Validating LLM-Based Digital Interviews for Individual-Level Public Opinion: A Chilean Election Study

OSF Preprints (OSF Preprints) 2026 Preprint

Authors: Matias Fuentes Becerra, Raymond Low, Piotr Kotlarz, raymond duch, Felipe Torres Raposo, Roberto Cerina
Cited by: 0

Abstract

A study to evaluate how accurately large language models (LLMs) can replicate individual-level political attitudes and demographics by analyzing social media data. The project builds on recent work by (Cerina and Duch ,2025 and Cerina and Duch, 2026) that used LLM-driven “digital interviews” of social media users to estimate public opinion and forecast elections.

Fields of Study & Subjects

Computational and Text Analysis Methods Sentiment Analysis and Opinion Mining Populism Right-Wing Movements

Publication details

Journal / source
OSF Preprints (OSF Preprints)
Publication date
3 January 2026
Document type
Preprint