Why are certain movies more successful in some markets than others? Are the entertainment products we consume reflective of our core beliefs? To answer these questions, we analyze international movie data and local tales using machine learning to measure a plot’s similarity to traditional motifs. Globally, films mirroring local folklore get wider screenings and earn more. Across US markets, movies matching ancestral stories attract more interest. Finally, films endorsing risktaking, vengeance-seeking, and traditional gender roles outperform across markets holding similar historical and contemporary norms. Choosing to watch modern adaptations of stories passed down from our ancestors on the big screen reinforces cultural identity. . Traditional opinion polling faces growing challenges such as declining response rates. Meanwhile, vast amounts of user-generated text—once seen as purely qualitative—offer untapped potential for understanding public opinion. This talk explores how large language models can transform free-form text into structured, survey-like insights, bridging the gap between qualitative and quantitative analysis. We’ll discuss the limitations of classical polling, the promise of AI-driven alternatives, and what this shift means for the future of survey research.
Rauh et al. (Tue,) studied this question.