Building Trust When Adopting New Technology
by Richard Twigg

When choosing a new technology for use in an electoral system, an election management body, or EMB, must consider a range of issues. Important factors include cost, supporting infrastructure, ease of use (for voters and election officials), accuracy, and dependability. These concerns are significant, but just as critical to success (and frequently overlooked) is the issue of public trust: not only trust in the e-voting equipment itself but also trust in the EMB providing it. This article will discuss how to think about public trust and how voting officials can cultivate it.

Build public trust in the election authority
A recent study by researcher Wolter Pieters about the acceptance of voting technology focused on the difference between trust and confidence.1 Pieters, who studies voting and technology, suggests that trust is based on an assessment of risks (and a decision about whether or not to accept them). In contrast, confidence arises more from familiarity than from understanding. If something, say a computer, performs reliably over time, we may have confidence in it even though we might not fully understand the risks of losing our data.

Pieters concluded that it is more important for the public to trust their EMB than the voting technology itself. Confidence in technology can be developed fairly quickly if it performs reliably and accurately. However, it is just as easy to lose if the system crashes several times.

In contrast, trust is part of an intra-personal relationship that can be nurtured by an EMB or government. It requires more effort to establish, but once it is gained, it is not easily lost. If an EMB is open about the risks and trade-offs between use of familiar and new technologies,and it demonstrates that it has established reliable backup procedures should the new technology fail, the public is more likely to trust it. In such a case, the EMB might retain public trust even if the system fails (and the system would therefore lose the public's confidence). However, if EMBs are not transparent in discussing the tradeoffs and do not demonstrate they are ready for system failures, citizens will not trust them and may feel disenfranchised. The loss of trust is difficult to overcome.

Build public trust in a new system by automating a known process
Once a decision has been made to implement a new election technology, the EMB should take care to keep as many similarities between the former and the new systems as possible. Studies have shown that if a technology is presented as an automated version of an existing process, the technology will be better received by citizens.2 Events that took place in Hamburg, Germany, in 2005 offer an example of this phenomenon.

Germany
Hamburg's ballot book was 25 pages long, and voters could choose 10 of 1,000 candidates. Given the population of Hamburg,this meant that the local election authority was dealing with 1.6 million ballot sheets and more than 8 million votes, and it was required to tally all votes by the end of the day. To achieve this goal, it would have needed four times the number of election officials on staff, at a cost of 6.8 million Euros.

Instead, the EMB decided to pilot a new digital pen and software technology developed by Anoto. This digital pen works in conjunction with digital paper—that is, paper with microdots on its surface. In its tip, the pen has a scanner that enables it to know where it is on the paper (thanks to the microdots). Hamburg election officials simply printed the ballot book—in its normal format—on digital paper. Voters marked their ballots as they had always done, but this time they used the digital pen. Each vote was recorded in the pen, which was then docked in a computer, which recorded the vote and cleared the pen for the next voter.

The pilot was so successful that the digital pen will be used in Hamburg's 2008 elections. Hamburg's election officials saw clear advantages in using digital pens because of their complex ballot, but citizens accepted the technology because of the familiarity of the process. They used the same format of ballot book as they had in the past. The automation of existing voting procedures helped instill trust in the new technology.

The Netherlands
The Netherlands also relied on familiarity with existing systems to ease the introduction of electronic voting machines and, later, Internet voting. Essentially, they automated the process voters were accustomed to using. Some of the reasons for their success include:

• Voters went to their regular polling stations to vote (the Netherlands did not establish new voting locations as some countries have, creating confusion);


• The screens of the electronic voting machines used a layout similar to that of the old paper ballots; and


• The government has established relationships of trust with citizens.

Estonia
Recently, Estonia successfully implemented Internet voting by using a process familiar to its citizens. All Estonians have ID cards that fulfill various functions: it is their health care card, their library card, and their driver’s license. When the Estonian government offered voters the option of online voting (it was not mandatory), officials designated this same card as citizens’ access to the Internet voting system. In addition, Estonia’s electorate is very technically literate as a result of the country’s commitment to IT (Estonia spends more on IT, as compared to its GDP, than any other country).The EMB knew its audience and used a familiar process to facilitate trust in the new voting system.

Between 2005 and 2007, there was a 30 percent jump in use of Estonia’s Internet voting system. To address concerns about voter intimidation, election officials allowed voters to vote repeatedly, but only the last vote counted. A manual re-vote was also allowed in case the e-vote was revoked.
In each of these examples, trust in the EMB— established through familiarity with an existing system—helped ensure the successful introduction of new voting technology. EMBs must sort through a variety of issues when choosing an appropriate election technology. Among these concerns, they should also consider whether the electorate trusts the EMB enough to accept the new technology and whether they can introduce it in a way that will reinforce that trust.

Contact: Richard Twigg, Information
Technology Director at IFES
Tel: +1 202-350-6720

Notes 1 Wolter Pieters, "Acceptance of Voting Technology: Between Confidence and Trust" (Radboud University, Netherlands, 2006). Available at http://www.cs.ru.nl/~wolterp/iTrust.pdf. 2 Ibid. 3 Melanie Volkamer and Roland Vogt give an overview of the technology and Hamburg's intention to use it in advance of its adoption there. See 'New Generation of Voting Machines in Germany: The Hamburg Way to Verify Correctness' (International Association for Voting Systems Sciences, 2006). Available at http:// fee.iavoss.org/2006/papers/fee-2006-iavoss-New-Generation-of-Voting-Machines-in-Germany.pdf.