Each election should have its own Readme in which we document the source of each dataset, a data dictionary, and whether we've done any processing.
Each election should have its own Readme in which we document the source of each dataset, a data dictionary, and whether we've done any processing.