open Secondary menu

Youth Engagement and Mobilization in the 2010 Toronto Municipal Election

6. Recommendations

This section includes two sets of recommendations: suggestions for further research in the field of youth electoral engagement, as well as best practices for youth mobilization initiatives. Although these recommendations are framed in the context of youth electoral engagement, many of them can be applied to voters of all ages. Conducting field experiments, collecting turnout data, providing election information and increasing accessibility are relevant to the electoral engagement of all Canadians, particularly in an era of continually declining turnout.

6.1 Research Recommendations

I. Election agencies should commission field experiments that use actual turnout records to study youth mobilization.

There is a need for field experiments that use actual turnout records to directly measure and compare the effectiveness of different youth mobilization strategies. Additional survey research into youth engagement is also needed, particularly in Canada (see next recommendation). However, when the dependent variable is direct participation in an election, there is no substitute for real-world turnout data.27

With proper implementation and data collection, there are a limitless number of turnout field experiments that can be conducted. Appendix F describes three experimental protocols that could serve as starting points for a turnout research program in Canada:

  • Testing different methods of delivering letters of attestation to students living in campus residences
  • Holding festivals at polling stations on election day
  • Variations of traditional door-to-door canvassing for young voters

II. Election agencies should commission and conduct additional research into the lifestyles and attitudes of Canadian youth.

The survey results presented in this report show the significant impact of life-cycle effects and residential patterns on engagement. Other studies have shown the large impacts of the transition to adulthood on turnout, sometimes on the order of 20 percentage points (Bhatti and Hansen 2010). These effects warrant further attention, particularly because the first few years of eligibility may be a critical period in the formation of voting behaviours (Johnston, Matthews and Bittner 2007).

II.1 Conduct research with large, randomly-selected samples of youth. Many of the findings described above could not have been identified using smaller samples or less detailed indicators. Large, representative samples allow for finer distinctions between subsamples based on age, place of residence, etc., while securing results that accurately represent the population as a whole.

II.2 Gather longitudinal data about attitudes among youth. Studies such as this one use age as a proxy, comparing younger respondents to older ones to assess the impact of life-cycle changes. However, by observing the same group of respondents over time, researchers can more precisely link changes in attitude to changes in behaviour.

II.3 Gather more information about the whereabouts of youth. Several key findings in this report involve the residential patterns of youth in Toronto. In order to better target youth during elections, be it for mobilization campaigns or the distribution of VICs, more information is needed about where young Canadians live and spend their time. Residential patterns differ from city to city, as well as between urban and rural communities. Understanding these patterns is crucial for any initiative that seeks to reach and mobilize youth.

II.4 Gather more information about immigrant communities. Two interview participants described issues surrounding citizenship as barriers to youth participation. More information about these communities is needed in order to properly assess these barriers and determine how they can be addressed.

III. Election agencies should collect consistent, age-segmented turnout data for all Canadian elections.

This research could have had a broader scope and more robust findings – without any substantial change in the budget or labour required – if consistent, age-segmented turnout data had been available. Patterns of youth engagement in different areas of the city, as well as the impact of mobilization activities and campaigns, could be analyzed in far greater detail with real-world turnout information.

Canada has an abundance of elections. In 2011, there will be six provincial or territorial elections, dozens of municipal elections in another three provinces, and Band Council elections across the country. Unfortunately, most provincial and municipal election authorities do not record turnout by age. What little data they do collect often lacks detail and cannot be compared to other jurisdictions due to methodological inconsistencies.

Consistently collecting turnout data has several advantages. One unintended consequence of a federal system is that it creates natural experiments: election agencies across Canada operate in a variety of contexts and under a variety of regulations. Consistent turnout data would open the door to countless comparative analyses. Researchers could analyze the influence of many factors on youth turnout, including registration systems, ID requirements, polling dates, ballot types, on-line voting, electoral competitiveness and more.

Elections Canada has already developed a secure and reliable procedure for gathering age-segmented turnout data during Canadian federal elections (Elections Canada 2010). The implementation of similar procedures for other elections would be a boon to the study of turnout in Canada and provide an invaluable source of longitudinal data about young voters.

