Controversy over voting technologies in U.S. politics came about in the 2000 presidential election. The inability to obtain a reliable vote count in Florida in the close presidential contest between Al Gore and George W. Bush triggered electoral reform, with a particular focus on election voting machinery. The 2000 presidential election illustrated to all the vulnerability of U.S. voting systems and how voting machines could play havoc with voter intent and electoral results. States today use five different voting technologies, and some jurisdictions employ more than one type. Because of heightened awareness and increased media attention to voting problems, choosing among these various voting technologies has become problematic and controversial.
I. A Range of Choices
II. Mechanical Lever Systems
III. Punch-Card Systems
IV. Optical Scan Ballots
V. Electronic Systems
I. A Range of Choices
State governments, being primarily responsible for election administration, have used a variety of balloting procedures and voting hardware, including oral voting, paper ballots, mechanical lever machines, punch-card systems, mark-sense or optical scanning ballots, and electronic systems. One of the least complex systems—the hand-counted ballot— was the first voting mechanism used. Hand-counted ballots allow voters to mark boxes next to the preferred candidates’ names. Even more simplistic, in the late 1700s and early 1800s, southern male voters would go before a judge and, when their names were called, would publicly state which candidates they supported. Because of voter coercion and intimidation, the secret paper ballot was introduced and became the primary voting device (Alvarez and Hall 2008, 15). Given the time-intensive nature of hand-counting paper ballots, the use of paper ballots has greatly declined. Paper ballots are used primarily in less populous counties; nationwide, the number of voters using them has been reduced from 2 million in the year 2000 to 300,000 (or 0.2 percent) in 2006 (Percy 2009, 2).
II. Mechanical Lever Systems
The mechanical lever machine, invented in 1892, was adopted by states to obtain quicker and more reliable election results. In using this machine, voters indicate their voting decisions by pulling a lever near a candidate’s name. The levers, connected to counting wheels, maintain a running tally of the votes cast for each candidate. Election officials obtain vote totals by reading the number of votes recorded by the counting device. Devised to alleviate problems associated with paper ballots, mechanical lever machines were also vulnerable to fraud and corruption. For example, election administrators with access to the storage devices could manipulate stored vote totals at any time during the election cycle. Mechanical lever machines make it difficult for the disabled to vote because, being large machines (comparable in size to a refrigerator), they can be hard to access by those with physical handicaps or those who are visually impaired. Also, lever devices are not easily adapted to provide ballots in multiple languages (Alvarez and Hall 2008, 16). By 1998, only about 15 percent of counties still used mechanical lever machines, affecting about 18 percent of the voting population. In 2006, mechanical voting systems were used in just 62 counties and covered less than 7 percent of the nation’s voting-age population (Knack and Kropf 2002).
III. Punch-Card Systems
In the 1960s, punch-card voting machines emerged. They were cheaper and allowed election officials to purchase more machines to accommodate larger precincts. Punch-card systems employ two types of ballots. The simpler ballot design—Datavote—allows voters to punch a hole on a card beside the candidate’s name using a punching tool with a metal shaft. The second type—Votomatic—provides the candidate’s name in a booklet attached to a mechanical holder under which voters insert a prescored punch card. Each hole on the punch card has a corresponding number. Voters punch a hole corresponding to the number of the candidate they wish to support. With a stylus or other device, voters punch holes at the appropriate spots on the card, forcing out the marked areas (also known as “chads”). In 1998 about 20 percent of counties, comprising 35 percent of the population, used punch-card technology. Following the 2000 election, in which this technology became an issue, the use of punch cards rapidly declined, and by 2006 only 13 counties used punch-card systems, affecting less than 0.5 percent of voters (Saltman 2006, 7–13).
