The constituent foundations of the rally-round-the-flag phenomenon

Despite the massive attention heaped on the rally-round-the-flag phenomenon by public opinion scholars, relatively little attention has been paid to its constituent elements. Yet, recent research has found that different groups of Americans respond differently to presidents' activities according to their interests and attentiveness. In this study, I disaggregate public opinion along two dimensions: political party and political sophistication. I argue that in responding to high-profile presidential activities abroad, different groups of Americans weigh various individual, contextual, and situational factors differently. I investigate all major U.S. uses of force between 1953 and 1998 and find that the propensity of different groups to rally does indeed vary according to individual and environmental circumstances. To explain these differences, I employ two models of public opinion. The first emphasizes the importance of threshold effects in explaining opinion change. That is, individuals who are closest to the point of ambivalence between approval and disapproval are most likely to change their opinion in response to external circumstances. The second emphasizes both the propensities of different types of individuals to be exposed to a given piece of information, and their susceptibility to having their opinion influenced by ally additional information. My results offer a more nuanced picture of the nature and extent of the rally phenomenon than has been available in previous studies. My findings also hold important implications for other related scholarly debates, such as whether, and under what circumstances, the use of force can successfully divert public attention from a president's domestic political difficulties.