Positioning Political Texts with Large Language Models by Asking and Averaging
Political Analysis · 2025-07 · 33 (3) · 274-282
DOI: 10.1017/pan.2024.29 ↗Kurz gefasst
Fachartikel in der Zeitschrift Political Analysis, online erschienen am 27.01.2025. Verfasst von Gaël Le Mens und 1 weiteren Autor:innen. Die Arbeit wurde bisher 42-mal zitiert (Crossref) und nennt 19 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Directorate-General XII, Science, Research, and Development, Ministerio de Ciencia y Tecnología, Ministerio de Ciencia, Innovación y Universidades.
Abstract (Originalsprache)
We use instruction-tuned large language models (LLMs) like GPT-4, Llama 3, MiXtral, or Aya to position political texts within policy and ideological spaces. We ask an LLM where a tweet or a sentence of a political text stands on the focal dimension and take the average of the LLM responses to position political actors such as US Senators, or longer texts such as UK party manifestos or EU policy speeches given in 10 different languages. The correlations between the position estimates obtained with the best LLMs and benchmarks based on text coding by experts, crowdworkers, or roll call votes exceed .90. This approach is generally more accurate than the positions obtained with supervised classifiers trained on large amounts of research data. Using instruction-tuned LLMs to position texts in policy and ideological spaces is fast, cost-efficient, reliable, and reproducible (in the case of open LLMs) even if the texts are short and written in different languages. We conclude with cautionary notes about the need for empirical validation.
Förderung
- Directorate-General XII, Science, Research, and Development
- Ministerio de Ciencia y Tecnología
- Ministerio de Ciencia, Innovación y Universidades
Zitieren
APA
Mens, G. L., & Gallego, A. (2025). Positioning Political Texts with Large Language Models by Asking and Averaging. Political Analysis, 33(3), 274-282. https://doi.org/10.1017/pan.2024.29
Vancouver
Mens GL, Gallego A. Positioning Political Texts with Large Language Models by Asking and Averaging. Political Analysis. 2025;33(3):274-282. doi:10.1017/pan.2024.29
BibTeX
@article{Mens2025,
author = {Mens, Gaël Le and Gallego, Aina},
title = {Positioning Political Texts with Large Language Models by Asking and Averaging},
journal = {Political Analysis},
year = {2025},
volume = {33},
number = {3},
pages = {274--282},
doi = {10.1017/pan.2024.29}
}
EUVIMED