In Florida, during the 2000 presidential election, use of punch-card voting systems led to undervoting (failure to vote in a race that the voter was eligible to vote in) and overvoting (or voting more times than permitted). Votomatic systems in particular were subject to alignment failures and malicious tampering with instructions. Also, voters with visual impairments found it hard to see the ballot with the punch card, and those with physical impairments found it difficult to use the stylus to mark their ballots (Alvarez and Hall 2008, 18). In 2000, punch-card voting systems, compared to optical scan, paper ballot, and machine lever ballots, had the highest rejection rate in Florida. In a study of rejected ballots in that state, approximately 4 percent of punch-card ballots were rejected. In such a situation, when different voting systems are used, voters in one jurisdiction do not have the same chance that their votes will be counted as do voters who live in another jurisdiction—effectively denying some voters the right to vote (Percy 2009, 7). An additional limitation to punch-card voting is that this type of equipment may influence the racial gap in proportion to voided ballots. African Americans cast invalid ballots more frequently than whites. Voting records from Louisiana and South Carolina, the only states to report voter turnout by race, indicate that punch-card voting systems produce a wider gap in frequency of voided ballots between blacks and whites than do mechanical lever and electronic machines. These findings have given jurisdictions further incentive to upgrade their voting systems (Tomz and Houweiling 2003, 58–60; Buchler et al. 2004, 523).
IV. Optical Scan Ballots
Marksense, or optical scan, ballots were introduced in the 1980s for computing election results, but the technology had been around for decades, as these machines were frequently used to grade standardized examinations. Similar to paper ballots, marksense ballots require voters to use a black marker to fill in a circle or box beside the names of preferred candidates. A scanning machine reads the dark marks on ballots and records the results. Although similar to a paper ballot, a marksense ballot is larger, allowing jurisdictions to provide information about candidates directly on the ballot rather than in a separate booklet. Two types of optical scanners in use are precinctcount optical scanning and central-count marksense equipment. Precinct-count optical scanning equipment allows voters to feed ballots directly into a reader, which can be programmed to return the ballot to the voter if the voter has mistakenly selected more than one candidate for the same office (Saltman 2006, 13). With central-count marksense equipment, on the other hand, voters drop ballots into a box, and election officials then feed the ballots into the counting machine. With this system, voters do not have an opportunity to correct faulty ballots. Increasing in popularity in the 1990s, the technology reached peak use in 2006, with 56 percent of all counties and almost 50 percent of the nation’s voting-age population embracing optical scan equipment (Utter and Strickland 2008, 46).
Like punch-card technology, optical scan ballots require voters to mark their choices on paper, and the paper is then scanned by an electronic device. Likewise, optical scanning voting systems face similar problems to those of punch-card voting. Voters are prone to undervote or overvote and may make mistakes with write-in candidates or find some other way to deface or spoil the ballot. Precinct-count machines can reduce these errors. Still, disabled voters may find it difficult to use optical scan technology if they have a handicap that makes using a pen or marking device unworkable. Voting with assistance raises concerns about coercion and fraud, because those who provide the assistance could potentially record preferences that do not match those of the actual voter. Because optical scan voting systems require that information specific to each precinct be printed on ballots to reflect all the possible choices voters would face and that the paper used be of high quality, it is hard to implement these systems in large, complex election jurisdictions. The increased complexity of printing these types of ballots across a large jurisdiction—which may require multiple formats and possibly multiple languages—raises the costs of using this technology. Because the technology is paper-based, some argue that this establishes a voter-verified paper trail and a means for recounting ballots for accuracy. However, the 2000 election demonstrated that having a marked ballot and interpreting voter intent are two different things. Voters may mark their ballots in ways that make their intent indiscernible, leading to delay in obtaining election results or no change in the recount of the vote (Alvarez and Hall 2008, 22).
