Selective Access to and Avoidance of Political Content Online?

Seattle.
The next speaker at AoIR 2011 is Ericka Menchen-Trevino, whose focus is on media selection practices online. She begins by noting the concerns that people don’t necessarily gain a full understanding of current political trends online, if they flock only to those Websites which already speak to their political preferences; this may give them a fundamentally skewed perspective on politics. Additionally, of course, people may also avoid exposure to political news altogether – so there’s a two-dimensional framework here, from low to high political partisanship and from low to high interest in the news.

There is also selective exposure, selective perception, and selective retention. In the first place, people may seek or avoid content which challenges their political views; indeed, the two don’t necessarily go together: just seeking out content even without avoiding other content already generates a selective exposure.

Ericka studied this through real-world observational Web data generated through her experimental proxy software, which tracks participating users’ Web activities, against the backdrop of the 2010 Illinois election campaign. Her participants were 30 years or older registered voters from the Chicago area; she focussed on 46 non-Net-native users with a range of political interests, gender, education, and other personal attributes. She tracked their Web usage, as well as conducting interviews with these participants.

Ericka tested participants’ awareness of key topics to test for selective avoidance; she also measured the percentage of political content containing specific keywords (e.g. ‘Obama’) within the total number of pages accessed by individual users, and studied the content of the visited Web pages which mentioned specific Illinois candidates to identify any political preferences.

17% of Ericka’s respondents split their votes between Democrat and Republican candidates; 26% of those who voted decided their vote during the campaign – so in those cases, not much political content avoidance is likely to have taken place. Interest in government and politics, online news access, ranking of local news amongst pages seemed not to have been significant indicators of avoidance of politics, either. Watching local TV news, the level of political interest, online news browsing, and the ranking of local and metropolitan news amongst the news type was significant as an indicator of campaign avoidance, however.

Avoiding the other side of politics took place in stops and starts. Some Democrats were completely unaware of the strong right-wing leanings of Fox News, for example; others were more selective, seeing their choices variously as ‘biased’ or just ‘better’. Very few visited overtly partisan campaign Websites; several read negative stories about candidates they voted for. So, participants had different levels of exposure to campaign claims, partly driven by interest and news media consumption routines; the Web was not promoting polarisation but such polarisation is possible at the level of retention.