Download the Full Report 

This account of the November 7, 2006 election draws on surveys from participants in Pollworkers for Democracy, reports from voters who called the Election Incident Reporting System and Voter Action, and reports collected by VotersUnite.Org from the national and local media. In all, we looked at 1022 reports of problems associated with electronic voting equipment from 314 counties in 36 states.

Many reports depicted multiple problems experienced by a single voter or pollworker; thus the total number of problems discussed in this paper is greater than the number of reports reviewed. While some reports reference a single incident, others reference widespread incidents (such as the 800 e-voting machines that malfunctioned in Westmoreland County, Pennsylvania). The first section of our report shows the geographical extent of problems. The second, a set of case studies, explores the problems in four locales in greater depth.

The mid-term election revealed that the promise of easier voting, more accurate tallies, and faster results with electronic systems has not been fulfilled. Voters in some jurisdictions waited in line for hours to cast their ballots. Others cast their ballots accidentally before they were done because they pressed the wrong button or left without casting their ballots because they didn’t press the right button. Many voters watched the machine highlight a candidate they didn’t select or fail to indicate a vote for a candidate they did select and were then blamed for not being able to use a computer correctly.

Many polling places couldn’t open on time because of machine failures, and complex procedures often left pollworkers frustrated and reluctant to serve again. Election directors were often forced to rely on voting equipment vendors to set up the election, administer it, and tally the votes because it was too complicated for their personnel to handle. Others blamed themselves for not following the poorly documented, non-intuitive procedures required to collect and tally the votes.

After the polls closed, poll workers and election officials struggled with a myriad of reporting problems. Many couldn’t retrieve data from memory cards or couldn’t get the tally software to combine totals from different computerized systems, while others couldn’t figure out why the software was subtracting votes instead of adding them, or adding them two and three times instead of only once; couldn’t determine for sure whether the first set of results was correct, or the second set, or the third; couldn’t explain why one out of every six voters didn’t have an electronic vote recorded for a hotly contested race; or why the machines recorded more ballots than the number of voters who signed in to vote.

Often hidden from public view, equipment malfunctions such as these have normally been exposed only when they are severe enough to attract media coverage. Reports from Pollworkers for Democracy and voters provided additional insight into the extent of these problems. The frequency of reports of difficulties retrieving results even casts doubt on the accuracy of the certified results in affected areas, particularly since it is reasonable to  assume that many such retrieval problems were never reported.

An increasing number of voters, poll workers, and election officials are finding the election process to be more difficult, not easier, and confidence in the final tallies has been undermined. While our source material is neither a complete list of problems nor even a representative sampling, the number of incidents and the broad range of problems reported is indicative of the widespread failure of electronic voting systems across the country and how this failure affected the experience of voters on November 7, 2006.