6.2 Electoral Engagement Recommendations

IV. Election agencies should plan and publicize their youth strategies earlier.

The youth mobilization conducted during the 2010 Toronto municipal election was a remarkable achievement. In the space of about three months, Toronto Elections brought together a network of youth partners from across the city, including many organizations that had never before been involved in electoral mobilization. Given the overwhelmingly positive response from interview participants, election agencies should continue this approach.

However, interview participants repeatedly said that their organizations could have been more active if they had been given more time to prepare. Election agencies should ensure that other stakeholders are aware of their youth strategies and have sufficient time to coordinate with them.

V. Stakeholders should adopt mobilization strategies that target unengaged youth.

The survey findings in this report suggest that unsolicited forms of personal contact mobilize youth to vote. This reflects similar findings in GOTV research, as well as the qualitative input of the interview participants. These findings suggest that organizations seeking to engage youth in elections should reconsider their mobilization strategies.

V.1 Election debates are not a highly effective mobilization tool. They may further other policy or educational goals, but the people who attend debates are also more likely to already be engaged. Debates also draw relatively small crowds given the relative commitment of time and resources.

V.2 For organizations seeking to mobilize youth, traditional activities such as canvassing may be more effective. For those still wishing to host events, integrating the election into an event that appeals to unengaged youth (e.g. a music event or community festival) may be a more promising approach.

V.3 Narrower mobilization strategies appear to be more effective. Our survey findings show significant differences between recent adults and young independents in terms of lifestyle, engagement, and residential patterns. Furthermore, youth who either live with a parent or have moved to the city centre report higher levels of civic and electoral engagement.

A one-size-fits-all approach to youth mobilization is unlikely to be effective at contacting these diverse groups. Youth who live independently in the city centre, for example, will have different needs and lifestyles than those living independently in the amalgamated suburbs (with the latter group being less engaged on the whole). Furthermore, the organizations that were most effective at mobilization in the 2010 municipal election focused on specific communities. Youth mobilization initiatives that are less ambitious and tailored for specific sub-populations may prove to be better investments.

These findings are relevant to all election stakeholders; the targeted strategies described above can just as easily be integrated into partisan campaigns as non-partisan ones, and they are also relevant when mobilizing groups other than youth.

VI. Election agencies should support non-partisan youth mobilization networks that convene, coordinate, and educate stakeholders.

When asked to provide feedback for election agencies, 10 interview participants recommended more on-the-ground mobilization. Others asked for more guidance and leadership from Toronto Elections. Given that most of these organizations do not have election-specific mandates, it is unreasonable to expect them to spontaneously develop large and highly effective electoral mobilization campaigns. They need an outside source of expertise, coordination and support.

The activities election agencies are willing to undertake are limited. As described in the interview findings, these limitations had a negative impact on Toronto Elections' ability to mobilize youth in the 2010 municipal election. If election agencies are sincere in their commitment to increase youth turnout in Canada, they need to collaborate with other youth stakeholders and support more effective mobilization initiatives.

Below, we recommend steps to be taken in order to provide the necessary expertise, coordination and support through a non-partisan youth mobilization network. This plan builds on Toronto Elections' recent youth initiative by creating a more structured network that is better able to meet the needs of youth and community organizations. Election agencies should support the creation of one of these networks in an upcoming election to serve as a test and proof of concept.

VI.1 Help youth and community organizations implement effective youth mobilization strategies. When asked for a subjective evaluation of the impact of their work during the 2010 municipal election, most interview participants vacillated. They described it as unknown or impossible to assess. This uncertainty suggests a general lack of confidence among organizations in their youth mobilization plans.

The primary role of a youth mobilization network would be to help organizations develop and deploy effective mobilization campaigns. Political operatives and social scientists may think extensively about turnout, but youth and community organizations do not. They are unlikely to be familiar with the research into voter mobilization strategies and their effectiveness. A youth mobilization network can share this specialized knowledge with organizations and help them apply it.

VI.2 Coordinate the work of youth and community organizations. Interview participants repeatedly cited collaboration as a best practice in their work, and a lack of preparation time as a barrier to being more active during the election. These organizational challenges affected the Toronto Elections partner network even though the date of the 2010 municipal election was fixed years in advance. They can only be aggravated for elections that occur on short notice.

