Measuring Legislators’ Ideology Using LLMs: Text Analysis of Korean National Assembly Transcripts

View at Publisher

Ask AI

Bookmark

Share

Key Points

This research aims to propose a new methodology for measuring legislators' ideology using large language models (LLMs).
Analyzed transcripts from the 17th to 20th National Assembly sessions.
Utilized the GPT-4.1-mini model to assign continuous ideology scores ranging from -1 (strongly progressive) to +1 (strongly conservative).
Compared LLM-based scores with traditional W-NOMINATE scores.
LLM-measured ideology showed a unimodal distribution, contrasting with the bimodal distribution of W-NOMINATE scores.
A significant convergent validity between LLM-measured ideology and W-NOMINATE scores was observed, especially in recent Assemblies.
Regression analysis indicated LLM scores consistently predicted W-NOMINATE scores without systematic bias by political faction.

Measures legislators’ ideology in Korean National Assembly, indicating a novel method for analysis.

No abstract yet. The journal hasn't published one we can show.

No takes yet. Share an insight, caveat, or question.

Jeon et al. (2026) studied this question.

View at Publisher

Ask AI

Bookmark

Share