Does Presidential Rhetoric Matter? Priming and Presidential Approval

Does presidential rhetoric affect presidential approval? Surprisingly, virtually no research has addressed this question--despite widespread recognition that presidents invest substantial resources to perfect their rhetoric (Edwards 2002), and clear evidence that approval fundamentally affects the president's power and policy-making success (Neustadt 1960; Canes-Wrone n.d.). In this article, we use a multi-method approach to demonstrate that presidents can use rhetoric to shape their own approval. What the president says matters for what the public thinks of him. We begin in the next section by discussing presidential approval and the effect rhetoric might have on approval. We then use a content analysis of a presidential speech, a laboratory experiment, and a nationally representative survey to test our expectations. We show that the president can influence his own approval by priming the standards on which he is evaluated. We also introduce the idea of image priming and explore how political knowledge affects issue and image priming (in different ways). Our results add a new dimension to the study of presidential approval, extend work on priming and public opinion, and raise intriguing questions about accountability. The Study of Presidential Approval Work on presidential approval constitutes one of the most progressive research endeavors in political science. The bulk of this scholarship focuses on aggregate trends in approval (Gronke and Newman 2003), documenting the causal importance of the economy, wars, media coverage, inter alia (e.g., Kernell 1978; Edwards 1990; Edwards et al. 1995; Nicholson et al. 2002). "Presidential drama," such as the occurrence of a major speech, also can impact trends in approval (e.g., Brace and Hinckley 1993). This suggests that presidential actions--but not necessarily the content of what the president actually says, once he decides to give a speech--can affect approval (McGraw et al. 1993). (1) Other work explores the dynamics behind individual level approval (e.g., Mutz 1994; Edwards et al. 1995). For example, media coverage can influence individual level evaluations of the president (e.g., Iyengar and Kinder 1987). Gronke and Newman (2003, 22) explain, however, that, relative to research on aggregate trends, "it is surprising ... that so little is known about the individual level determinants of presidential approval." Another surprising aspect of scholarship on presidential approval is that it has not investigated how presidential rhetoric (i.e., what the president says) affects approval. (2) Edwards and Eshbaugh-Soha (2000, 4-3) explain that scholars "make numerous inferences regarding the impact of the president's rhetoric on public opinion [but they] virtually never provide evidence for their inferences about the president's impact. ... [Many studies] have examined public evaluations of the president, but not the president's influence on those evaluations." In what follows, we fill these gaps by exploring how presidential rhetoric affects individual level approval. Priming Approval How might presidential rhetoric affect approval? We build on media effects research to argue that the presidential rhetoric shapes approval via priming. We next describe priming theory and how we propose to extend it. What Is Priming? Miller and Krosnick (2000, 301) explain that "[p]riming occurs when media attention to an issue causes people to place special weight on it when constructing evaluations of over-all presidential job performance" (Iyengar et al. 1984). Scholars have amassed a large body of experimental and survey evidence of media priming (e.g., Iyengar and Kinder 1987; Krosnick and Brannon 1993; Miller and Krosnick 2000). For example, individuals exposed to news stories about defense policy tend to base their overall approval of the president on their assessment of the president's performance on defense. Thus, if these individuals believe the president does an excellent (poor) job on defense, they subsequently display high (low) levels of overall approval. …