Does the Introduction of Online Voting Create Diversity in Representation?

Abstract

The Internet’s effect on political communication is omnipresent. However, very few jurisdictions around the globe allow their citizens to cast their ballot online. What are the electoral consequences of this reform? Research, so far, has mainly looked at security considerations and effects on turnout. In this research note, we broaden the scope of prior studies by examining the effect of online voting on diversity in representation. Using the voting results of municipalities in the Canadian province of Ontario both before and after the implementation of online voting, we test whether this reform has increased the representation of women and youth. We do not find that Internet voting has any significant impact on which candidates are elected, with both the gender and age of elected mayors being constant across online and traditional elections. We further find that the number of woman candidates does not increase with online voting.

Keywords

online voting turnout elections Canada

Introduction

The Internet is omnipresent in people’s personal, economic and political lives. As a means of personal communication, messenger services such as Instagram, Twitter or Facebook have largely replaced traditional means of communication such as mail or the telephone (De Wet et al., 2016). In the business world, the Internet is the medium that primarily connects companies and clients (Castells, 2002). And in the political world, the Internet has become one of the most, if not the most, important medium to connect politicians to voters (Dahlgren, 2005; Sullivan and Bélanger, 2016). Beyond its communication function, the Internet can also be a means of conducting elections. Several jurisdictions have so far adopted online voting, including the national elections in Estonia, and the municipal elections in parts of Canada (Alvarez et al., 2009; Goodman and Stokes, 2020). Several more have conducted trials including Switzerland (Petitpas et al., 2021) and the United Kingdom (Norris, 2004).

The Internet as a means of conducting elections is gaining traction slowly, and in turn so is academic interest in the topic. One strand of research asks the question whether electronic voting is as secure as traditional means, looking both at the security of in-person voting machines (e.g. Wolchok et al., 2010) and remote online voting (e.g. Rubin, 2002). Another strand looks at political participation, posing the question whether e-voting increases turnout (Germann and Serdült, 2017; Solvak and Vassil, 2018). We add a third layer to previous research and examine whether the introduction of e-voting increases diversity in representation. Specifically, we examine the effect of online voting on the gender and age of elected mayors in municipal elections in the Canadian province of Ontario. We find no evidence that online voting increases the diversity of elected mayors; online and traditional elections, on average, elect mayors of similar age and gender profiles. We also test if online voting leads to more women candidates running compared to traditional elections and likewise find no evidence to support this.

This short article proceeds as follows: first, we situate our study within the budding literature on the electoral consequences of online voting. Second, we succinctly present our case, the municipal elections in Ontario. Next, we present our data and methodology. Finally, we display our descriptive and inferential results, followed by a conclusion.

Theory

Theoretically, the literature sees online voting as a potentially novel innovation that can make the electoral process more diverse and less skewed towards a certain demographic (i.e. middle-aged to senior men of the dominant ethnicity; for example, Brady et al., 1995). A simple rational choice model would predict that online voting would make the process of voting less costly. Voters could cast their ballot on their own terms with no need to going to the polling station and potentially need to wait in line (see Blais, 2000). As the costs of in-person voting are not distributed evenly across the population, lowering these costs could change the demographic composition of voters, and thereby influence electoral outcomes (Alvarez and Nagler, 2000). For example, many people with physical disabilities will incur far higher than average costs in physically going to a polling place to cast a ballot. In allowing them to easily vote from home, online voting decreases their cost of voting to a much greater extent than for a person without mobility restrictions (see, for example: Miller and Powell, 2016). A similar logic could increase the electoral participation of many demographics with higher costs of voting, including people who work long or unusual hours, people who cannot easily afford transportation to a polling station, and people with caregiving responsibilities. Given the greater caregiving responsibilities faced by many women, and the lower wages and greater reliance on public transit for both women and young people (Heisz and Schellenberg, 2004), online voting could also favour the participation of these two groups. In addition to reducing the cost of voting, online voting might be particularly appealing to younger electors (see Gabel et al., 2016), who are more likely to be familiar with the Internet and comfortable using it for everyday tasks.

