Why do municipal electors not vote?
Voter turnout in Canada has declined in the last several decades and is now among the lowest in the western world. In this paper, we examine, through telephone surveys, the reasons why electors did not exercise their right to vote at the municipal level. When non-voters were asked why they did not vote, the most frequent reason was lack of information about candidates and the poor quality of the candidates. The person most likely to vote is a senior, upper income, born outside Canada, owning their own home, and politically and community involved. These results are not dissimilar to those in a recent survey of federal electors.
Keywords: Elections, voting
La participation electorale au Canada a diminuee depuis les dernieres decennies et elle est desormais parmi la plus basse des pays occidentaux. Dans cet article nous examinons, par le biais d'enquete telephoniques, les raisons pour lesquelles les electeurs n'ont pas exerces leur droit de vote lors des elections municipales. La raison principale pour ne pas voter parmi les repondants qui ne se sont pas presentes aux urnes etait le manque d'information ainsi que la faible qualite des candidats. La personne la plus susceptible de voter est une personne du troisieme age, ayant un niveau de revenu eleve, proprietaire, nee a l'exterieur du Canada et impliquee sur la scene politique ainsi que dans sa communaute. Ces resultats sont conformes a ceux obtenus lors d'une enquete recente sur les electeurs aux elections federales.
Mots cles : Elections municipales
In the past several decades, voter turnout in Canada has declined for all three levels of government--federal, provincial, and municipal. Numerous studies have examined the reasons for not voting in federal elections (summarized in Bakvis, 1991). Most recently, Jon M. Pammett and Lawrence LeDuc were commissioned by the Chief Electoral Officer of Canada to examine why a significant number of Canadians did not vote in the recent federal election. The reasons included "lack of interest in the election, negative attitudes toward politics, and personal/administrative factors" (Pammett and LeDuc, 2003:1).
Although much research, both national and international, has been done on national elections, very little analysis has been done of local elections where the turnout is even lower. To date, the only Canadian study of municipal elections that focuses on reasons for non-voting is Peter Boswell's (1991) analysis of the 1990 mayoralty election in St. John's. The Canadian Federation of Mayors and Municipalities (1967) also produced a study of voter turnout, but did not analyze the reasons for non-voting.
The purpose of this paper is to determine why electors do not vote in municipal elections. The results will then be compared to those of Boswell in St. John's and the Pammett and Leduc studies of why so few people vote in federal elections. Unfortunately there are no provincial studies to make comparisons with.
To determine why electors do not exercise their right to vote, we surveyed voters in the City of St. Catharines, a typical Ontario city of 130,000 population, located in the Niagara Region of Ontario. The survey took place in January 2004, shortly after the November 2003 municipal election. Respondents were chosen at random from the local telephone directory. The sample size of 455 produces a level of accuracy of plus or minus five percentage points, nineteen times out of twenty. University students trained in interview techniques conducted the interviews.
In Ontario, elections for mayor and councillors are held every three years. St. Catharines, like virtually all other municipalities, does not have a party system and like other large municipalities has a ward system. The mayor is elected at large. Voter turnout in municipal elections has decreased over the years and is now in the 30 percent vicinity.
To enable us to make comparisons with studies of federal voting patterns, we adapted the questions from the Pammett and Leduc survey of the Canadian public as to why they did not vote in recent federal elections. Respondents were also asked to give reasons as to why others did not vote and in cases where they did not vote, their reason for not doing so. The socio-demographic characteristics of the respondents were also identified.
According to the official records of the returning officer, the turnout at the election was 30%. Normally, such a low turnout would not be surprising. However, given ideal weather conditions and a hotly contested three way mayoralty election with a controversial $16 million arena issue, this low turnout was inconsistent with conventional wisdom, supported by a voter survey by the Canadian Federation of Mayors and Municipalities, suggesting a much higher turnout rate. The low turnout may have resulted from voter fatigue arising from a recent provincial election, just one month prior to the municipal election.
One of the most significant findings of the survey was that 63% of the respondents indicated that they had voted, whereas the actual turnout was only 30%. These results are similar to the St. John's survey cited above where 56% of survey respondents indicated that they had voted, but the official turnout was 44%. These studies are the only two that we are aware of that compare the actual turnout rate to the reported rate. Virtually all other studies merely acknowledge the possibility of such a discrepancy and do not consider the impact of the discrepancy on their results.
