National Register of Electors. Updates: November 2017 Annual Lists

Transcription

1 National Register of Electors Updates: November 2017 Annual Lists

2 Table of Contents Introduction...3 Overview Background Updating Quality Coverage Currency Accuracy Interpretation Reliability of Quality Estimates Future Plans Conclusion...13

3 Introduction In accordance with the Canada Elections Act, Elections Canada provides lists of electors (containing name, address and unique identifier) to members of Parliament by November 15 of each year and, upon request, to each federal registered party for any electoral district in which the party endorsed a candidate in the last general election or by-election. This document provides an overview of the primary sources of information used to update the National Register of Electors, as well as information on the quality of the lists. National Register of Electors Updates: November 2017 Annual Lists 3

4 Overview The data extracted for this purpose reflect updates to the National Register of Electors made on the basis of information received from federal, provincial and territorial administrative sources, from lists of electors provided by other electoral agencies and from the Elections Canada Online Voter Registration Service. Since November 15, 2016, some 380,000 new electors were added to the Register. The majority (291,000) are electors aged 18 to 24. Records for some 270,000 deceased electors were removed from the Register, and 34,000 duplicate records were deleted. In addition, address changes were made for over 3.3 million electors to reflect their moves. The current lists also reflect improvements made to address information. The November 2017 annual lists contain 93.2 percent of the 27.2 million eligible electors (see Table 1). Some 86.5 percent of the eligible electors, or 23.5 million electors, are registered at their current residential address. Among registered electors, 92.9 percent are listed at their current residential address. Table 1. Registered and Eligible Electors, and Estimates of Quality, Lists of Electors, November 2017 and November 2016 Annual Lists November 2017 November 2016 Registered electors 25,926,769 25,884,187 Eligible electors 1 27,182,615 27,083,333 Coverage 2 (%) Currency (%) Accuracy (%) The electoral population is estimated using information from the 2016 Census of Population and National Household Survey, adjusted for census net undercoverage, as well as demographic components of growth. 2 The coverage for 2017 was adjusted from 95.4 percent to 93.2 percent to account for deceased electors, duplicates and non Canadian citizens. The adjustment is similar to those made in previous years. Currency and accuracy were also adjusted accordingly. 4 National Register of Electors Updates: November 2017 Annual Lists

5 1. Background The National Register of Electors is a database of Canadians who are qualified to vote in federal elections. The Register contains the name, gender, date of birth, address and unique identifier of each elector. Elections Canada uses the information in the Register to produce preliminary lists of electors for federal elections, by-elections and referendums, and to provide lists to members of Parliament and, on request, to registered political parties in November of each year. The Canada Elections Act also allows Elections Canada to provide Register data to other jurisdictions (provinces, territories, municipalities and school boards) for electoral purposes, where data-sharing agreements have been signed. 2. Updating According to Statistics Canada, some 14 percent of elector information changes every year. These changes involve moves (11 percent), new 18-year-olds (2 percent), new citizens (less than 1 percent) and deaths (1 percent). To reflect these demographic changes, the National Register of Electors is regularly updated using federal, provincial and territorial administrative data sources. It may also be updated using electoral lists from other jurisdictions. Revisions made to lists of electors during federal elections are another important source of Register updates. Since November 15, 2016, the Register has been updated using the following sources: 1. Canada Revenue Agency: The name, address and date of birth of tax filers who indicated that they are Canadian citizens and who consented to their information being provided to Elections Canada. For the 2016 taxation year, 81 percent of tax filers consented on their T1 tax returns. 2. Immigration, Refugees and Citizenship Canada: The name, address, date of birth and gender of new Canadians who consented to their information being provided to Elections Canada once their citizenship was granted. As of June 2017, some 96 percent of new citizens consented to this transfer. 3. National Defence: The name, gender, date of birth and address for voting purposes of Canadian Forces Regular Force members, obtained from individual Statement of Ordinary Residence forms. Also, the removal of deceased members. 4. Provincial and territorial motor vehicle agencies: Driver information from each province and territory except for Quebec, where this information is obtained from the Quebec provincial electoral list. National Register of Electors Updates: November 2017 Annual Lists 5

