Party Matters: Enhancing Legislative Embeddings with Author Attributes for Vote Prediction

TL;DR

This paper improves legislative vote prediction by incorporating sponsor attributes into neural models, demonstrating that metadata enhances generalization across different congressional sessions.

Contribution

It introduces a method that combines bill text with sponsor ideologies in neural networks, significantly improving prediction accuracy over prior models.

Findings

Adding sponsor attributes boosts accuracy by 4%

Metadata is crucial for cross-session generalization

Neural models outperform previous state-of-the-art

Abstract

Predicting how Congressional legislators will vote is important for understanding their past and future behavior. However, previous work on roll-call prediction has been limited to single session settings, thus did not consider generalization across sessions. In this paper, we show that metadata is crucial for modeling voting outcomes in new contexts, as changes between sessions lead to changes in the underlying data generation process. We show how augmenting bill text with the sponsors' ideologies in a neural network model can achieve an average of a 4% boost in accuracy over the previous state-of-the-art.