By providing a means of casting a ballot that is more flexible, and possibly more appealing, than traditional pencil-and-paper voting, online voting might not only induce more voters to participate in an election, but these potential shifts in the voting demographic could possibly result in a corresponding increase in the diversity of candidates elected. Affinity voting, that is to say voting for a candidate with whom a voter shares identity traits such as gender, could be a major driver of votes (see Bird et al. (2016) for a study on a Canadian municipal election). According to Badas and Stauffer (2019), this particularly applies to elections in non-partisan contexts. For the Canadian context, this could imply that by increasing the proportion of younger and women voters, implementing online voting could increase the number of younger and women mayors elected by increasing affinity voting in favour of these groups.¹

H1. Elections with online voting will see more women elected than traditional elections.

H2. Elections with online voting will see the election of younger candidates in greater numbers than traditional elections.

Case and Context

In this article, we examine the effects of online voting on diversity in representation using the example of municipal elections in Ontario. In Ontario, voters typically elect one or more councillors representing their district in the municipality, as well as voting directly for the municipality’s mayor. Municipal elections in the province are also non-partisan, with candidates having no formal affiliations with either other candidates in the election or with federal or provincial parties. In addition, election campaigns in smaller municipalities are often poorly publicised, and often focused on rather idiosyncratic local issues, making categorising candidates on a left-right spectrum impractical for voters. As we have no way to systematically determine the partisan leanings of candidates, we limit our analysis of election results to the age and gender of the mayor elected. We specifically chose mayors and not councillors due to better availability of census data for entire cities as opposed to districts, as well as a greater number of acclaimed positions at the councillor level.

Two more methodological considerations make Ontario’s mayoral elections a good choice for analysis. First, Ontario represents a case with a large number of elections in similar societies conducted simultaneously, but with varying implementation of online voting. In fact, e-voting has been introduced incrementally in Ontario, with the first municipalities adopting it in 2003 (e.g. Peterborough). In later elections, more and more municipalities, but not all, have implemented it (see Table 1 for information on e-voting adoption in our sample). This allows us to compare the gender and the age of mayors in municipalities that have adopted e-voting against those that have not adopted this reform. Other jurisdictions such as the national elections in Estonia (where the whole country has adopted e-voting simultaneously) do not allow for such a comparison.

Table 1.

Elections Included in the Sample.

Year	No online voting	Online voting	Total
2000	131	0	131
2003	140	10	150
2006	140	18	158
2010	133	41	174
2014	85	98	183
2018	70	124	194
Total	699	291	990

Second, in Ontario, there is no clear pattern between municipalities that have adopted e-voting and those that have not adopted it, yet. Ontario’s largest cities, Toronto (pop. 2.5 million) and Ottawa (pop. 1 million) have not yet adopted online voting, but many small rural municipalities have stuck with paper ballots as well. In addition, Markham (pop. 300k), one of Ontario’s larger and most diverse cities, was among the very first cities in Canada to adopt online voting. Although not the subject of much study, the expressed motivations of municipalities to adopt online voting generally include increasing accessibility, simplifying elections, and lowering costs (see Goodman et al., 2010b). As the demographics of the two groups are reasonably similar (see Table 2), we judge the municipalities with and without online voting to be very easily comparable.

Table 2.

Demographics of Municipalities Using and Not Using Internet Voting in 2018.

