Reports of Media Polarization Effects Have Been Greatly Exaggerated

Via Brendan Nyhan on Twitter, a new paper by Mike LaCour arguing that partisans are far less likely to consume exclusively partisan television news than many accounts would have it.

Results indicate that even under the most lenient denition of ideological selectivity { greater than 50% of news consumption from like-minded programming { only a small subset of individuals (14% of Democrats and 24% of Republicans) engage in this behavior. Using the most stringent definition of of ideological selectivity { over 90% of news consumption from like-minded programming includes an even smaller subset of individuals (3% of Democrats and 7% of Republicans). The evidence presented thus far suggests that a lot of people avoid partisan news sources altogether in favor of centrist mainstream sources. The majority of individuals are far from having an exclusive liberal or conservative news diet.

LaCour provides this graph of news consumption, which suggests that on average Republicans have a slightly more ‘balanced’ news diet than Democrats, but that there is a significant group of Republicans (the bump towards the right) who consume heavily partisan media.

As an aside: LaCour lists the 2010 paper that Eric Lawrence, John and I wrote on blog consumption as one of the pieces making the case for media polarization. This isn’t quite the argument we’re making – we are quite specific in claiming that our data characterizes readers of political blogs (a small subset of the population) rather than the population as a whole. As I’ve argued in a overview paper:

[t]hese results may tentatively suggest that a more diffuse version of Prior’s sorting process is taking place. On the one hand, there is a broad population, which includes many partisans but is exposed to a broad variety of online sources with different ideological perspectives. On the other hand, there is a more speciï¬c and highly politically aware subgroup that preferentially seeks out partisan information via blogs and other means and is less likely to be exposed to dissenting opinions. Although the latter group is far smaller than the former, it is likely to be more involved in politics, and hence more influential than its size would suggest. The Sunstein hypothesis is more likely to apply to the small group of highly politically aware people who, e.g., read blogs than to the wider population described by Gentzkow & Shapiro (2010).

Update: Have restored this post since people linked to it when it was up, but it turns out that John already blogged an earlier version of this paper back in 2010.

[Cross-posted at The Monkey Cage]

Henry Farrell

Henry Farrell is an associate professor of political science and international affairs at George Washington University.