Elite Polarization in European Parliamentary Speeches: a Novel Measurement Approach Using Large Language Models

TL;DR

This paper introduces a novel, scalable method using Large Language Models to measure elite polarization in European parliamentary speeches by analyzing directed sentiment, providing a new tool for cross-national political research.

Contribution

It develops the Elite Polarization Score, a new measurement capturing mutual out-party negativity in parliamentary debates, validated across three countries and multiple decades.

Findings

The measure is conceptually distinct from mass affective polarization and other phenomena.

Validation shows high accuracy and no false discoveries across countries.

The method is multilingual, scalable, and requires no task-specific training.

Abstract

Theories of democratic stability, populism, and party-system crisis often point to a form of polarization that comparative research rarely measures directly: hostile relations among political elites. Existing comparative measures capture adjacent phenomena, including mass affective polarization, or elite ideological distance, but not directed mutual elite evaluation. This paper introduces the Elite Polarization Score, a measurement of out-party evaluations in parliamentary speech. Large Language Models identify political actors mentioned in parliamentary debates, recover speaker-target pairs, estimate the sentiment directed at each actor, standardize heterogeneous references into party dyads, and aggregate these evaluations into party- and parliament-level measures of mutual out-party negativity. The validity of the approach is demonstrated on parliamentary corpora from the United Kingdom, Hungary, and Italy, covering up to four decades of debate. The resulting measure is conceptually distinct from mass affective polarization, elite ideological polarization, incivility, negative campaigning, and general sentiment. Evidence from the UK case study shows that it is also empirically distinct from mass affective polarization, elite ideological polarization, and incivility. Extreme negative evaluations can also be used to locate pernicious polarization rhetoric. Validation across three countries finds no false discoveries, sentiment estimates accurate to roughly 10 percent of the scale range, and AI sensitivity that meets or exceeds that of human coders in two of three settings. Because the algorithm is multilingual, requires no task-specific training, and can be aggregated by party and quarter, it provides a scalable basis for future cross-national research on what produces elite polarization and what elite polarization itself produces