| Sumario: | Conflict scholars commonly employ public opinion surveys to understand the causes and consequences of violence. However, surveying in wartime presents a distinctive set of challenges. We examine two challenges facing polling in countries at war: under-coverage of national samples and response bias. Although these issues are acknowledged in the literature on surveying methods, they become significantly more pronounced in war zones due to the geographic clustering of violence and the heightened sensitivity surrounding certain opinions. We illustrate these challenges in the context of the ongoing Russia–Ukraine war, drawing on original panel survey data tracing the attitudes of the same people in Ukraine prior to and after Russia's full-scale invasion in 2022. We show that unit and item non-response bias in surveys conducted during the war are related to respondents' political orientation, particularly their support for NATO membership measured in 2019. We conclude with lessons for those employing survey methods in wartime, and point to steps forward, in Ukraine and beyond.
|
|---|