Secondary menu

Explaining the Turnout Decline in Canadian Federal Elections: A New Survey of Non-voters


This report explores the major findings of a survey of Canadians carried out in April 2002. It was designed by the authors in co-operation with Elections Canada and conducted by Decima Research. The sample design called for a short screening interview with a large number of Canadians (5 637) and a longer interview to be continued with 960 reported voters in the 2000 federal general election and 960 reported non-voters in that election.1 In this way, interviews were obtained with a much larger group of non-voters than is customary in election-related surveys of the Canadian public. This allowed the examination of their reasons for not voting in considerable detail, and more confidence to be placed in the accuracy of the results than is possible when smaller numbers of respondents are involved.

The survey was designed to measure a wide variety of explanations for not voting, both in general and in reference to the increase in not voting that has occurred in the last three federal elections.

After a brief introduction illustrating the phenomenon of turnout decline, this report examines the reasons survey respondents gave for not going to the polls in the 2000 general election. The report then examines a series of correlates of not voting, starting with socio-demographic factors. Here, and throughout the report, particular attention is paid to the factor of age. In conjunction with socio-demographics, a number of important attitudinal and behavioural factors in voting/not voting are examined in succeeding sections: civic duty and political interest; efficacy of the vote; feelings of community and participation (commonly referred to as "social capital"); and attitudes towards electoral reform. Finally, three topics are investigated at greater length. The first is the political attitudes of youth, including the ways in which young people might be encouraged to take a greater interest in politics. The second is the possible improvement in the voting rate that might be produced by use of the Internet for electoral registration and/or for voting itself. Finally, the report offers a more detailed analysis of personal and administrative factors involved in not voting, including matters relating to the National Register of Electors.

A system of corrective weights was calculated for the dataset by Decima Research, and the weighted data is used for most tables in this report. The total weighted N for non-voters in the 2000 general election is 1 097.