open Secondary menu

Electoral Insight – Aboriginal Participation in Elections

Electoral Insight – November 2003

Aboriginal Participation in Canadian Federal Elections: Trends and Implications

Aboriginal Participation in Canadian Federal Elections: Trends and Implications

Daniel Guérin
Senior Analyst, National and International Research and Policy Development, Elections Canada

This article analyzes the participation of Aboriginal electors in the 2000 Canadian federal general election.Footnote 1 The differences between turnout rates for Aboriginal people and other Canadians and the differences across provinces and territories are analyzed using unpublished electoral data for First Nations reserves in each province and territory. Finally, certain implications of this analysis for the development of measures to encourage Aboriginal participation in federal elections are discussed.

Trends in Aboriginal turnout

Data on voting participation by Aboriginal people in Canada are quite limited. As Bedford and Pobihushchy have pointed out in their study of Aboriginal participation in the Maritimes, "very little attention has been directed at Indian (or more generally, Aboriginal) voter participation in Canadian politics by students of electoral participation."Footnote 2

The few available studies show that, on average, the turnout of Aboriginal people at federal elections is lower than that of other Canadians.Footnote 3 That said, turnout among Aboriginal voters varies greatly across provinces and communities. In some areas, turnout among Aboriginal voters is higher than that of the Canadian population as a whole.

Historically, Aboriginal people and their communities in Canada have faced a series of obstacles to electoral participation. The extension of the franchise to "registered Indians" is relatively recent. Nevertheless, a great deal of progress has been made since 1960, the year the federal government first allowed First Nations people living on reserves to vote at the federal level without having to give up their status under the Indian Act. It is important to recall that the lower turnout of Aboriginal voters observed since 1960 is not so unusual if we consider that it often takes several decades for newly enfranchised people to exercise their right to vote at a rate similar to that of the majority (African-Americans, for instance).

In addition, a significant number of Aboriginal people, as individuals and communities, still regard participation in non-Aboriginal elections or plebiscites as a threat to their unique rights, their autonomy and their goals of self-governance. Such persons hold a philosophical belief about the legitimacy of Aboriginal self-governance that differs fundamentally from that of the Canadian government; and may view other (non-Aboriginal) governments as irrelevant, even alien. Joan Carling suggests that national party system and electoral processes can be viable and meaningful to certain groups of Indigenous peoples in various national settings if there is democratic space within nation-states that provides an equal playing field for the participation of Indigenous peoples (in general); and that creating this democratic space should include recognition and respect for Indigenous peoples' systems of decision-making and mechanisms for self-governance.Footnote 4

Whatever their reasons for non-participation in Canadian federal elections, promoting greater involvement by Aboriginal people must be situated within a complex historical, cultural and political context. Thus, while the concern here is to understand the barriers to Aboriginal electoral participation in Canadian society, it is to be noted that not all such barriers are externally induced. Indeed, as discussed in other articles in this issue, some are more a result of voluntary political choices of Aboriginal people than the accessibility and administrative efficacy of various programs implemented by Elections Canada.

Factors behind low Aboriginal turnout

Evidence to date on Aboriginal voter turnout suggests the following questions: What structural and political factors account for the weak participation of certain Aboriginal peoples in federal elections? What cultural factors are related to Aboriginal participation in federal elections? What are the attitudes and values associated with not voting among Aboriginal people?

According to the Royal Commission on Electoral Reform and Party Financing, Aboriginal turnout depends on a number of factors related to the context of each election, such as the presence or absence of debate about issues that are relevant to Aboriginal people, and, in particular, the presence of Aboriginal candidates. The Commission identified a series of additional factors to explain the traditionally low participation of Aboriginal peoples, which may be grouped into several categories, and are based on the on- and off-reserve distinction.Footnote 5

On reserves:

  • Social conditions and other socio-demographic factors
  • Cultural factors
  • Political factors
  • Problems of communicationFootnote 6

Off reserves:

  • Geographic dispersion makes outreach challenging
  • Weak social connectedness
  • Social conditions

Empirical data on the socio-psychological characteristics of the Aboriginal voter population are essential for understanding the lower turnout of Aboriginal people at federal elections. Also essential are data on factors related to the demographic structure of Aboriginal communities. As noted above, a significantly higher proportion of the Aboriginal population is young (under 25) than in the Canadian population as a whole. This age structure tends to have a strong negative impact on Aboriginal participation, as it is well-known that youth vote at a rate significantly lower than older groups.Footnote 7 Independently of age, knowledge about the electoral process is also demonstrated by research to have a significant influence on the likelihood of voting.