One possible reason for the discrepancy between actual voting and reported voting is a non-random sample. We therefore checked our sample in terms of recent Census data and found the sample demographics to be representative of the population of the City. Therefore, the only other explanation is that non-voters either have poor memories or are reluctant to admit to non-voting. The poor memory issue could have been heightened in this case because of a provincial election about a month before the municipal election. The significance of this distortion is that the sample of those who admitted to non-voting is reliable bur the voting sample is contaminated with non-voters. Thus, if we compare the reasons cited by stated voters with the admitted non-voters, any differences that exist would be even greater had some non-voters not been classified as voters.
With respect to voting patterns by different groups, the question arises as to whether the discrepancy between those who claimed to have voted and those who actually voted distorts the results. For example, one age group could very well have inflated their numbers but there is no reason to assume that memory loss or misrepresentation is more common to one age group than another. In other words, memory loss and misrepresentation is assumed to be random with respect to age and therefore the voting numbers, although inflated, are still meaningful in terms of voting turnout by age categories. The same would be true for other voting groups.
Reasons Cited for Not Voting
Respondents were asked an open-ended question as to why they felt that others had not voted. The responses were recorded and classified according to the reasons cited in the Pammett and LeDuc federal study. The responses are summarized in Table 1. Polling station problems included not being able to get to the polling station and not knowing when and where to vote.
As Table 1 indicates, the most frequent reason assigned to others for not voting is apathy. Other frequently cited reasons are lack of information about candidates and issues, vote meaningless, laziness, and poor candidates. The Table also allows us to compare the responses of voters versus non-voters. The most significant difference was that voters attributed low turnout to the election being meaningless to a much greater extent than non-voters. Another major difference was that voters considered non-voters to be lazy whereas non-voters were less critical of others for not voting and instead cited lack of information about candidates and issues and candidates who were not appealing and not to be trusted.
Non-voters were then asked why they personally did not vote. As Table 2 indicates, laziness was not cited by anyone. Instead, non-voters considered themselves to be too busy and cited polling station problems and candidates who were not to be trusted. These results are similar to the St. John's results. However, the St. John's non-voters manifested considerable less negativity toward candidates than did the St. Catharines non-voters. Conversely, St. Catharines voters were much less likely to refrain from voting for personal or administrative reasons.
Table 3 compares municipal responses with the federal responses. To make the comparison, we combine the reasons cited by respondents in Table 1 in the municipal survey into three major groupings: 1) lack of interest due to apathy and the meaningless of the election; 2) negativity because of poor candidates and lack of issues; and 3) personal/administrative reasons such as too busy, out of town and polling station problems. For federal responses, the reasons for not voting are almost equally divided between lack of interest, negativity and personal/administrative. In the case of municipal voters, lack of interest is by far the most cited reason (51.6%) followed by negativity (35.8%) and personal/administrative reasons. It is interesting to note that federal candidates are not perceived to be of higher quality or more trustworthy than municipal candidates.
(i) Relationship Between Not Voting and Age
As indicated in Table 4, age and voting participation are positively correlated; seniors are the highest voting group (79.0%) followed closely by middle-aged voters (70.6%), followed by a significant drop for the 25-44 age group (53.3%) and an even greater drop for the 18-24 category (26.3%). A comparison of our results with federal results indicates a similar pattern although the age group differences at the municipal level are less extreme. The St. Catharines pattern is similar to that reported in St. John's; both studies indicated that people under 25 were the least likely to vote, although in St. John's voting peaked among the 45-55 age group.
Given the different voting patterns by the age groups, we then compared the reasons given by the age groups. For the sake of brevity, we do not present the results in this paper. Senior citizens cite illness much more often than the younger age groups whereas too busy is the most cited reason by the younger age groups. Unfortunately, the sample size of the various age categories was insufficient to permit meaningful comparisons to be made.
(ii) Relationship Between Not Voting and Education
A commonly held view is that better educated people are better informed and therefore more likely to vote. Lambert, Curtis, Kay and Brown (1988) using data from the 1984 federal election, measured two types of political knowledge, factual (knowledge of provincial premiers) and conceptual (ability to interpret left and right), and found both to be associated with higher levels of education. If "political knowledge can be thought of as an important precursor of political action, such as voting" (p. 360), then we would expect both education and community involvement to be positively correlated with voting turnout.
Table 5 presents the voting turnout classified by the respondents' level of formal education. Contrary to expectations, those with more formal education did not have higher voting patterns. Although there was very little difference between groups, those with the least formal education had the highest voting turnout. One possible reason might be that upper income groups were more forthcoming in admitting to not voting. Another possible explanation for this counter-intuitive result might have been that the controversial arena issue was of less interest to the higher educated groups, and so did not serve as a motivator for them to vote.
