Machines Versus Humans: The Counting and Recounting of Pre-Scored Punchcard Ballots

Abstract

The counting of ballots, especially punchcard ballots, has received a great deal of attention in the years following the 2000 presidential election in Florida. Much of the research literature has focused on various measures of how accurately voting machines record voter intentions, with studies of the relative accuracy rates across voting machines (e.g., Caltech/MIT Voting Technology Project 2001), studies of voting accuracy across groups of the electorate (Alvarez and Sinclair 2003), and studies that examine the variability in voting machine accuracy across both machine types and voter types (Alvarez, Sinclair and Wilson 2002; Ansolabehere 2002; Tomz and Van Houweling 2003).