Aimed Emotions: Strategic Affective Appeals Across Diverse Campaign Media

• In Event: Understanding Emotions

Abstract

Emotional appeals are essential tools for political candidates. Drawing on affective intelligence theory, a rich empirical literature has demonstrated the distinct effects of different emotional appeals on mass political behavior, namely that appeals to fear provoke information-seeking and reduced reliance on routine (i.e., partisan) habits, whereas positive appeals to enthusiasm and negative appeals to anger have the opposite effect. This raises a question of whether political candidates are actually strategic in deploying emotional appeals in their campaign messages. One possible means of strategic behavior is for candidates to take advantage of audience characteristics. For example, because audiences on social media platforms like Twitter are heavily skewed towards partisan supporters, enthusiasm and anger appeals should dominate; in contrast, candidates’ messages on the debate stage should more frequently involve fear appeals to promote political learning. Similarly, because television advertisements in battlegrounds and uncontested media markets serve different strategic purposes, emotional appeals in advertising should vary with the partisan lean of a particular media market. In this paper, I use a machine learning approach to classify the emotional content of tweets and debate statements by presidential candidates in the 2012, 2016, and 2020 elections. Combining these data with emotional classification of television ads from the Wesleyan Media Project, I investigate whether presidential candidates exhibit strategic use of emotional appeals across message channels, and whether these strategic patterns conform to the expectations of affective intelligence theory.

Author

• Andrew R.L. Trexler, Duke University