Text-Based Ideal Points for the U.S. Congress

TBIP provides researchers with a novel measure of ideology based on texts.

About

TBIP are a novel measure of ideology via text-based ideal points that infer ideological positions based on text alone. The model extends prior work by Vafa et al. (2020) and earlier iterations of Gaynor et al. (2026), which utilizes Latent Dirichlet Allocation (LDA) via Gibbs sampling to generate outputs based on topic-word distribution (β) and document topic distributions (θ). The resulting ideal points capture not only the topics legislators engage with, but also the ideological framing and word choices used when discussing those topics. As a result, ideal points are determined jointly by topic engagement and topic-specific word polarity, allowing the model to estimate stable ideological positions without relying on party labels or other external ideological indicators.

Data

Datasets available for download. Browse datasets.

Papers

Working papers and publications. Browse papers.

Visuals

Interactive visualizations. Browse visuals.

Research team

Principal investigators

SoRelle W. Gaynor, University of Virginia
Pranav Goel, Northeastern University

Research assistants

Samuel Du, University of Virginia (graduate)
Lakshay Kansal, University of Virginia
Musayab Razaq, University of Virginia