V. Electronic Systems
Based on the microprocessor technology that emerged in the 1970s, electronic voting machines or direct recording electronic (DRE) devices allow voters to use touch-screen technology to vote. A ballot, displayed on an electronic device, displays vote choices. Then voters push buttons or touch spots on the surface of a computer screen to indicate their vote choices. Voters may write in a candidate, using the keyboard to type the name. Some DRE machines count votes as soon as they are cast, while others record ballot images that compute vote totals when the polling stations close (Jones 2001). Voter preferences are stored electronically, and if the machine is programmed correctly, voters have no chance to overvote. DRE systems can be programmed to display ballots in any language to accommodate minority voters for whom English is a second language. These systems can be designed to aid the disabled, granting improved access. Compared to punch-card and lever machines, DRE systems can reduce voter rolloff —the failure to indicate a vote choice for all offices and measures on a ballot—by up to 26 percent in judicial elections (Fischer 2001). In 2006, about 38.4 percent of voters (or 65.9 million voters) used DRE voting systems (Percy 2009, 8).
This type of technology has also generated its share of controversy. After the 2000 elections, manufacturers of DRE machines lobbied states to purchase their technology as a panacea for all their voting system problems. In conjunction with this, Congress passed the Help America Vote Act (HAVA), which gave states incentives to replace their punchcard and lever machines. In their rush to get the HAVA monies, many state governments did not engage in comprehensive evaluations of the DRE machines. If they had done so, they would have found various security and reliability problems. Critics argue that under DRE voting technology, only a very small number of individuals can inspect and verify the machines’ correct operation. Thus the integrity of a jurisdiction’s voting system rests with very few people. Some believe that the machines are subject to manipulation and error and that even the machines that print out a complete record of the votes cast can be trusted only to the extent that the software is transcribed in a valid fashion (Percy 2009, 8–9). The Caltech / MIT Voting Technology Project expresses numerous concerns including the following: (1) no longer being able to “see” the vote count (no auditable trail), (2) the inability of numerous people to examine the various stages of the voting process, (3) the vesting of authority in electronic machine manufacturers with little federal oversight on the technology, (4) companies that produce the machines can program them to record and count votes according to their partisan political interests, and (5) failures in the DRE machines that have led to lost votes and altered election outcomes (Alvarez and Hall 2008, 35).
As of 2009, about 89 percent of localities across the United States used either optical scan paper ballots or an electronic system. Less than 7 percent of election jurisdictions used lever machines and paper ballots. With almost $1.26 billion spent through HAVA to replace other voting technologies such as punch-card, mechanical lever, and paper ballots (Wolf 2008), clearly a possibly irreversible shift has occurred in the use of voting technologies in the United States.
Ruth Ann Strickland
- Alvarez, R. Michael, and Thad E. Hall, Electronic Elections: The Perils and Promises of Digital Democracy. Princeton, NJ: Princeton University Press, 2008.
- Buchler, Justin, Matthew Jarvis, and John E. McNutty, “Punch Card Technology and the Racial Gap in Residual Votes.” Perspectives on Politics 2 (September 2004): 517–524.
- Fischer, Eric A., “RL 30733: Voting Technology in the United States: Overview and Issues for Congress.” Congressional Research Service Report for Congress. March 21, 2001. http://usa.usembassy.de/etexts/gov/voting.pdf
- Jones, Douglas W., “Evaluating Voting Technology.” Testimony before the United States Civil Rights Commission, Tallahassee, FL, January 11, 2001. http://homepage.cs.uiowa.edu/~jones/voting/uscrc.html
- Knack, Stephen, and Martha Kropf, “Who Uses Inferior Voting Technology?” PS: Political Science and Politics 35 (September 2002): 541–548.
- Percy, Herma, Will Your Vote Count? Fixing America’s Broken Electoral System. Westport, CT: Praeger, 2009.
- Saltman, Roy G., The History and Politics of Voting Technology: In Quest of Integrity and Public Confi dence. New York: Palgrave Macmillan, 2006.
- Tomz, Michael, and Robert P. Houweiling, “How Does Voting Equipment Affect the Racial Gap in Voided Ballots?” American Journal of Political Science 47 ( January 2003): 46–60.
- Utter, Glenn H., and Ruth Ann Strickland, Campaign and Election Reform, 2d ed. Santa Barbara, CA: ABC-CLIO, 2008.
- Wolf, Richard, “Voting Equipment Changes Could Get Messy on November 4.” USA Today (October 29, 2008).