Given the different voting patterns by the education groups, we then compared the reasons given by the groups. Once again, the sample size was insufficient to make meaningful comparisons.
(iii) Relationship Between Not Voting and Family Income
Table 6 presents the voting turnout for different income groups, The results indicate higher voting patterns by those having higher family incomes. These results are similar to the federal results.
Given the different voting patterns by different income groups, we then compared the reasons given by the groups. Once again, the sample size was insufficient to make meaningful comparisons. Two observations, however, could be made: lower income groups cited polling station problems as the major reason for not voting whereas upper income groups cited that they were too busy as the major reason for not voting.
iv) Relationship Between Not Voting and Property Ownership
We also examined the relationship between voter turnout and whether the elector was a renter or homeowner. The results indicate that homeowners were more likely to vote (70.7%) than were renters (51.1%). Once again, these results are similar to those in the St. John's study. These results are not unexpected given that homeowners are more concerned with property taxes than renters. In addition, homeowners may on average have lived in the community longer and are more aware of local issues whereas renters may be more transient.
Given the different voting patterns by renters and homeowners, we then compared the reasons given by the groups. Once again, the sample size for the most part was insufficient to make meaningful comparisons, The only meaningful difference was that renters cited polling station problems whereas homeowners did not.
(v) Relationship Between Community Involvement and Voting Recent research by Henry Milner (2001) examines the theoretical and empirical relationship between political information and voter turnout. In the analysis of international differences in voter turnout, Milner concludes that countries that have the most politically knowledgeable voters have the highest voting turnout and that, when information levels are high, individuals are more likely to vote. A similar finding by Bean (1989) concluded that voter turnout is also related to community involvement which is not surprising since one would assume those active in the community would be among the most knowledgeable. Empirical research by Junn (1995) found such a relationship between political activity and knowledge of local officials. Boswell (1991) found that voters were more likely to have attended a public meeting or have contacted City Hall.
To examine the relationship between community involvement and voter turnout, we asked the respondents if they had participated in any local activity and if they had participated in any local group or club. The results are summarized in Table 8. In both cases those who were either politically or community involved had a higher turnout than those who were not.
Given the different voting patterns by activists and non-activists, we then compared the reasons given by the groups. The only significant difference was that those not active in the community cited "candidates can't be trusted" as the major reason for not voting whereas community activists did not. When political activists did not vote they cited too busy as a reason more often than those politically inactive.
(vi) Relationship Between Place of Birth and Voting
As indicated in Table 9, voting turnout was higher for non-Canadian-born voters than for Canadian born voters. Thus it appears that non-Canadian-born electors take their democratic responsibilities at the local level more seriously than Canadian-born voters. In terms of reasons for not voting, Canadian-born voters cite polling station problems and being too busy as reasons whereas non-Canadian-born voters are more apt to cite candidates can't be trusted as their reason. These results are in contrast to the federal ones where those born in Canada have a higher voting frequency.
(vii) Relationship Between Years in Community and Voting
As Table 10 indicates there is a positive correlation between voting and number of years individuals lived in the community. This also corresponds to Boswell's (1991) finding that people who had lived in St. John's for their entire lives were more likely to vote in the municipal election. The results are not surprising since one would expect newcomers to be less informed and less involved in the community. Furthermore, years lived in the community might also be a function of age where, as previously reported, there is a correlation between voting and age in which case one would also expect a correlation between voting and years lived in the community.
Given the different voting patterns by residents, we then compared the reasons given by the groups for not voting. The only significant difference was that polling station problems were cited more often by those who lived in the community less than two years.
This paper reports the results of a telephone survey examining why electors did not exercise their right to vote at the municipal level, The first significant finding is the large discrepancy between the actual voter turnout and the reported turn out. Researchers involved in voter surveys have always suspected voter self-reporting. However in our case, the recorded voting was twice the number that actually voted. The non-voter sample is genuine since respondents would have no bias to report that they did not vote. The same cannot be said for the voting sample. Researchers should therefore be careful in their interpretation of the voting sample.
When electors were asked why people don't vote, the most frequent reasons assigned to others for not voting were apathy, lack of information, the meaninglessness of the vote, and electors being lazy. When non-voters were asked why they didn't vote, the most frequent reasons were that they were too busy, polling station problems and non-trustworthy candidates. Laziness was never mentioned. Thus, non-voters were kinder to themselves than were voters when asked why the turnout was low. At the federal level, the reasons for not voting are almost equally divided between lack of interest, negativity and personal/administrative. For municipalities, however, lack of interest is by far the most cited reason for not voting. It is interesting to note that federal candidates are not perceived to be of higher quality or more trustworthy than municipal candidates.
