Statistical Advances in Post-Election Ballot Auditing

Abstract

Post-election audits are an integral part of the broader election audit process. Depending on the auditing method, they may detect miscounts in the official tabulation of votes, or limit the risk of certifying an incorrect outcome. In this chapter, we review the statistical methods for post-election auditing. We start by illustrating the limitations of fixed percentage statistical audits, which currently constitute the most common type of auditing method among states that require or allow post-election audits. Then, we described the statistics of more sophisticated auditing methods where the probability of drawing an auditing unit increases with the chance of error associated with that unit. After that, we discuss a new type of post-election auditing method that has received broad support risk-limiting audits, that have a minimum pre-specified chance of leading to a full recount when the outcome is incorrect. We also discuss additional statistical issues in post-election auditing, including the relevance of the auditing unit and alternative approaches to random sample generation. We conclude with a discussion of the limitations of statistical approaches to post-election auditing described earlier in the chapter, and difficulties that might be encountered while trying to apply them in real world elections.