Personalizing Ideology in the News: The Impact of News Aggregators and Collective Filtering

Abstract

This paper finds state-specific 2- and 3-word phrases that reflect the Republican and Democratic language most commonly used by Congress people from that state by performing content analysis regressions on the 2009-2010 Congressional Record. The relative percent of Republican “buzz word” search for each metro area within a state is then compared to the percent of the population from that metro who voted Republican in the 2008 Presidential Election. I find a significant correlation between the percentage of Republican voters in a metro and the relative percentage of Republican search terms in that same area.