	Internet voting		No Internet voting
	Median	Mean	Median	Mean
Population	10,109	24,179	11,292	113,737
Population density	28.8	250.59	41.45	453.11
Median age	46.6	47.2	43.7	44.82
Median household income	64,213	65,428	62,336	65,370
% with degree	11.83	12.59	10.58	13.32
% immigrants	7.9	9.66	9.24	14.22
Employment rate	58.6	57.99	59	57.61

As the first instances of online voting began in 2003, we include all municipal elections since 2000 (2000, 2003, 2006, 2010, 2014 and 2018). We begin in 2000 to allow at least one election without online voting for all municipalities, as a point of comparison. Within this period, we included all municipalities which provided or published data on the age and gender of mayors, whether or not the municipality eventually adopted online voting (see Table 1).

Data and Methods

Our first two main dependent variables are the gender of the mayor elected (measured as a binary male/female variable, with men coded 0 and women 1) and the age of the mayor elected (in years). We collected both dependent variables from the municipality in question, either through consulting published results or requesting the data directly from the municipal clerk. Our main independent variable is the availability of an online voting platform, coded 0 for no online voting and 1 for online voting available. While municipalities provided gender data for a total of 713 elections, our age data are more limited and available for only 320 elections.²

In our regression models, we also include a series of eight relevant control variables. We first control for acclaimed mayoral elections and hypothesise that acclaimed mayors are long sitting experienced incumbents, who do not draw contenders. Given women’s low levels of representation in municipal governments (Tolley, 2011), we expect such seasoned politicians to be senior men, which, in turn, should be unbeneficial for the election of women and youth (Schwindt-Bayer, 2005). We operationalise acclaimed elections through a dummy variable, coded 0 for contested mayoral elections, and 1 for uncontested mayoral elections. We further control for a series of seven demographic variables, obtained from the Canadian Census.

We control for population and for population density, measured in people per square kilometre. Women and youth might face a double disadvantage in rural municipalities. For one, voters in smaller and more rural communities might be more conservative as compared to city dwellers, and therefore might be more likely to vote for a senior and middle-aged men (Smith et al., 2012). Furthermore, outgroups such as youth and women might be less likely to run in such a rather unfriendly political environment (Crowder-Meyer, 2013). In addition, we control for the municipality’s median age, as older citizens tend to vote for older candidates and possibly for men (Webster and Pierce, 2019).

We also include education in our models, operationalised as the percentage of residents who had completed a university degree; more educated people are generally more likely to have more progressive values than less educated individuals. This should include a greater openness to support young and women candidates (Stockemer and Sundström, 2016). We similarly include the percentage of residents who were born outside of Canada; we take this variable as a proxy for the diversity of the municipality (Hooghe et al., 2009). Finally, we control for municipalities’ employment rates, as an additional indication of the economic health of a municipality.

In total, we present four main regression models. Models 1 and 2 predict the gender of the elected mayor, and Models 3 and 4 the age. Because of the binary nature of the dependent variable gender (zero for men and one for women), regression models 1 and 2 use binary logistic regression analysis. The third and fourth models, measuring mayors’ age, are linear regression models. We further add municipality fixed-effects to the first and third models, which generate a more conservative model that controls for other unobserved municipality-level characteristics (Allison, 2009). To account for other election-specific factors between the various elections we study, we also include election-year dummies in all models.

Results

The introduction of online voting makes no observable difference in terms of diversity in electoral outcomes. Overall, we find no effect of the availability of Internet voting on electoral results. In terms of gender, both online and in-person elections elect men at an overwhelming, but almost identical, rate. Specifically, men were elected as mayor in 71.08% of elections where online voting was available, compared to 71.09% of elections that only allowed for in-person voting. Ontario also shows a slight decrease in the number of women elected under online voting (see Table 3). A logistic regression also finds no effect from online voting on the probability of electing a female mayor (see Models 1 and 2 in Table 4).

Table 3.

Descriptive Statistics for Mayoral Election Winners.

	Percentage	N
Women mayors, overall	18.01	622
Women mayors, no online voting	18.40	451
Women mayors, online voting	16.96	171
Age of mayor, overall	56.92	299
Age of mayor, no online voting	56.33	228
Age of mayor, online voting	58.83	71

Table 4.

