open Secondary menu

Survey of Candidates Following the 40th General Election

Appendix

Methodology Note

This research consisted of a census survey of candidates of the 40th federal general election.

The following specifications applied to the survey:

  • A mixed-mode data collection strategy was used to survey candidates. The primary mode of data collection was telephone interviewing, with additional self-administered methodologies (i.e. online, fax, mail, and e-mail) included to help maximize the response rateFootnote 13. In total, 851 interviews were conducted by phone, 23 onlineFootnote 14, and 3 by mail.
  • The sample was provided, in electronic format, by Elections Canada based on the information contained in the candidates' nomination papers. It contained the candidate's complete name, full mailing address with postal code, phone number(s), where available, and party affiliation and federal riding. Phoenix augmented the list by completing as many missing phone numbers as possible.
  • In advance of the fieldwork, a notification letter was sent on Elections Canada letterhead to all candidates. The letter explained the background and purpose of the research/consultation, introduced Phoenix as the firm conducting it, offered assurances of confidentiality, and encouraged participation. The letter, signed by Marc Mayrand (Chief Electoral Officer), contained a phone number at Elections Canada for those who wished to validate the legitimacy of the research/consultation and answer any questions that candidates may have about it. As well, a 1‑800 number was included in the letter for candidates to use to call Phoenix to schedule an interview (if they so desired).
  • All 1,601 candidates were contacted by phone (except those for whom no phone numbers were available) and invited to complete the survey at that time or to schedule one at a more convenient time. Those for whom phone numbers could not be obtained were sent a mail package, containing a modified background letter and a copy of the questionnaire. The modified letter contained a password and URL to access the survey, in addition to the 1‑800 number to contact Phoenix to schedule an interview.
  • Candidates who initially declined to take part in the research were offered the opportunity to complete the survey via an alternate self-administered mode of data collection. This was also available to respondents who specifically requested a self-administered method.

The following specifications also applied to the survey:

  • The questionnaire, designed in close consultation with Elections Canada, was pre-tested in English and in French, and interviews were recorded for review by Phoenix and Elections Canada. This resulted in making adjustments that limited the response burden on candidates and ensured further data quality. The results of these interviews were included in the final dataset.
  • As an additional quality control measure, top-line frequencies were run after 50 interviews were completed in order to confirm that respondents were being routed correctly through the programmed survey.
  • A detailed interviewer briefing note was prepared by Phoenix (and approved by Elections Canada) to brief interviewers and guide the data collection process.
  • A small team of elite interviewers was used for this study. In terms of experience, all individuals assigned to this study had worked for at least two years conducting telephone interviews.
  • All interviewing was conducted in the respondent's official language of choice.
  • Calling was conducted at different times of the day and the week to maximize opportunities for contact and up to 10 call-backs were attempted before a sample record was retired. Voicemail messages were left on the third call attempt, referencing the letter and providing a 1‑800 number for candidates to contact Phoenix to schedule or conduct the interview.
  • In cases where contact resulted in a refusal, a senior interviewer re-contacted the candidate between three and five days following the refusal to emphasize the importance of his/her participation and attempt to schedule an interview by phone or alternative method. If the candidate still declined to take part, no further contact attempt was made. Similarly, respondents who said they would complete the survey using one of the self-administered methods, but who did not actually complete it (after the reminder regime has run its course), were contacted one last time by a senior interviewer to try to complete the interview.
  • The sample was carefully monitored throughout the data collection period to ensure effective sample management.
  • Sponsorship of the study (i.e. Elections Canada) was revealed.
  • The survey was registered with the National Survey Registration System, and respondents were informed about this.

The following table presents information about the final call dispositions for this survey, as well as calculation of the response rate (using MRIA's Empirical formula):

40th FGE Candidates Call Distribution
Total numbers attempted 1601
Out-of-scope invalid (i.e. not in service, faxes, pagers, businesses) 107
Unresolved (U) 539
No answer/answering machine 539
In-scope - Non-responding (IS) 2
Language barrier 0
Incapable of completing (ill/deceased) 2
Callback (respondent not available) 0
Total asked 953
Refusal 75
Termination 1
In-scope - Responding units (R) 877
Completed interview 877
Refusal rate 8%
Response rate 59%

Sub-group Characteristics

The following is a breakdown of the sub-groups reported on in this study.

Demographic Banners (All Questions):

  • Region (admin data):
    • Atlantic Canada
    • Quebec
    • Ontario
    • The Prairies (SK/MB/Alberta)
    • British Columbia
    • Territories
  • Age (Q39):
    • Under 30 years
    • 30‑49
    • 50 years and older
  • Language (admin data):
    • French
    • English
  • Gender (admin data):
  • Number of Candidacies (Q40):
    • One
    • Two
    • Three or more
  • 40th General Election Outcome (admin data):
    • Elected
    • Not elected
  • Political affiliation (admin data):
    • HOC represented party
    • Other political parties
    • No party affiliation/independents

Attitudinal Banners (all questions):

  • Overall satisfaction with administration of election (Q1)
    • Satisfied (scores of 4‑5)
    • Neutral (3)
    • Dissatisfied (scores of 1‑2)
  • Satisfaction with returning officer performance in own riding (Q2)
    • Satisfied (scores of 4‑5)
    • Neutral (3)
    • Dissatisfied (scores of 1‑2)
  • Satisfaction with overall service quality from Elections Canada (Q35)
    • Satisfied (scores of 4‑5)
    • Neutral (3)
    • Dissatisfied (scores of 1‑2)

In addition to the above sub-groups, the following were run for specific sections and sub-sections of the questionnaire:

Nomination Requirements Sub-section Only:

  • Ease of complying with nomination requirements (Q3)
    • Relatively easy (Very easy/moderately easy)
    • Relatively difficult (Not easy at all/not very easy)
  • Attendance at financial requirements information sessions (Q8)
    • Self or representative attended (includes official agents, campaign managers, and other representatives)
    • No one attended

Registration, Voting and Counting Processes Sub-section Only:

  • Satisfaction with elector registration (Q9)
    • Satisfied (scores of 4‑5)
    • Neutral (3)
    • Dissatisfied (scores of 1‑2)

Services and Products Provided Section Only:

  • Attendance at all candidates meeting (Q18)
    • Self or representative attended (includes official agents, campaign managers, and other representatives)
    • No one attended

Footnote 13 A mixed-mode approach not only increases the likelihood of contacting hard-to-reach respondents, it offers all respondents the opportunity to take part in the research using a method that they find convenient.

Footnote 14 Before going live, the programming was tested by Phoenix and reviewed by Elections Canada to ensure proper functioning.