Turnout in recent federal elections

In March 2001, Ipsos-Reid reported the findings of a survey of 556 Aboriginal people, commissioned by Elections Canada, about their participation in the November 2000 federal election. Respondents were from the northern areas of the provinces of British Columbia, Alberta, Saskatchewan, Manitoba, Ontario and Quebec, and other regions in these provinces and in Atlantic Canada with known heavier concentrations of Aboriginal people, as well as an over-sample of 150 Aboriginal people residing north of the 60th parallel.Footnote 8 In that survey, seven in ten Aboriginal respondents (70 percent) indicated that they voted in the 2000 election. The proportion of respondents who say they voted may be inflated, however.Footnote 9

In an internal Elections Canada study on the participation of Aboriginal people at federal elections prior to 2000, Jean-Nicolas Bustros points out that the only empirical evidence on Aboriginal turnout rates available for the whole country is provided by the detailed poll-by-poll reports of the Chief Electoral Officer. However, he observes that evidence can only be collected from polls with an exclusively Aboriginal population. Using this evidence, he reported a participation rate in such polls of 41 percent in the 1992 referendum, 38 percent in the 1993 general election and 40 percent in the 1997 general election. Bustros adds that Aboriginal voters who cast, or omit to cast, their vote in "mixed" polls could not be traced; nor was it possible to determine turnout rates for the large urban Aboriginal population.Footnote 10

Bedford and Pobihushchy examined trends in voter turnout among Aboriginal people for federal, provincial and band elections in New Brunswick, Nova Scotia and Prince Edward Island between 1962 and 1993. They found significant variations in turnout rates for on-reserve Status Indians. For example, participation rates were:

  • 1988 federal election: 17.8 percent in New Brunswick, 54 percent in Nova Scotia and 72.8 percent in Prince Edward Island
  • provincial elections: 27.6 percent in New Brunswick in 1991, 45.2 percent in Nova Scotia in 1993 and 78.4 percent in Prince Edward Island in 1993
  • band elections: in New Brunswick and Nova Scotia, consistently around the 90 percent mark from 1972 to 1992

The authors used voting results only from those polls situated entirely within the boundaries of reserves; results from other polling stations would have included information on participation by both First Nations and non-Aboriginal persons. Nevertheless, Bedford and Pobihushchy's research method for assessing Aboriginal voter participation may be considered the "gold standard" for such research.


During the 1992 federal referendum, Elections Canada's information program for the country's many Aboriginal groups included publishing the referendum question about the Charlottetown Accord in 37 Aboriginal languages. This booklet was available at all polling stations.

Analysis of Aboriginal turnout in the 2000 federal election

Empirical data on turnout rates in the 1992 referendum and recent federal elections, especially for exclusively Aboriginal polls, show a noticeable variation across Aboriginal communities and regions of the country. Bustros indicated that in northern polls, where Aboriginal candidates were present, turnout rates were comparable to and even exceeded those of non-Aboriginal communities; however, in the larger, southern Aboriginal communities, turnout rates were generally much lower than the Canadian average.Footnote 11

New analysis conducted by Elections Canada's National and International Research and Policy Development Directorate illustrates similar variations during the last federal election. Using the same methodology as the Bedford and Pobihushchy study, this analysis examined participation rates among Aboriginal people living on reserves at the 2000 federal election.

The analysis included the 264 First Nations reserves whose boundaries corresponded exactly with the federal electoral polling divisions at that time. Overall, the turnout rate for all 296 polling stations covered by the study was 47.8 percent – 16 percent lower than the turnout among the general population during the same election.Footnote 12 At the same time, as shown in Figure 1, there was considerable variation in turnout rates across provinces and territoriesFootnote 13 in the 2000 election:

  1. High-turnout provinces and territories: Prince Edward Island, Saskatchewan, Nunavut,Footnote 14 Alberta and British Columbia
  2. Medium-turnout provinces and territories: Northwest Territories, Ontario, Newfoundland, New Brunswick and Nova Scotia
  3. Low-turnout provinces: Manitoba and Quebec

More information about the composition of the Aboriginal population in Canada can be found on the Statistics Canada Web site at www.statcan.ca/english/Pgdb/demo38a.htm.

Although comparable data are not available, there is some evidence to suggest that turnout is lower in the various urban centres with significant proportions of Aboriginal people. Current estimates are that about half the Aboriginal population does not reside on reserve lands.Footnote 15 For example, according to a 2001 Ipsos-Reid survey,Footnote 16 Aboriginal people living in urban areas were three times less likely to have said they voted in the 2000 federal election than those living on reserves. It is also worth noting that the "youth factor" has to be taken into account when interpreting these results. According to 2001 Canadian census figures, approximately 50 percent of the Aboriginal population was 24 years of age or younger, compared to 31 percent of the general Canadian population.Footnote 17 If this factor were taken into account, the overall difference between the turnout rate of Aboriginal people and that of the general population would probably be reduced. Finally, under-registration of Aboriginal electors may mean that available turnout rates are somewhat higher than they would be if coverage were more complete.