We also found voter turnout to be a function of age, income, home ownership, community and political involvement but not the level of education. These results are similar to those in a recent survey of federal electors. A significant difference however was the influence of the number of years lived in Canada. The federal study found that "being new to Canada, as measured by whether respondents were born in this country or not, is associated with lower voter turnout" (Pammett and LeDuc 2003, 28). In our case, we found that those not born in Canada had a higher voter turnout than native-born Canadians. Another difference was that education was found to be the most important predictor of voting in the next federal election but not in the municipal election.
Given that municipal politics is "closest to the people," why then is the voter turnout less at the local level than at the provincial or federal level? One possible explanation is the more intense and longer media coverage of federal elections compared to the weaker coverage of local elections. A second possible explanation is the large amounts spent on federal and provincial campaigns compared to the minimal amounts spent on local elections where donations are not tax-deductible. Another possible explanation is that federal and provincial issues are of more importance to electors. The lack of political parties in municipal elections may also be a reason. In the United States, cities where national parties compete have higher voter turnouts than cities without parties. National parties are able to mobilize voters more effectively and to simplify the choices voters face (Peterson 1981).
Given the reasons cited for not voting, what type of efforts can be made to increase the turnout? The recent federal study by Pammett and LeDuc (2003, p. 73) concludes "that improvements in education and information to prospective voters are the best methods of interesting young people in politics and elections." Consistent with this recommendation is Milner's (2001) review of the literature that concludes that civic literacy increases civic engagement and voter turnout. The review also cites research that high levels of newspaper reading contribute to civic engagement, whereas high levels of television consumption have the opposite effect (Pumam 1995). Thus the recent decline in newspaper circulation does not augur well for voter turnout.
Campaigns to improve civic education and to increase information to the electorate, be it in the form of posters, flyers or advertising, are common. Targeting young people is less common. However, a program to improve civic engagement among young people, "Youth Vote 2003," a variation of the successful "Kids Voting USA" program, was recently introduced in Ontario's schools (Kids Voting Canada 2005; Kids Voting USA 2005). A parallel election campaign is run for students with the hope that students, by voting on the same day as their parents, will encourage their parents to vote. At the federal level, a youth text-messaging program for 18-24 year olds has also been implemented.
Several proposals to improve the convenience of voting are under consideration. For example, in California, the rescheduling of local elections to coincide with state and national elections was found to be effective in increasing turnout (Zoltan, Lewis, and Louch 2002). In Ireland, rescheduling the voting day to weekends was also found to increase turnout. In Canada, voting via the internet or telephone has been implemented in several smaller municipalities. However, the high cost of security software to prevent fraud and hacking appears to be a deterrent to its widespread usage.
With respect to future turnout, the trend is not clear. Although seniors tend to have high participation rates, it does not necessarily follow that, as the population ages, turnout will increase. It may very well be that the young who are not voting now will become non-voting seniors in which case there could be both a senior non-voting demographic in addition to the traditional young non-voting group. If this indeed occurs, one could very well expect proposals to introduce compulsory voting similar to that in Belgium and Australia. However, to do so would be very difficult given that a majority of Canadians are opposed to compulsory voting (Pammett and LeDuc 2003).
Bakvis, Herman, ed. 1991. Voter Turnout in Canada. Toronto: Dundurn.
Bean, Clive. 1989. Orthodox Political Participation in Australia. Australia and New Zealand Journal of Sociology 25 (3): 451-479.
Boswell, Peter. 1991. Municipal Non-Voters: An Analysis of the 1990 City of St. John's Election. Paper presented at the Annual Meeting of the Canadian Political Science Association, Kingston, Ont., June 2.
Canadian Federation of Mayors and Municipalities. 1967. Survey of Municipal Voting in Fourteen Canadian Urban Centres of 100,000 of More Population (May).
Junn, Jane. 1995. Participation in Liberal Democracy, What Citizens Learn from Political Activity. 1995 Meeting of the American Political Science Association. New York, N.Y.
Kids Voting Canada. 2005. http://www.kidsvotingcanada.com. (Accessed April 7, 2005).
Kids Voting USA. 2005. http://www.kidsvotingusa.org. (Accessed April 7, 2005).
Lambert, R.D, J.E. Curtis, B.J. Kay and S.D. Brown. 1988. The Social Sources of Political Knowledge. Canadian Journal of Political Science 21: 359-75.
