2000 General Election Post-event Overview
Appendix: Survey Methodologies
To reach Aboriginal people, Ipsos-Reid targeted northern areas of British Columbia, Alberta, Saskatchewan, Manitoba, Ontario and Quebec. Where known, areas with heavier concentrations of Aboriginal people were also targeted in these provinces and in Atlantic Canada. As a result of this pre-selection process, the survey is not truly representative of the national Aboriginal population, but is instead a proxy.
Included in the survey is an over-sample of 150 Aboriginal people residing north of the 60th parallel. The methodology involved conducting a telephone survey of 556 Aboriginal people in November and December 2000. The margin of error for a sample of this nature is ± 4.2 percent, 19 times out of 20.
The sample has been weighted and is representative of Canada's age and sex composition in accordance with 1996 Census data for Aboriginal people.
Elections Canada, with the collaboration of Ipsos-Reid, conducted a mail-back survey of members of the academic community in January and February 2001.
Elections Canada has compiled, over the years, a list of 160 academics in Canada who have indicated their interest in sharing information on electoral matters with the Office of the Chief Electoral Officer. Forty-nine of these academics (31 percent) responded to the questionnaire. The completed questionnaires were returned to Elections Canada, and the data were analyzed internally.
2000 Canadian Election Study
The principal co-investigators of the 2000 CES are André Blais and Richard Nadeau (Department of Political Science, Université de Montréal), Elisabeth Gidengil (Department of Political Science, McGill University) and Neil Nevitte (Department of Political Science, University of Toronto).
The four researchers are responsible for the design and conduct of the 2000 CES, for which they received funding from the Social Sciences and Humanities Research Council of Canada. Elections Canada and the Institute for Research on Public Policy are partners in this project. Elections Canada sponsored 17 questions in the survey.
The 2000 CES was conducted over three different panels. The first panel, called the Campaign Period Survey (CPS), was conducted during the electoral campaign. The second panel, the Post-Electoral Survey, was conducted in the days following the election. The third panel, the Mail-Back Survey, was a questionnaire sent by mail to volunteer respondents in the months following the election. The sample was distributed across the panels as follows:
A weighting factor was applied to the data for the number of electors in households, and for provinces' and territories' electoral weights.
|2000 CES Panels||Respondents (n)||% of CPS sample||Margins of error|
|Campaign Period Survey||3 651||100%||± 1.6%|
|Post-Electoral Survey||2 862||78%||± 1.8%|
|Mail-Back Survey||1 536||42%||± 2.5%|
Ipsos-Reid conducted telephone interviews with 60 randomly selected candidates from registered parties represented in the House of Commons. The interviews were conducted between January 9 and February 12, 2001.
The margin of error for a sample of this nature is ± 11.3 percent, 19 times out of 20.
Ipsos-Reid conducted a random telephone survey of Canadian citizens 18 years old and over, that took place between November 28 and December 11, 2000. A total of 2 500 Canadians was surveyed.
The sample has been weighted and is representative of Canada's age and sex composition in accordance with 1996 Census data. The margin of error for a sample of this nature is ± 2.0 percent, 19 times out of 20.
|Survey of Electors||Respondents (n)||Margins of error|
|18-34 Years||1 400||± 2.5%|
|Over 35 Years||1 100||± 2.9%|
|Total||2 500||± 2.0%|
Elections Canada, with the collaboration of Ipsos-Reid, conducted a post-electoral study with 99 representatives among 1 083 ethnocultural associations.
The data were collected through a self-administered mail questionnaire from 99 representatives. As a lead-up to the mail questionnaire, Elections Canada sent a letter to the potential respondents within the targeted ethnocultural associations. A follow-up reminder card was also sent to potential respondents. The fielding dates for the survey were between March 23 and May 22, 2001.
Political Party Representatives
Ipsos-Reid conducted telephone interviews with 20 randomly selected representatives of registered political parties, including national directors, agents and members of Elections Canada's Advisory Committee of Political Parties. There were 11 registered political parties in the 2000 general election. Interviews were conducted between January 9 and 25, 2001.
Ipsos-Reid collected the data for this study through a self-administered mail questionnaire sent to Elections Canada's list of 301 returning officers, following the November 2000 general election. In total, 259 returning officers responded to the questionnaire in time for the study.
Ipsos-Reid used a combined mail and telephone methodology for this study. For the associations representing people who are blind or visually impaired, and people with developmental or physical disabilities, a short telephone survey was used. For the associations representing people who are deaf or hard of hearing, a self-administered mail questionnaire was used.
Telephone interviews were conducted with 165 randomly selected associations representing Canadians who are visually impaired, physically disabled, developmentally delayed, those with psychiatric and learning disabilities, and those with low literacy skills. Additionally, 33 mail-back surveys were received from the associations representing Canadians who are deaf or hard of hearing, for a total of 198 respondents.
The margin of error of a sample of this nature is ± 7.1 percent, 19 times out of 20. The fielding dates for this survey were between January 16 and February 23, 2001.
Registered Third Parties
Ipsos-Reid conducted telephone surveys with 21 representatives from the 49 registered third parties. Before scheduling the interviews, Ipsos-Reid sent a letter to third parties as an invitation to participate in the survey. The interviews were conducted between January 11 and January 25, 2001.
Web Site Survey
Participants were recruited from Ipsos-Reid's Internet Panel. Ipsos-Reid selected 991 panellists (including 198 youth Internet panellists between 18 and 24 years old), instructed them to visit Elections Canada's Web site, and asked them to complete an evaluation survey.
The youth component was asked to visit the youth section of the Web site. Initially, respondents were asked to spend some time becoming familiar with the site, and were asked to locate some specific features and to complete some site-specific navigational exercises. Partway through the survey period, it was determined that respondents required some additional instruction on how to locate certain specific features. At that point, additional respondents were invited to participate in the on-line survey, were given specific directions about the location of those site-specific features, and were encouraged to perform the exercises. In both instances, respondents were asked to assess the site and answer the on-line questionnaire.
The Internet panellists had been pre-screened to determine that they had voted in the last election and/or intended to vote in the general election of November 27, 2000. The on-line survey took place between November 27 and December 22, 2000.
The margin of error for a sample of this nature is ± 3.0 percent, 19 times out of 20.