and voters have marked their ballots in ways that prevent the vote-counting equipment from reading and counting the vote. National Survey Results We estimate that 32 percent of jurisdictions nationwide had no written instructions, either from the state or local jurisdiction, to interpret voter intent, such as marks on paper ballots or partially punched chads on punch card ballots. GAO Mail Survey of Jurisdictions What constitutes a proper mark on a ballot can differ depending on the type of voting method used. State guidance on what is a proper mark on any specific type of ballot, such as optical scan, varies, and guidance on how to interpret variations from proper ballot marks also varies. For example, state guidance to local election officials varied from general to specific regarding how to determine voter intent when a ballot could not be read by the vote-counting equipment. In some cases, poll workers or other election officials make that determination at the voting precinct. Forty-seven of the 50 states and the District of Columbia have laws with provisions for a recount, and they vary among the states. According to responses to GAO's mail survey, election officials in 42 jurisdictions in 16 states identified a total of 55 recounts (some jurisdictions identified more than 1 recount) for state or federal office from 1996 through 2000.13 All but one recount involved recounting every precinct in the jurisdiction. According to the jurisdictions, twenty-seven of these 55 recounts were required by state law, and 16 were conducted at a candidate's request. The remainder were for a variety of reasons, such as court order. Regardless of the reasons for the recount, whether it occurred before or after the certification of the vote count, who conducted the recount, or the methods used for the recount, the jurisdictions reported that none of the recounts altered the original outcome of the election. 13 GAO included only responses from the mail survey that were verified by phone with the jurisdictions. Challenges The following are some key challenges that election officials identified for vote counting: • counting absentee, provisional, and early voting ballots to include eligible voters and maintain the integrity of the vote counting process; ⚫ interpreting variations when ballots are not properly marked, a task that is particularly important when votes are close; and ⚫ completing the results of a recount in a close or contested election in a fair, accurate, and timely manner. Voting Technology Four of the five methods by which votes are cast and counted in the United States involve technology-lever machine, punch card, optical scan, and DRE. The fifth-paper ballot-does not. The four methods that involve technology were used in 98 percent of all precincts nationwide. GAO examined the technologies used in these voting methods according to a range of characteristics, including accuracy, ease of use, efficiency, security, testing, maintenance, and cost. With respect to accuracy, ease of use, efficiency, and security, GAO's analysis of vendor-provided data showed little difference among DRE, optical scan, and punch card equipment. DRE rated slightly better than optical scan, which in turn rated slightly better than punch card. GAO's analysis of jurisdiction-reported data on the various types of technologies revealed more distinguishing differences, although still not strikingly different, with DRE rating better than the other voting methods. Figure 2 compares vendor- and jurisdictionreported data on the various types of technologies. The differences among voting equipment reported by local election jurisdictions can be attributed, in part, to the differences in the equipment itself. However, they also can be attributed to the people who use the equipment and the rules or processes that govern its use. In each case, different opportunities exist for voter misunderstanding, confusion, and error, which in turn can affect the equipment's performance in terms of accuracy, ease of use, and efficiency. Despite these differences, the vast majority of jurisdictions across the country were satisfied with their respective methods of voting in the November 2000 election. From its national mail survey, GAO estimates that 96 percent of jurisdictions nationwide were satisfied with the performance of their voting equipment. More specifically, about 99 percent of DRE jurisdictions, 95 percent of optical scan jurisdictions, and 97 percent of the remaining jurisdictions (those that used lever, punch card or hand-counted paper ballots) were satisfied or very satisfied with their voting method (hand-counted paper ballots does not use voting equipment). National Survey Results We estimate that 96 percent of jurisdictions nationwide were satisfied with the performance of their voting equipment during the November 2000 election. GAO Mail Survey of Jurisdictions However, because many jurisdictions indicated they did not collect data on the accuracy of their voting equipment, this sense of satisfaction may be largely based on officials' perceptions of how their respective equipment performed. Further, most of the 27 local election jurisdictions GAO visited did not collect actual performance data for the voting equipment used in the November 2000 election. National Survey Results We estimate that about 48 percent of jurisdictions nationwide collected data on the accuracy of their voting equipment for the November 2000 election. GAO Mail Survey of Jurisdictions Voting equipment's performance is not the only equipment characteristic germane to effective election administration. All voting equipment is influenced by testing, maintenance, and cost issues, each of which also involves people and processes. Properly testing and maintaining voting equipment are required if its optimum performance is to be achieved. Also, the overriding practical consideration of the equipment's life-cycle cost versus its benefits, which affects and is affected by all the characteristics, must be considered. Newer voting equipment and methods beyond the voting equipment used in The capital cost for replacing existing voting equipment with optical scan GAO's vendor survey showed that although some vendors include certain software costs in the unit cost of the voting equipment, most price other software separately. Using software cost information available in August 2001, GAO estimates that these other software costs for DRE, optical scan, and punch card equipment can run as high as $300,000 per jurisdiction. The higher costs are generally for the more sophisticated software associated with election management systems. Because the software generally supports numerous equipment units, the total software cost per unit varies depending on the number of units purchased or the size of the jurisdiction. The cost estimate for each approach used a set of assumptions that may overestimate the needs and costs for some jurisdictions and underestimate the needs and costs for other jurisdictions. These assumptions and limitations are discussed in more detail in the text that accompanies each estimate. Challenges The following are some key voting technology challenges: |