Milner, Henry. 2001. Civic Literacy: How Informed Citizens Make Democracy Work. Hanover, NH: University Press of New England.
Pammett, Jon H. and Lawrence LeDuc. 2003. Explaining the Turnout Decline in Canadian Federal Elections: A New Survey of Non-voters (Match). Elections Canada.
Peterson, Paul. 1981. City Limits. Chicago: University of Chicago Press.
Putnam, Robert. 1995. Bowling Alone: America's Declining Social Capital. Journal of Democracy 6: 65-78.
Zoltan, Haznal, Paul Lewis and Hugh Louch. 2002. Municipal Elections in California: Turnout, Timing and Competition. Public Policy Institute of California.
Department of Economics
Councillor, City of St. Catharines
Department of Political Science
Dean of Social Sciences
Table 1: Electors' Opinions as to Why Voter Turnout was Low Reasons Percentage of Respondents Voters Non-voters Electors are lazy 11.6 13.0 9.2 Not interested; didn't care; apathy 26.1 26.1 26.2 Vote meaningless; doesn't count, 13.9 17.4 7.8 election foregone Lack of information about 16.5 15.4 18.4 candidates and issues No appealing candidates 9.9 9.5 10.6 Candidates can't be trusted 9.4 7.9 11.3 Too busy with work/school/family 6.3 4.3 9.9 Illness 0.0 0.0 0.7 Polling station problems 3.5 3.6 3.5 Other--unclear, unclassifiable 2.5 2.8 2.1 Sample size 395 254 141 Table 2: Non-voters' Reasons for Not Voting Percentage Reasons of Non-voters Lazy 0.0 Not interested; didn't care; apathy 3.9 Vote meaningless; doesn't count, election 2.0 foregone Lack of information about candidates and issues 2.0 No appealing candidates 2.0 Candidates can't be trusted 12.7 Too busy with work/school/family 28.4 Illness 6.9 Polling station problems 18.6 Other--unclear, unclassifiable 23.5 Sample size 102 Table 3: Comparison of Reasons Cited for Not Voting in Federal Elections and Municipal Elections Reasons Percentage of Respondents Level Federal Level Municipal Lack of interest Not interested; didn't care; apathy 37.2 51.6 Vote meaningless; doesn't count; results foregone Forgot; unaware Negativity Poor candidates 34.4 35.8 Candidates not to be trusted No candidate information Personal/Administrative Too busy 37.3 9.8 Illness Polling station problems Other Religious 3.9 2.5 Unclassifiable reasons Table 4: Voting Turnout for Different Age Groups Age group Percentage Voting Turnout 18-24 26.3 25-44 53.3 45-64 70.6 65+ 79.0 Sample size 452 Chi-square level of significance .000 Table 5: Voting Turnout for Different Education Groups Level of Education Percentage Voting Turnout Primary 70.0 Secondary 64.0 University or college 62.3 Postgraduate 62.5 Sample size 454 Chi-square level of significance .871 Table 6: Voting Turnout for Different Income Levels Level of Income Percentage Voting Turnout Under $30,000 57.7 $30,000 to $50,000 60.0 $50,000 to $100,000 62.6 Over $100,000 69.7 Sample sie 342 Chi-square level of significance .666 Table 7: Voting Turnout for Renters and Home Owners Tenancy Percentage Voting Turnout Renters 51.1 Home Owners 70.7 Sample size 444 Chi-square level of significance .000 Table 8: Political Activity, Community Involvement and Voting Panel A: Political Activity Percent Voting Turnout Politically involved 75.2 No political involvement 57.7 Sample size 464 Chi-square level of significance .001 Panel B: Community Involvement Percent Voting Turnout Community involved 70.1 No community involvement 60.1 Sample size 463 Chi-square level of significance .039 Table 9: Voting Turnout for Canadian Born and Non-Canadian Born Voters Place of Birth Percent Voting Turnout Canada 62.0 Non-Canada 67.7 Sample size 462 Chi-square level of significance .304 Table 10: Years in Community and Voting Years in Community Percent Voting Turnout Less than 2 years 40.7 2-10 years 59.0 11-25 years 63.4 More than 25 years 67.5 Sample size 466 Chi-square level of significance .000
|Printer friendly Cite/link Email Feedback|
|Author:||Kushner, Joseph; Siegel, David|
|Publication:||Canadian Journal of Urban Research|
|Date:||Jan 1, 2006|
|Previous Article:||The NIMBY syndrome and the health of communities.|
|Next Article:||Characteristics of university student housing and implications for urban development in mid-sized cities.|