| Despite an abundance of literature demonstrating that immigration and crime are unassociated, public opinion often reflects the contrary. I examine a source that could contribute to this disconnect between research and public opinion -- media framing -- particularly, how the specific way that news outlets talk about immigration and crime, along with where they are located geographically, influence how prominently these stories are covered. I employ content analysis of newspaper articles from 2008-2012, which I geo-locate and pair with structural covariates gathered from several other data sources. I use multilevel models to analyze the effect of article-level framing and county-level contextual characteristics on article prominence in newspapers. Findings reveal that newspapers in counties with less immigrants and less crime are more likely to prominently feature articles discussing immigration and crime. Furthermore, articles with negative frame of immigration-crime are more likely to be put on the front page, regardless of contextual characteristics. I discuss implications for literature and policy, along with limitations of my study and suggestions for future research. |