We can draw a number of lessons from the available research on Aboriginal participation in Canadian elections.

  • Participation of Aboriginal people at federal elections is usually lower than that of the general population, although there are significant variations across the country.
  • Aboriginal turnout is affected by the context of each election, including whether there are Aboriginal candidates, and the presence or absence of debate about issues that are important to Aboriginal people.
  • Various factors, often reinforcing one another, partly explain lower turnout rates of Aboriginal people.
    These are:
    • lower socio-economic and educational levels
    • disengagement from, or even opposition to, participation in the federal election process
    • geographical dispersion, particularly in northern Canada
    • cultural diversity (e.g. large number of languages)
    • difficulties in reaching the significant Aboriginal population not living on reserves

Aboriginal participation in the 2000 federal election

This examination of Aboriginal participation in the 2000 federal election is one of the first empirical analyses conducted in all the provinces and territories for a single federal election. Using a methodology similar to that adopted by Bedford and Pobihushchy in their pioneering study of Aboriginal participation in the Maritimes, it was demonstrated that the question of Aboriginal participation in federal elections is more complex than many observers of the political scene have generally believed.

According to our analysis, the Aboriginal participation rate in the 2000 federal election was 48 percent. This is 16 points below the rate for the Canadian population as a whole. These results tend to confirm a point that has been widely acknowledged in studies of Aboriginal participation in federal elections, i.e. their habitually low level of participation. However, our results for individual provinces and territories also tend to reveal a wide variation in the Aboriginal participation rates in federal elections.

In fact, there are three different groups of provinces and territories. Aboriginal electors in the first group (four provinces and one territory) had a turnout rate comparable to or slightly lower than the overall rate. The average turnout rate for Aboriginal electors in this group in the 2000 election was 56.3 percent.

In the second group of four provinces and one territory, the turnout rate for Aboriginal electors was more than 10 points lower than the overall turnout rate, that is, an average of 43.1 percent. Finally, there is a small group of two provinces (Manitoba and Quebec) where the turnout rate for Aboriginal electors was significantly lower than the rate for the Canadian public as a whole – 36 percent, which is roughly 30 points lower than the overall rate during that election.

The research could not determine the reasons for these significant variations. However, a number of hypotheses are possible. First, one might think that these variations, to a great extent, reflect the fact that Aboriginal participation in federal elections depends largely on cultural and social factors. This hypothesis can be called the socio-cultural explanation. The precise mechanism of these influences is difficult to judge at present, given the almost total lack of research on this question.

Another possible explanation is that the participation rates are influenced more by the specific context of each election in each community than by long-term forces such as those that form the basis of the socio-cultural hypothesis. This explanation can be called the contextual hypothesis. It is also plausible, inasmuch as research has shown in the past that variables specific to the local electoral context, such as the presence of Aboriginal candidates or of debates on issues relevant to members of the First Nations, can have a marked influence on the turnout rate of Aboriginal people.

It is possible that a combination of several explanatory factors lies at the root of the variations in Aboriginal participation rates in Canada. That is why it would be advisable to develop a multivariate (and preferably multilevel) analytical model that takes into account factors relating to individuals, as well as factors relating to the social and political environment. In any case, the research reported in this article allows us to envisage a further step in the investigation of this question, that is, an analysis of other federal elections using the same analytical method.

Conclusion

Aboriginal Languages by Community, 1996
Aboriginal Languages by Community, 1996
The map Aboriginal Languages by Community, 1996 shows the distribution and population of Aboriginal communities categorized by language family. The Aboriginal languages of Canada belong to 11 major language families. Most families consist of separate but related member languages, and each member language may include several dialects. The Atlas of Canada. © 2003.

In closing, some of the limitations of this research should be pointed out. The trends derived from this research are based primarily on data gathered on First Nations reserves. In order to extend the conclusions to the entire Aboriginal population of Canada, it will be necessary in subsequent analyses to include data on the electoral participation of Aboriginal people living off reserves, and particularly in cities. For the moment, there is little data available. It would thus be advisable to include larger samplings of these segments of the Aboriginal population in upcoming electoral research and other surveys.

Finally, a word about some of the possible implications of this research for the development of programs to encourage participation in federal elections. The fact that we noted significant variations in participation rates across provinces and territories means it is necessary to develop measures adapted to the different conditions of the varied communities and provinces/territories. This will require qualitative information to complement the quantitative data gathered through research such as this. Such information could come from consultations with the various groups concerned, Aboriginal officials and opinion leaders, as well as other electors living in the diverse Aboriginal communities throughout Canada. These consultations could also serve to test any initiatives being developed and to ensure a reasonable level of acceptance among the communities targeted by such measures.

Notes

Footnote 1 The author thanks Asifa Akbar, Analyst, National and International Research and Policy Development, Elections Canada, for her noteworthy contribution to this research.

