.D. dissertation. June 2016. Major: Mass Communication. Advisors: Brendan Watson, Seth Lewis. 1 computer file (PDF); v, 149 pages.For nearly half a century, mass media research has understood that news content that is conflict-oriented is more likely to appear in news outlets produced in structurally heterogeneous communities than in structurally homogeneous communities. For individuals living in homogeneous communities, their ability to find news about their own communities in regional-level media — especially metropolitan-based newspapers — has meant that they had access to conflict-oriented news. However, decades of financial pressures have weakened regional metros’ ability to continue to distribute to and cover non-metropolitan communities, especially the structurally homogeneous communities most reliant on a regional source for conflict-oriented news. When metro newspapers pull back distribution and coverage of non-metro communities, residents are expected to display lower levels of political knowledge, and preliminary studies have suggested lower political participation as well. This dissertation uses computer-assisted content analysis and multilevel statistical modeling to add nuance to the understanding of structural pluralism and metropolitan coverage, as well as specifically support the assumption that news content about specific communities contributes to voter turnout in those communities.encomputer-assisted content analysisknowledge gapnews effectsnewspaper pullbackpolitical engagementstructural pluralismThesis or Dissertation