An Introduction to our Methodology
We'll have a lot more to say as about how we approach the evaluation of elections as this project progresses. We will be using our collaboration with the Orange County Registrar of Voters to implement a number of methods that have been used in past elections, and in other locations, to evaluate the performance of an election process.
Many of the tools we will use in our 2018 study have been discussed in a book that project PI Michael Alvarez co-authored with Lonna Atkeson and Thad Hall, Evaluating Elections. There we focused on methodologies for evaluating elections mainly from the perspective of voters and pollworkers, and for the examination of early and election-day voting procedures. A number of the tools that we will use for the forensic analysis of election returns data were discussed in a book edited by Alvarez, Hall and Susan Hyde, Election Fraud. An especially promising method that we are hoping to deploy this year is the use of machine learning methods in forensic analyses of elections data, as discussed in a book chapter authored by Alvarez, Ines Levin, and Julia Pomares, "Using Machine Learning Algorithms to Detect Election Fraud." We'll also be examining various auditing procedures used by election officials, using approaches discussed in the book that Hall, Atkeson, and Alvarez edited, Confirming Elections.
As the project moves forward, we'll continue to post updates about the methodologies we are using and what we are finding when we implement them.