Partisan Identification as a Predictor of Cortisol Response to Election News

Abstract

Partisan effects on media consumption have been widely documented, with considerable attention given to partisan influences on selective exposure. Although researchers have debated the degree to which selective exposure drives media consumption in general and partisan consumption of the news in particular, one of the hypothesized mechanisms linked to this phenomenon seems inherently plausible: Exposure to disliked news coverage can generate psychological discomfort. The physiological effects of this hypothesized discomfort were examined by determining how political partisanship influences release of the stress hormone, cortisol, following exposure to news coverage of a presidential election. The study was conducted in the week following the 2008 election. Participants were students at a large, mostly conservative state university who read news coverage about the election victory of Barack Obama or a set of control news stories. Results indicated that conservative political identification was associated with more negative and less positive emotional responses and with a spike in salivary cortisol levels. Contrary to predictions, however, the cortisol spikes appeared to operate independent of self-reported emotional distress. The implications of these results are considered as they relate to selective exposure and the physical health of partisans who follow political news.

Keywords:

• partisanship
• political attitudes
• political communication
• social stress
• psychological stress
• cortisol

Notes

1. See http://www.princetonreview.com/best-press-release.aspx.

2. Participants observed a series of stimulus words on a computer screen (e.g., eagle) and were asked to list the first association that came to mind (e.g., bird). No feedback was given regarding the appropriateness of associations. The advantage of including this pilot is that it helped establish a stable baseline by engaging participants in a reasonably neutral task that potentially distracted them from outside stressors.

3. Empirically it may be difficult to examine this question in the weeks following a presidential election. It is quite conceivable that even seemingly “neutral” political coverage would prime concerns about the recent election outcome. It was out of consideration of this issue that we invested our statistical power in the simple two-group comparison in this investigation and did not include any political content in the control group.

4. A potential extension of this work will be to see if physiological stress predicts other forms of political avoidance and disengagement, such as is found with low voter turnout. Whereas this issue is typically framed in terms of “apathy” or disinterest, our framework suggests that it might be the highly invested individuals who avoid the ballot box, as a means of managing stress (see French, Smith, Guck, Alford, & Hibbing, 2011; Lavine et al., 1999; Waismel-Manor et al., 2011).