Yunus Emre Tapan (Northeastern University) Zeynep Elif Koç (Atilim University) Tuba Unlu Bilgic (University of St Thomas)
Abstract
Measuring public opinion in International Relations has attracted increasing scholarly attention, with anti-Americanism emerging as a pivotal area of focus. This paper reviews existing frameworks for conceptualizing and measuring anti-Americanism and introduces a novel approach utilizing Large Language Models (LLMs). By examining Turkish public opinion over the past decade, the study employs a mixed-methods design that integrates qualitative insights with computational techniques to capture the multifaceted nature of this phenomenon. The paper demonstrates that LLMs, when guided by a well-constructed and contextually tailored codebook, can serve as a reliable and effective complementary tool for analyzing public opinion. It concludes by encouraging scholars to harness the reasoning capabilities of LLMs to advance the study of complex social phenomena, offering a pathway for integrating traditional and computational methodologies in social science research.