Footnote 2 David Bedford and Sidney Pobihushchy, "On-Reserve Status Indian Voter Participation in the Maritimes," Canadian Journal of Native Studies Vol. 15, No. 2 (1995), pp. 255–278.

Footnote 3 R. Gibbins, "Electoral Reform and Canada's Aboriginal Population: An Assessment of Aboriginal Electoral Districts," in R. A. Milen, ed., Aboriginal Peoples and Electoral Reform in Canada, Vol. 9 of the Research Studies of the Royal Commission on Electoral Reform and Party Financing (Toronto: Dundurn Press, 1991), p. 160; Elections Canada, "Elections Canada Initiatives Concerning Aboriginal Electors: Elections Canada 1992–1999" (Presentation to the Assembly of First Nations, January 1999); Elections Canada, Thirty-fifth General Election 1993: Official Voting Results (Ottawa: Chief Electoral Officer of Canada, 1993); Chief Electoral Officer of Canada, Thirty-sixth General Election 1997: Official Voting Results (CD-ROM: Catalogue No. SE-1-1997-MRC).

Footnote 4 Joan Carling, "Indigenous Peoples' Involvement in National Politics," in Kathrin Wessendorf, ed., Challenging Politics: Indigenous People's Experiences with Political Parties and Elections (Copenhagen: International Work Group for Indigenous Affairs, 2001).

Footnote 5 Royal Commission on Electoral Reform and Party Financing, Final Report, Vol. 1 (Ottawa: Supply and Services Canada, 1991), pp. 168–170.

Footnote 6 It is a well recognized fact that "less than adequate communications media are responsible for the diminished awareness and interest of Aboriginal people in the electoral process." (Royal Commission on Electoral Reform and Party Financing, Final Report, Vol. 1, Section "Equality and Efficacy of the Vote," p. 170.) See also Valerie Alia, "Aboriginal Peoples and Campaign Coverage in the North," in R. A. Milen, ed., Aboriginal Peoples and Electoral Reform in Canada, Vol. 9 of the Research Studies of the Royal Commission on Electoral Reform and Party Financing (Toronto: Dundurn Press, 1991), pp. 105–106.

Footnote 7 See Jon H. Pammett and Lawrence LeDuc, "Explaining the Turnout Decline in Canadian Federal Elections: A New Survey of Non-voters," (Ottawa: Elections Canada, March 2003), p. 20.

Footnote 8 As a result of this pre-selection process the survey is not truly representative of the national Aboriginal population but is simply a proxy; there is a greater margin of error for subgroups of the survey population; the margin of error for this sample was plus or minus 5 percentage points 19 times out of 20; the sample was weighted and is representative of Canada's age and gender composition according to the 1996 census data for Aboriginal people, but has not been weighted according to the actual voter turnout in the November 2000 election. Ipsos-Reid, Elections Canada Survey on Aboriginal Participation, Ottawa, June 2001.

Footnote 9 The difference between the proportion of respondents who say that they voted in the general election and the estimated proportion of Aboriginal people who voted may be attributed to response bias and acquiescent bias.

Footnote 10 Jean-Nicolas Bustros, "Electoral Participation of Aboriginal People: Summary of Previously Conducted Research and Analysis" (Ottawa: Elections Canada, Legal Services Directorate, March 2000).

Footnote 11 Gibbins, "Electoral Reform and Canada's Aboriginal Population," p. 160.

Footnote 12 We use here the turnout rate calculated after the Register was purged of duplicates, i.e. 64 percent (61 percent before the elimination of duplicates).

Footnote 13 Yukon was excluded from the analysis because there was insufficient valid data.

Footnote 14 In Nunavut, 22 of 25 Inuit communities were included in our study. It is worth noting that in Nunavut, there are 22 Inuit communities and no First Nations reserves. In Nunavut, the turnout rate is 10 percentage points higher than in the Northwest Territories, where we included 23 First Nations reserves and 5 Inuit communities.

Footnote 15 It is difficult to estimate the total Aboriginal population reliably, but a figure of 900,000–1 million is often presented as a plausible approximation.

Footnote 16 Elections Canada Survey on Aboriginal Participation.

Footnote 17 Statistics Canada, Aboriginal Population by Age Groups, 2001 census, www.statcan.ca/english/Pgdb/demo38a.htm; Statistics Canada, 2001 Census, Age and Sex, for Canada, Provinces, and Territories, www12.statcan.ca/english/census01/products/highlight/AgeSex/Page.cfm?Lang=E&Geo=PR&View=1&Table=4a&StartRec=1&Sort=2&B1=Median&B2=Both.


Note: 

The opinions expressed are those of the authors; they do not necessarily reflect those of the Chief Electoral Officer of Canada.