A youth mobilization network is a long-term initiative. It would build trust and maintain links with partner organizations before and after the election campaign, rather than repeating the process of outreach, relationship-building and coordination each time an election is held. It falls outside the traditional mandate of election agencies, but a stable network is a more effective way to engage stakeholders from the community and non-profit sectors.

The coordination of non-partisan mobilization through a network has additional benefits. First, it helps to reduce duplication of effort among the members of the network. Organizations can share resources they have developed, collaborate, or find partners to host events and plan activities. This is precisely the type of behaviour that occurred within the Toronto Elections network, and it should be supported.

Convening a network also provides a way for member organizations to give feedback to and start a dialogue with election agencies. Many of the findings and comments contained in this report could have been secured through a conversation between election agencies and interview participants. However, for this type of conversation to happen, there must be a relationship with mutual trust.

VI.3 Expand knowledge of youth mobilization among all stakeholders. As identified in the research recommendations, there is a need for more real-world data about how to engage youth in elections. Youth mobilization networks should gather this information and disseminate it to other electoral stakeholders. For example, a coordinated network could easily implement any of the experimental designs discussed in Appendix F.

There are limits to the systemic impact of non-partisan mobilization initiatives. Elections are ultimately about partisan competition. Political parties and candidates will always be the primary source of mobilization in elections. Unfortunately, partisan organizers are not always far-sighted: their priority is to gain a relative advantage over opponents in the current election. Although youth are the largest untapped group of voters in Canada today, the conventional wisdom is that campaigns should focus on persuading known voters rather than reaching out to new ones.

However, the self-interest of parties and candidates can also be harnessed to mobilize non-voters. By testing and refining cost-effective youth mobilization strategies, other stakeholders can impact the system as a whole. As Green and Gerber (2008) have articulated, political campaigns have an interest in this type of information:

Our perspective on how to raise voter turnout is rather different. Examine a range of GOTV tactics and figure out which ones are effective and cost-efficient. By demonstrating what works (and what does not), this investigative approach provides an important signal to those engaged in electoral competition. If the market for campaign services learns from a reliable source that a particular GOTV tactic is a more cost-effective way of garnering votes, we eventually will see campaigns allocate more resources to this tactic.

Even with the limitations of our survey dataset, this research found that certain forms of contact from a candidate made youth more likely to vote. Supporting the creation of a pilot youth mobilization network will produce more comprehensive and robust research.

VII. Election agencies should improve youth coverage in the voters' list.

Youth are less likely to receive a voter information card (VIC). An Ipsos Reid survey commissioned by the City of Toronto for the 2010 election found that 65% of Torontonians recalled receiving a VIC at their current address, whereas 26% responded that they did not receive a card (Ipsos Reid 2010). That compares with 48% and 47%, respectively, in our sample. Given that these cards serve as a basic source of information about the election for voters, this relatively low coverage of youth is troubling.

Our findings show that youth who live independently, reside in the city centre, or changed addresses recently are less likely to receive a VIC. To supplement the existing process for municipal elections, targeted enumeration could be conducted in communities that have younger and more mobile populations. The success of any efforts to engage or inform potential voters hinges on a reliable knowledge of that population.

VIII. Non-partisan stakeholders should clearly define non-partisanship policies and their implications.

The interview process revealed a significant amount of confusion and concern surrounding non-partisanship policies during elections. Further attention to this topic is certainly warranted, as it was never part of this project's research design and its impact may be underplayed.

VIII.1 Stakeholders in elections should clarify their definitions of non-partisanship. For example, what makes an election event non-partisan? Do all candidates have to be present? Or should no candidates be present? Clearly, there is no litmus test for non-partisanship. However, organizations that rely on government support would benefit from clearer guidelines and articulations of policy in this area, particularly when concerns surrounding funding are at play.

Discussion: Managing risks related to non-partisanship

Several participants identified potential negative consequences associated with the perceptions of youth mobilization. As this report includes a number of recommendations concerning youth mobilization, a discussion of how to mitigate these risks is appropriate.

These risks are to some extent inevitable. By nature, every election produces both winners and losers. Commentators and interested parties attribute these results to a variety of causes: the weather on election day, the tone of media coverage, etc. Initiatives that mobilize youth, and particularly those that prove effective, are not exempt from this type of commentary and speculation.

