Datasets for building stratification frame for US 2026 Midterm election

This repository contains the raw datasets used to create the Stratification Frame used to predict US midterm 2026 voting preferences using synthetic sampling (research soon to be submitted for publication). The datasets are archived here to facilitate reproducibility. Unless otherwise noted, the datasets were obtained from publicly available third-party sources. Ownership and copyright remain with the original data providers. Please cite the original dataset providers when using these data. Please also refer to the terms and conditions of use on the respective source websites, if using for any purpose other than reproducibility. Depending on the data source, there are explicit prohibitions against combining these with other data sources to identify individuals, prohibitions against commercial use, and prohibitions against use for nefarious, partisan or gerrymandering purposes. Please refer to the source websites for details. Further information including sources and attributions of each dataset is provided in the accompanying data_sources.md file.