Report of the Chief Electoral Officer of Canada – Following the Pilot Project on the Use of an Assistive Voting Device in the November 29, 2010, By-election Held in Winnipeg North
5. Feedback, Costs and Findings
Elections Canada focused primarily on feedback from voters, election workers and representatives of electors with disabilities to conduct its evaluation. It also conducted a survey of electors following the by-election to validate the level of awareness regarding the availability of the AVD.
As indicated earlier, this feedback was used to draw conclusions based on the criteria of awareness, user experience, extent of use, integrity, operational effectiveness and usefulness (see Table 1).
Feedback from assistive voting device users
Elections Canada gathered information from voters as they were using the AVD. Of the five electors who used the device, four were satisfied with their experience. Two of the five preferred to have the assistance of a family member, who validated their vote. All five required assistance while using the device. Most encountered difficulty following the system's instructions and saw a need for improvements in system functionalities.
One elector expressed reluctance to use the device in the future, finding it too slow. It took approximately 10 minutes for each elector to vote using the device. Three potential users decided against the AVD because they were in a rush.
Comments from election workers
Many electors who came to vote at the polls were offended at being asked whether they required assistance. Election officers had to explain why they were asking the question. Some deputy returning officers stopped asking it.
The AVD presented a number of challenges to election workers. An issue at several polling sites and the returning office was space configuration and the set-up of the device. The device screen had to be in a position that ensured secrecy. At the same time, there had to be an electrical outlet nearby as well as sufficient room for electors using wheelchairs.
There was a very narrow window of time between the issuing of the list of confirmed candidates, 19 days before election day, and the configuration and deployment of the equipment in the returning office, which the agency had set for 15 days before election day. Within this time many tasks had to be completed, including the production of sound files of candidate and party names, testing and sealing of the device, shipping of the device from Toronto to Winnipeg and training of AVD operators.
While there was sufficient time to program the audio file with the list of candidates before advance voting began on the 10th day before election day, the delivery of the devices to these sites the day they opened, their installation and their testing within a short time frame proved to be a challenge. The devices also had to be removed promptly at the close of polls.
Feedback from stakeholders
The agency held a post-election meeting with organizations representing electors with disabilities. For the meeting, it invited groups that had taken part in the pre-election town hall forum as well as other national organizations. Overall, the groups felt that this was not the right device, but that the pilot project was a step in the right direction by Elections Canada toward improving the accessibility of the electoral process and that it should explore other options.
In its surveys following the November 29 by-elections, Elections Canada asked Winnipeg North electors whether they were aware of the AVD pilot project. Of the 451 respondents, 26 percent remembered hearing about the availability of an AVD for electors with disabilities. Of these, 55 percent had heard about it through television and newspapers; 14 percent read about it in a brochure, postcard or pamphlet; 11 percent heard about it when they voted; and 9 percent heard about it through the radio.
The following table provides costs related to the implementation of the pilot project. The costs include deployment of the equipment and provision of a number of services by Dominion Voting Systems. It is important that the costs of a pilot project in a single electoral district not be extrapolated to a general election conducted in 308 electoral districts. In a general election, these technical support services would have to be delivered according to a different, cost-effective business model. Costs are naturally a concern but are not the determining factor in the provision of an assistive voting service, since the rights of electors are at stake.
|Item|| Cost ($)
|1. Deployment of AVDs|
|2. Elections Canada in Ottawa – Staff and travel expenses||37,460|
|4. Supplies, shipping and printing||1,273|
The experience of organizing and conducting a pilot project during an electoral event has confirmed the value of testing a service model for electors before making recommendations for legislative changes. A pilot project improves Elections Canada's ability to provide operational feedback on its effectiveness and engages organizations representing affected electors in the process. Elections Canada will certainly make use of this model in future to test other electronic voting methods, with the approval of parliamentarians.
The agency has concluded that the type of AVD used in the pilot project in Winnipeg North is not a solution that lends itself to electoral events held at the federal level.
Elections Canada will therefore not proceed further with analysis of this device. Other solutions may better serve electors.
Given these findings, Elections Canada has decided not to proceed with a business case to determine the costs of using this device in a general election and the implications of using this technology over the next five years, as requested by the Senate committee. Elections Canada agrees that these are fundamental questions that it would pursue rigorously, were the agency to propose extending the use of assistive voting technologies to federal general elections.