Scratching Below the Core Network Surface: Capturing Hidden Exposure to Political Disagreement and Difference
23 Pages Posted: 15 Jun 2016
Date Written: June 13, 2016
One of the most common means by which network scholars have studied exposure to disagreement is through the name generator (e.g., Huckfeldt & Sprague, 1995; Mutz, 2006). Unfortunately, name generator methods suffer from a number of limitations, including severe limitations in the number of alters who can be identified. Accurately assessing the extent of political disagreement in discussion networks has become an important topic of recent research. Building on this work, the present paper used data from two national surveys – the 2012 U.S. Comparative National Elections Project (CNEP) survey and a May 2015 Race & Politics representative survey including a black oversample – to test the potential of a simple and efficient modification of name generator methods to probe for “hidden” exposure to difference. The findings highlight limitations of the traditional name generator, and suggest promise for the new approach. Over half of those for whom name generator results suggest no exposure to disagreement were able, in a follow-up, to name a discussion partner who supported an alternative candidate (2012 CNEP), and over a quarter of those whose name generator results indicate no exposure to a racially different alter were able to name someone of the opposite (Black vs. White) race with whom they’d talked politics (2015 Race & Politics). These findings suggest that mixed or diverse networks are far more common, and purely homophilous networks are far less common, than prior name generator research (Mutz, 2006) suggests. The implications for network measurement and democratic theory are discussed.
Keywords: networks, race, disagreement, name generator
Suggested Citation: Suggested Citation