Ideology and Policy Preferences in Synthetic Data
The Potential of LLMs for Public Opinion Analysis
Bibliographic Data
| ID | 21224969 |
|---|---|
| Authors | Keyeun Lee (0009-0001-2794-5712), Jaehyuk Park (0000-0001-8227-1143), Suh-hee Choi (0000-0002-4711-7383), Changkeun Lee (0000-0002-5508-1122) |
| Year | 2025 |
| Volume | 13 |
| Publication date | 2025-05-22 |
| Open Access | Yes |
| Type | ARTICLE |
| Venue | Media and Communication (JOURNAL) |
| Journal identifiers | ISSN: 2183-2439 • E-ISSN: 2183-2439 |
| Publisher | Cogitatio (PUBLISHER • PT) |
| DOI | 10.17645/mac.9677 |
| OpenAlex | W4410566969 |
| Language | EN |
| Citations received | 2 |
| References cited | 26 |
This study investigates whether large language models (LLMs) can meaningfully extend or generate synthetic public opinion survey data on labor policy issues in South Korea. Unlike prior work conducted on people’s general sociocultural values or specific political topics such as voting intentions, our research examines policy preferences on tangible social and economic topics, offering deeper insights for news media and data analysts. In two key applications, we first explore whether LLMs can predict public sentiment on emerging or rapidly evolving issues using existing survey data. We then assess how LLMs generate synthetic datasets resembling real-world survey distributions. Our findings reveal that while LLMs capture demographic and ideological traits with reasonable accuracy, they tend to overemphasize ideological orientation for politically charged topics—a bias that is more pronounced in fully synthetic data, raising concerns about perpetuating societal stereotypes. Despite these challenges, LLMs hold promise for enhancing data-driven journalism and policy research, particularly in polarized societies. We call for further study into how LLM-based predictions align with human responses in diverse sociopolitical settings, alongside improved tools and guidelines to mitigate embedded biases
Development economics · Economics · Ideology · Political economy · Political science · Politics · Positive economics · Public economics · Public opinion · Social science · Sociology · Computational and Text Analysis Methods · Electoral Systems and Political Participation · Law · Opinion Dynamics and Social Influence · Psychology · Social Psychology
Can AI language models replace human participants?
Using cognitive psychology to understand GPT-3
Human or machine? The perception of artificial intelligence in journalism, its socio-economic conditions, and technological developments toward the digital future
AI language models cannot replace human research participants
Policy Ideology in European Mass Publics, 1981–2016
Cultural Foundations of Contentious Democracy in South Korea
Out of One, Many
Political polarization in Korea
| Citations per year | 2 |
|---|---|
| Citation span | 2025 - 2026 (2) |
| Highly cited | No |