6 5. Provincial and territorial vital statistics agencies: Information on deceased persons from each province and territory except for Quebec, where this information is obtained from the Quebec provincial electoral list. The Canada Revenue Agency and provincial and territorial motor vehicle agencies also provide information on deceased persons. 6. Provincial and territorial electoral lists: Elector information from the permanent list of electors maintained by the Directeur général des élections du Québec. Also, the list of electors received from Elections PEI (November 2016 plebiscite) and Elections Nova Scotia (May 2017 general election). 7. Elector-initiated updates: Requests from electors to have their information updated or added to the Register, in addition to requests to opt out of the Register entirely or to opt out of sharing their information with other jurisdictions. Since the establishment of the Register in 1997, there have been some 4,000 requests to opt out of the Register entirely and some 150 requests to opt out of sharing with other jurisdictions. The electors who have requested to opt out entirely do not appear on the November 2017 lists. 8. Online Voter Registration Service: Since November 15, 2016, some 4,153 changes of address and some 27,937 registration requests were submitted through this online service. 9. Removal of duplicates: Verifications are made regularly to identify and remove duplicates (individuals for whom more than one record appears) from the Register. Since November 15, 2016, some 34,000 records were identified as duplicates and were removed from the lists. Recent changes from the federal by-elections held on October 23, 2017, in the electoral districts of Lac-Saint-Jean (Quebec) and Sturgeon River Parkland (Alberta) are still being processed and therefore are not included in the November 2017 lists of electors. 6 National Register of Electors Updates: November 2017 Annual Lists

7 3. Quality The main indicators used to measure the quality of information in the National Register of Electors are coverage, currency and accuracy. These indicators are described below, along with the corresponding estimates for the November 2017 lists. Refer to section 3.4 for information on how to interpret these measures. 3.1 Coverage Coverage measurements involve determining what proportion of the population is represented on the lists. This entails comparing the number of electors on the lists with an estimate of the total number of Canadian citizens aged 18 and over. This makes it possible for Elections Canada to assess the quality of the Register relative to its national registration target of 92 percent. Elections Canada prepares estimates of eligible electors based on information provided by Statistics Canada and Immigration, Refugees and Citizenship Canada. A demographic approach is used to derive estimates at the electoral district level using detailed information on new adults, deceased electors, new citizens, electors who move between electoral districts and electors leaving the country permanently. This information is used to estimate demographic changes. The sum of the electoral districts provides provincial and national estimates. Coverage estimates for Canada and the provinces and territories are shown in Chart 1 and Table 2. Comparing registered electors with eligible electors estimates indicates that the November 2017 lists include some 93.2 percent of Canadian electors which exceeds the Elections Canada target of 92 percent. Some 30,000 electors do not appear on the lists because their address has not yet been assigned to an electoral district and polling division. Coverage may vary by region because of various factors: Demographic changes; Availability of data sources used to update the Register; Rates of consent to share information; Information for some people does not appear in any administrative files used for updating purposes; Some records cannot be positively matched with information in the Register; Lapse of time between the demographic change and the time the information is provided to and processed by Elections Canada. National Register of Electors Updates: November 2017 Annual Lists 7

8 The coverage of lists in the Atlantic provinces, Quebec, Ontario and British Columbia is above or equal to the national target (92 percent). Coverage of lists in Manitoba, Saskatchewan and Alberta is lower than the national target. Among the territories, only Yukon has a coverage exceeding the national target. Higher demographic growth partly explains the lower coverage of the lists in the Northwest Territories and Nunavut. 8 National Register of Electors Updates: November 2017 Annual Lists

9 Registration of young electors poses additional challenges as the information from administrative sources may be available to Elections Canada only several months after they have reached the eligibility age. Chart 2 illustrates that the gap between registered electors and eligible electors decreases steadily with age. 18-year-olds, who have just become eligible to vote, have the largest gap with only 142,000 registrations out of 385,000 eligible electors. The gap decreases considerably for 19-year-olds (241,000 registrations out of 396,000 eligible electors) and continues to decrease up to the age of After that age, the Register contains information for nearly all eligible electors. 1 According to the Retrospective Report on the 42nd General Election of October 19, 2015 (Chief Electoral Officer of Canada, 2016, page 19), Canadians aged 18 to 34 benefited the most from registration and revision activities during the 42nd general election. Despite successful outreach to this age group, youth aged 18 to 34 remain registered at a lower rate than older electors. For example, among 18-year-olds, registration rates increased from 27 percent at the call of the election to 60 percent on final voters lists, an increase of 33 percentage points. National Register of Electors Updates: November 2017 Annual Lists 9

