Consultation and Evaluation Practices in the Implementation of Internet Voting in Canada and Europe
Summary of Evaluation Practices
Post-election evaluation in the European cases is much simpler to pinpoint than it is in the Canadian municipalities because of the existence of international agencies that take on the task of election observation and the writing of official reports. These approaches are compared with the Canadian cases in Table 3. In Europe, Internet voting operations are evaluated as part of an overall consideration of the electoral practices in a country. Primary among evaluating agencies is the Organization for Security and Co-operation in Europe Office for Democratic Institutions and Human Rights (OSCE-ODIHR), which makes a particular effort to send observer missions to states conducting Internet elections. In all of the European cases as well, these missions were actively invited and facilitated by the states concerned, in order to secure a favourable independent judgment on their election and to benefit from any recommendations made for improvement. Criteria used by OSCE range across security and secrecy issues, from those involving access to the ballot to the transparency of the counting process. OSCE does not conduct research for its reports other than that which can be accumulated from other sources or observed during its missions.
More expanded sets of evaluation criteria may be used in jurisdictions where additional evaluations are commissioned. In Norway, the Institute for Social Research has examined the potential effects of Internet voting on turnout, even though this was not an overt goal of the implementation of the trials. In Switzerland, the project being undertaken by Professor Pascal Sciarini for the canton of Geneva will also look at voting turnout, and will consider the use of the Internet to vote by particular subgroups of the population. In both Norway and Switzerland, the intention of the additional evaluations is to allow academic research to be conducted about the use of the Internet to vote, and the public attitudes toward it. Research tools involve not only public opinion surveys but also focus groups and specific consultations with subgroups of the population.
In Canada, evaluation practices seem to be developing over time. As more jurisdictions sign on to Internet voting, they benefit from the work done by their predecessors, learning about best practices, useful model features, and those items that did not work as well as hoped. As time passes, evaluation seems to be becoming increasingly more rigorous. The City of Markham, with its comprehensive surveys and data analysis, and the City of Edmonton, with its detailed evaluation of the Internet voting consultations through surveys, seem to be the most advanced in this regard. In some of the Nova Scotia municipalities (Cape Breton Regional Municipality and the Town of Truro), there is a broad discussion that takes place among groups of different people. This is not carried out in the form of an official survey (aside from election workers), but rather takes the shape of informal discussions.
For many Canadian municipalities, systematic, documented and research-based post-election evaluation does not take place. It may be that municipal election officials put a lot of effort into a pre-election consideration of whether Internet voting is a desirable policy, and when events seem to transpire as planned during the voting, and satisfaction appears to be high, they do not see a pressing need for a more fulsome, and potentially costly, evaluation process afterward. For many municipalities, regular election evaluation may be composed of a short report to council and election post-mortem meetings, but beyond that, notes stay on a computer or in a filing cabinet until the next election.
To be sure, in those municipalities that have used Internet voting in multiple elections, modifications are made between elections, which indicates that evaluation is taking place. Markham, for example, added voters' date of birth as an additional security credential for registration in 2010 as a direct consequence of reflection and evaluation. However, it seems that once the operation of Internet voting passes the initial test without succumbing to potential pitfalls, it becomes part of the current operational procedure. Realization that more rigorous evaluation would be desirable seems to be taking hold in many municipalities, however, and may lead to a retooling of evaluation procedures and processes. The decision to move forward, however, is not always a foregone conclusion. Recent decisions by the cities of Kitchener, Ontario, and Edmonton not to proceed with Internet voting in forthcoming elections, but rather re-evaluate at a later date, is evidence of this.
It needs to be pointed out that the current Canadian experiences with Internet voting have all been at the municipal level. Trials at the provincial or federal level would be subject to greater public scrutiny than municipalities, if only because they would have more extensive geographical reach or implications. Consideration of Internet voting by the British Columbia panel certainly involves the evaluation situation. Internet voting is in its infancy, but a consensus is developing that its evaluation procedures must be detailed and rigorous. The attention it attracts and the changes it imposes to the traditional election process create pressure to develop thorough, transparent, and documented evaluation procedures and policies for Internet voting.
