Electronic vote counting equipment makes it easy for a few persons to electronically manipulate vote counts and virtually impossible to independently audit vote count accuracy. Without routine independent audits of hand-countable voter verified paper ballots, insiders have freedom to manipulate vote counts with negligible possibility of detection.
This new exit poll discrepancy function allows us, for the first time, to know what patterns of exit poll discrepancy result from various combinations of
- vote miscounts,
- partisan exit poll completion rate differences, and
- random sampling error (by simulation).
These patterns can be compared with actual discrepancy patterns produced in elections to determine what causes are consistent with actual election discrepancy.
This paper also outlines statistical methods to determine the number of precincts with significant discrepancy, and how to test precinct-level exit poll and official election results data for consistency with partisan exit poll response rate explanations. Using these methods, mathematicians or statisticians can determine if vote miscounts are indicated prior to when candidates concede future elections.