The simplest way for any youth mobilization initiative to mitigate negative perceptions is to be both non-partisan and transparent. For example, the youth mobilization network proposed earlier should not be linked to any partisan stakeholders, be it through membership, funding, or other affiliations. It should operate independently, with member organizations formally committing to non-partisanship in their electoral activity.

Similarly, any organization involved in non-partisan youth mobilization should publicly and proactively disclose their membership, strategy, and plans for the election. This provides others with the opportunity to raise any concerns before activities take place.

These steps can help to address perceptions of partisanship or impropriety when mobilizing youth. Beyond that, any negative response is tied to youth engagement itself. However, if any election stakeholders genuinely disagree with the notion that more young Canadians should vote, surely the onus is on them to show why election agencies and other organizations should share that view.

IX. Election agencies should provide adaptable and redistributable election information.

Fully half of the interview participants described taking on-line resources from Toronto Elections and modifying them to suit their organizations and campaigns. Election agencies should encourage this adaptation of resources by election stakeholders. There are two immediate steps that can be taken in this direction.

IX.1 Public information about elections should be provided with a Creative Commons or comparable license. This would encourage others to adapt the information without fear of infringing on any copyrights, while still requiring them to properly attribute the source of the information. Given that the practice is already widespread, this would serve primarily as a symbolic gesture.

IX.2 Public information about elections should be provided in open and editable formats. This eases the actual process of adaptation and sharing by removing the inconvenience of extracting information from a "closed" resource.28 Similarly, public election information should be provided on-line whenever possible – a problem that was identified by debate organizers in the 2010 Toronto municipal election.

There is a perceived risk in using open formats, as the information provided may be reproduced inaccurately. However, the practice of third-party adaptation and dissemination is already widespread with closed formats. It is also unpreventable: in the Internet era, election agencies cannot monopolize information about voting. By responding to this practice, rather than ignoring it, election agencies can engage with those who are adapting information and secure their cooperation in reproducing it accurately and attributing it properly.

X. Stakeholders should provide information about candidates and their platforms.

If unsolicited feedback indicates a clear need, then this recommendation deserves special attention. Interview participants repeatedly identified a need for more information about candidates and their platforms. They also identified this as a challenge of their work, as they were often unable to answer questions about candidates in the election.

Information about candidates, parties and platforms should be compiled and distributed as a public service. This information should come from a reputable, non-partisan source. Political parties and candidates, for obvious reasons, tend to provide skewed descriptions of their platforms. Yet in order for this information to be trusted by those involved in non-partisan mobilization, it must be compiled by an organization with no interest in the outcome of the election.

Unfortunately, providing this information is likely to be deemed too risky by election agencies. However, a number of organizations, including Apathy is Boring, already develop platform summaries on an ad hoc basis. Similarly, on-line voting compass projects ask voters to identify their priorities and, based on information compiled by experts in the field, provide analyses of how an individual's opinions relate to the positions of different candidates. These non-partisan initiatives already exist, but they require consistent external support to expand the scope and improve the quality of this work.

XI. Election agencies and other stakeholders should increase voting accessibility.

XI.1 Polling stations should be open longer during municipal elections. Interview participants from the amalgamated suburbs raised concerns with polling station hours for the 2010 Toronto municipal election. This was a significant issue in their communities, as most residents commute to work or school, leaving only a narrow (and therefore crowded) window in the evening for them to vote.

XI.2 Students living in campus residences should receive letters of attestation.During the 2010 municipal election, the City Clerk's Office cooperated with several university administrations and student unions to provide these letters, as many students in university residences lacked another proof of address. Interview participants responded positively to this program, and we recommend that it be continued in future elections. Distributing these letters also provides an opportunity to mobilize students, which requires further discussion and coordination between all stakeholders, including college or university administrations and student unions.


27 Researchers in other jurisdictions, such as the United States and United Kingdom, are also able to use validated voting records in their experimental research. Because these records report turnout at the individual level, they make it much easier to study the behaviour of specific groups such as young voters. Similar data is not currently available in Canada.

28 A useful rule of thumb is to never provide a public resource as a PDF. If the goal is to share and diffuse the information in a document, it should be provided in an editable format.