We the People: An Empirical Analysis of the U.S. Political Climate

Overview

This project uses polling datasets to create micro-projections on how voters will behave during the 2026 United States Midterm Election and the 2028 United States General Election. These projections are translated into candidate and party win probabilities through the use of R, Python, and Excel. Data was imported and standardized across a variety of sources, aggregated with weighting to account for recency and sample size, then converted into win probabilities with respect to the margin of error. Elections with party candidate nominations will sum to 100 percent within their own party. In the Senate set, North Carolina is projected to lean Democratic at 74.1 percent for Roy Cooper, while Ohio and Michigan lean Republican. Maine's election is most dependent upon specific Democratic nominees, with Susan Collins favored against Janet Mills. The model accurately predicted Democratic nominee Henry Cuellar, who won the Texas Congressional District 28. Gubernatorial results range from California and Minnesota Democratic strongholds to a near toss-up in Georgia at 51 percent Republican. For 2028, the model currently identifies J.D. Vance as the leading Republican nominee at 50.5 percent and Gavin Newsom as the leading Democratic nominee at 32.5 percent, with a slight Democratic lead in the early election party outlook.

Competition history

• ISEF 2026 Technology Enhances the Arts · Entry TECA007

Resources

• ISEF abstract page ↗

Related projects

Closest projects by meaning, across every fair and year in the corpus.