The Multiple Effects of Casualties on Public Support for War: An Experimental Approach

Abstract

P ublic support for a conflict is not a blank check. Combat provides information people use to update their expectations about the outcome, direction, value, and cost of a war. Critical are fatalities--the most salient costs of conflict. I develop a rational expectations theory in which both increasing recent casualties and rising casualty trends lead to decreased support. Traditional studies neither recognize nor provide a method for untangling these multiple influences. I conduct six experiments, three on the Iraq War (two with national, representative samples) and three with a new type of panel experiment design on hypothetical military interventions. The results of hazard and ordered logit analyses of almost 3,000 subjects support a rational expectations theory linking recent casualties, casualty trends, and their interaction to wartime approval. I also examine the effects of the probability of victory, information levels, and individual characteristics on the support for war, and contrast results from representative and convenience samples.War has an immediate effect upon the attitude of mind of everyone who is brought into connection with it.

Wars, Presidents, and Popularity: The Political Cost(s) of War Re-Examined

“…Later studies consistently follow this choice (e.g., Kernell, 1978;Gronke and Newman, 2003;Eichenberg and Stoll, 2006;Kriner, 2006;Karol and Miguel, 2007;Gartner, 2008). Still, casualty numbers are not the only war characteristic affecting people's opinion about their leader.…”

“…Extensive research shows that the number of casualties suffered in military conflicts undermines incumbent popularity (e.g., Mueller, 1973;Kernell, 1978;Ostrom and Simon, 1985;Gronke and Newman, 2003;Kriner, 2006;Eichenberg and Stoll, 2006;Voeten and Brewer, 2006;Karol and Miguel, 2007;Gartner, 2008). While casualty counts capture the public's appropriate sensitivity to human losses, an exclusive focus on such data to assess the political cost of wars might be incomplete.…”

“…counts (e.g., Gartner andSegura, 1998, 2000), we report results using both indicators. In line with previous work, we expect casualties to depress incumbent popularity (i.e.…”

Assumed Transmission in Political Science: A Call for Bringing Description Back In

“…Theoretically, the war support literature often interprets this pattern as evidence that the American public perceives fewer benefits from war as its human costs mount. This presumed relationship is usually tested with aggregate data, using time-series models predicting war support as a function of the actual occurrence of American war deaths, or with experiments that ensure subjects are exposed to casualty information (e.g., Gartner 2008;Gelpi 2010). However, ordinary Americans learn about casualties not from time-series datasets or from experimental interventions but from news coverage.…”

Cry Havoc: Rhetorical Mobilization and Foreign Policy Decision Making During War-threatening Crises

“…Mueller introduces two hypotheses that continue to attract scholarly attention. The "casualties hypothesis" (Burk 1999) claims that public support for war is negatively correlated with of casualties (contributors and critics include Berinsky 2007;Gartner 2008;Gelpi, Feaver & Reifler 2005;Kull & Ramsay 2001;Slantchev 2004). The "rally-'round-the-flag effect,"…”