Jurisdiction | Internet elections held (number, type) | Type of evaluation used | Agents involved or consulted: Internal agents | Agents involved or consulted: External agents | Evaluation criteria used |
---|---|---|---|---|---|
Estonia | 2 Local 2 National 1 Extra-national |
Public opinion surveys Reports and papers (i.e. academic analyses, Organization for Security and Co-operation in Europe Office for Democratic Institutions and Human Rights [OSCE-ODIHR], National Electoral Committee) |
E-voting Committee National Electoral Committee |
Academics IT experts OSCE-ODIHR Expert working group |
Legality Security Transparency Observability Cost |
Switzerland | Numerous cantonal referendums | Reports (i.e. Federal Chancellery, OSCE-ODIHR) Ongoing analysis by Professor Pascal Sciarini |
Federal Chancellery | Academics IT experts Commission for the Evaluation of Public Policy OSCE-ODIHR |
Turnout Sub-groups Public attitudes Determinants of online voting Legality Security Transparency Observability Cost |
Norway | 1 Local | Reports (i.e. ISF, OSCE-ODIHR) Focus groups Interviews Public opinion survey |
Ministry of Local Government and Regional Development Election administrators |
IT experts International Foundation for Electoral Systems Institute for Social Research OSCE-ODIHR |
Legality Security Transparency Observability Cost |
City of Edmonton | None (mock Jellybean Election) | Reports (i.e. risk assessment report, security report, Citizens' Jury on Internet Voting report, public opinion survey response report, report from the roundtable advisory meetings) Roundtable meeting with city project team |
City project team Centre for Public Involvement team Municipal administration Centre for Public Involvement staff |
Auditor (Seccuris) | Usability Functionality Security Auditability Voter privacy |
City of Markham | 3 Local | Online voter surveys and 2010 candidate survey (Delvinia) Markham candidate surveys Reports (i.e. Delvinia reports, city evaluation report) Audits and documentation (i.e. audit report, program code evaluation, internal security audit) Internal lessons learned sessions |
Municipal administration Election officials |
Ryerson University Security company Delvinia Online voters Candidates |
|
Halifax Regional Municipality | 2 Local (2 regular elections, 1 by-election) | Audits and documentation (i.e. independent audit, auditor's report, independent security audit) Reports (i.e. report on recount in district 3*, report for council and Service Nova Scotia and Municipal Relations, report from Internet voting vendor) Internal meetings |
Municipal administration Electoral board HRM election staff Internal groups (i.e. IT, communications, marketing, accessibility committee) |
Independent auditor Security company Internet voting vendor |
Accessibility Transparency Cost Security Maintaining the integrity of the vote |
Cape Breton Regional Municipality | 1 Local | Meetings with key staff Poll worker survey Discussions and interviews with select persons and groups Internal election review folder Audits and documentation (i.e. independent audit, auditor's report) Reports (i.e. report of the Returning Officer, report from Internet voting vendor) |
Municipal staff Poll workers E-vote call centre staff |
Auditor (Chief Information Officer, Cape Breton University) Internet voting vendor Senior citizens' clubs The media Candidates Other Nova Scotia municipalities |
Safety Security Reliability Accessibility |
Town of Truro | 1 Local | Meetings with key staff Discussions and interviews with select persons and groups Internal election review folder Collection of media clippings Returning Officer's Report Report from Internet voting vendor Independent audit Auditor's report |
Election staff Auditor Chief Administrative Officer |
Internet voting vendor News media Candidates Nursing home workers and those who worked other public access points (i.e. the library) |
Voter turnout Accessibility |
*This is not a regular part of the evaluation protocol of elections, but rather was requested by a candidate after a close race separated the top two candidates by less than one percent of the vote.