10 3.2 Currency The second measure of the quality of the lists is the currency of the information on electors, particularly the address information. Measuring the currency of the lists comes down to answering the following question: What proportion of eligible electors are registered at their current address? The currency of lists is estimated by comparing the number of updates made to electors addresses in the Register with the estimated number of moves in each region of the country. Updating addresses presents a number of challenges. In addition to the factors listed above, the high number of moves and the variations by region could also explain differences in currency. That is why the national currency target, at 80 percent, is lower than the target for coverage. Table 2 also provides estimates of currency for Canada and the provinces and territories. Some 86.5 percent of eligible electors across the country are registered at their current address. The currency of lists is above the national target (80 percent) in all provinces except for Saskatchewan and Alberta. Among the territories, only Yukon exceeds the national target. Lower currency in the Northwest Territories and Nunavut is partly explained by its lower coverage and higher mobility rate. Various measures are implemented at election time to improve these lists. 3.3 Accuracy Accuracy of the electoral lists is derived by dividing the currency estimates by the coverage estimates. It represents the proportion of registered electors listed at their current address. In other words, it answers the question, What proportion of electors on the lists is at the current residential address? Whereas currency indicates the proportion of the eligible electors registered at their current address, accuracy measures the proportion of registered electors listed at their current address. Refer to section 3.5 for information on how to interpret these measures. 10 National Register of Electors Updates: November 2017 Annual Lists

11 Among electors appearing on the November 2017 lists, it is estimated that 92.9 percent of the electors on the lists have current residential address information. In all provinces and territories, 9 (or more) registered electors out of 10 are listed at their current address. Table 2. Estimated Percentages of Coverage, Currency and Accuracy 1, Canada, Provinces and Territories, November 2017 Annual Lists of Electors Coverage (national target 92%) Currency (national target 80%) Accuracy Canada Newfoundland and Labrador Prince Edward Island Nova Scotia New Brunswick Quebec Ontario Manitoba Saskatchewan Alberta British Columbia Yukon Northwest Territories Nunavut The estimates are adjusted for deceased electors, duplicates and non Canadian citizens. The quality estimates are subject to sampling errors. See section 3.5, Reliability of Quality Estimates. National Register of Electors Updates: November 2017 Annual Lists 11

12 3.4 Interpretation To better understand the practical impact of the quality measures of the lists, it is important to consider how the data are used in a federal election. Shortly after the election is called, voter information cards (VICs) are mailed to electors appearing on the preliminary lists of electors that are extracted from the Register. 9 (or more) out of 10 registered electors (accuracy) will be sent VICs addressed to them that correctly indicate where they can vote. Returning officers provide candidates with a Statement of Quality of the Preliminary Lists of Electors document during an election. This document provides quality measures and estimated revision volumes for their electoral district. For example, lower currency estimates may result in a higher number of revisions or polling day registrations. The document also contains accounts of recent initiatives for improving the lists and targeted revision plans. The Chief Electoral Officer instructs each returning officer to discuss list quality issues with candidates during the candidates briefing meeting, which must be held shortly after all candidates have been confirmed. Candidates are encouraged to raise questions and issues concerning the revision process and to advise the returning officer if they believe areas of the electoral district could benefit from additional targeted revision efforts. 3.5 Reliability of Quality Estimates Elections Canada s quality estimates are subject to periodic validations. Since 2005, the agency has conducted five data quality confirmation studies, using random samples of approximately 40,000 electors selected from the Register, to determine whether their information is correct and up to date. Independent estimates of quality can also be derived from these studies. The results of the most recent study carried out in fall 2014 indicate coverage estimates identical to those obtained using the quality model at the national level and a difference of 2 percentage points for currency and accuracy. For the provinces, differences are minor and fall within the confidence intervals associated with the study. For the country as a whole, the confidence interval for the coverage estimate is ±0.1, 19 times out of 20, and variations for the provinces/territories are between ±0.1 and ±3.2. The confidence interval for the currency estimate is ±0.5 for Canada as a whole, and variations for the provinces/territories are between ±0.8 and ±5.0. The results of the study also serve as a basis for producing the quality estimates. 12 National Register of Electors Updates: November 2017 Annual Lists

13 4. Future Plans Elections Canada is exploring the possibility of providing the lists of electors on a USB key instead of a CD-ROM in November In addition, the agency is planning to implement an online portal to give federal political entities (e.g. members of Parliament and political parties) secure online access to lists of electors and associated products. Once the portal is available, Elections Canada will no longer mail lists of electors. 5. Conclusion Elections Canada estimates that the November 2017 annual lists from the National Register of Electors include some 93.2 percent of the 27.2 million eligible electors, and that some 86.5 percent of the eligible electors are registered at their current residential address. The proportion of registered electors listed at their current residential address is estimated to be 92.9 percent. Although quality estimates are high, it is important that electors and political entities understand that revision procedures are essential to maintaining this level of quality. National Register of Electors Updates: November 2017 Annual Lists 13