Inferring Roll Call Scores from Campaign Contributions Using Supervised Machine Learning

Abstract

This paper develops a generalized supervised learning methodology for inferring roll call scores for incumbent and non-incumbent candidates from campaign contribution data. Rather than use unsupervised methods to recover the latent dimension that best explains patterns in giving, donation patterns are instead mapped onto a target measure of legislative voting behavior. Supervised learning methods applied to contribution data are shown to significantly outperform alternative measures of ideology in predicting legislative voting behavior. Fundraising prior to entering office provides a highly informative signal about future voting behavior. Impressively, contribution-based forecasts based on fundraising as a non-incumbent predict future voting behavior with the same accuracy as that achieved by in-sample forecasts based on votes casts during a legislator's first two years in Congress. The combined results demonstrate campaign contributions to be powerful predictors of roll-call measures of ideology and resolve an ongoing debate as to whether contribution records can be used to make accurate within-party comparisons.

Keywords: ideal point estimation, supervised machine learning, Congress, campaign contributions