abstract
- Do surveys provide reliable information on public opinion in interstate disputes? Politicians and critics frequently rely on surveys to understand public policy preferences in disputes. We provide empirical evidence that casts doubt on the reliability of such surveys: respondents may falsify their preferences when they favor contentious approaches, such as aggressive military activities and concessions to adversaries. Concessions are often perceived as anti-societal, as they sacrifice national interests for the benefit of an adversary, while aggressive military options conflict with widely shared anti-war norms prevalent across many societies. These approaches are therefore deemed socially undesirable, and the members of society feel pressure to conform their expressed preferences accordingly, even in anonymous online settings. A list experiment evaluating Japanese voters¿ opinions on the territorial dispute with China shows that the use of a direct question format underestimates support for territorial concessions and military base construction on the disputed territories by 16¿20 percentage points compared to the list format. These findings carry important implications for scholars, policymakers, and practitioners. Democratic elites who rely on survey data to gauge public preferences may systematically misread their constituents, potentially missing both opportunities for peaceful resolution and risks of unexpected public support for escalation.