Multivariate Regression Models Standard Errors in Parentheses.

	(1)Gender	(2)Gender	(3)Age	(4)Age	(5)# Women candidates
Online voting	−0.112 (0.586)	−0.0807 (0.455)	−1.033 (2.070)	−0.787 (1.679)	0.483 (0.319)
Mayor acclaimed	−0.848 (0.793)	−0.877 (0.676)	1.446 (2.675)	0.509 (2.163)	−0.926 (1.027)
Population	−0.0000575 (0.0000416)	−0.00000133 (0.00000178)	−0.0000107 (0.0000351)	−0.000000625 (0.00000351)	−0.00000160 (0.000000992)
Population density	−0.0000292 (0.00638)	0.0000234 (0.000650)	−0.00995 (0.0108)	−0.000758 (0.00207)	0.000209 (0.000283)
Median age	0.133 (0.206)	0.134 (0.0813)	1.336* (0.555)	1.022*** (0.276)	0.0240 (0.0579)
Median income	−0.000162* (0.0000778)	−0.00000300 (0.0000324)	0.000269 (0.000241)	0.000112 (0.000115)	0.0000122 (0.0000223)
Percentage with university degree	0.0302 (0.118)	0.0329 (0.0578)	0.0124 (0.507)	−0.151 (0.215)	−0.0248 (0.0350)
Percentage immigrants	0.0977 (0.132)	0.0392 (0.0410)	1.168*** (0.348)	0.317* (0.132)	0.0234 (0.0192)
Employment rate	0.189 (0.138)	0.0703 (0.0581)	0.0943 (0.333)	0.224 (0.193)	0.00197 (0.0515)
Total candidates					0.189* (0.0758)
Constant		−11.80* (5.878)	−30.32(40.25)	−9.036(19.19)	−3.444(4.814)
Observations	226	612	293	293	106
Groups	45	151	80	80
(Pseudo) R²			0.0764	0.1559	0.1355
Log likelihood	−72.40	−246.25
Municipal fixed effects	Yes	No	Yes	No	No
Election-year dummies	Yes	Yes	Yes	Yes	No

p < 0.05; **p < 0.01; ***p < 0.001.

The results in terms of the age of mayors elected are very similar; the average age of mayors elected in elections with no online voting available is approximately 1.5 years younger than that of elections with online voting, although this difference is not statistically significant (see Table 3). Our linear regression models (see Models 3 and 4 in Table 4) likewise find no association between online voting and the age of mayors elected. In sum, our data do not allow us to confirm our two hypotheses; on the metrics we are able to observe – the gender and age of elected mayors – the introduction of electronic voting does not appear to have any influence on the results of elections.³

While our analysis shows that younger candidates and women are no more likely to win office under online voting, it is possible that online voting could have an impact on the candidate pool. By potentially giving them the ability to reach voters who otherwise might not have voted, introducing Internet voting could entice more women to run for office, even if it does not make them more likely to win. To test this possibly, we run an additional model which incorporates data on political candidacies from the Canadian Municipal Elections Database (Lucas et al., 2021a). The database contains candidate-level data on elections in over 2000 Canadian municipalities, including candidate gender, which was manually coded from archival records (Lucas et al., 2021b). Unfortunately, the database includes only data on candidate gender, and not age. In total, we were able to extract data on the number and gender of mayoral candidates for 106 of the municipalities in our sample (69 with online voting and 39 without), for the 2018 elections only. To test the relationship between online voting and the number of women running for mayor, we fit a Poisson model with the total number of women candidates as the dependent variable. When controlling for the same control variables as in our main models, as well as the total number of overall candidates, we find no significant effect of allowing online voting (see Model 5 in Table 4).

Discussion

When it comes to Internet voting, there is a growing disconnection between theory and empirics. Backed up by rational choice theory, we can derive the following three postulates: (1) Internet voting should increase turnout, (2) Internet voting should lead to more diversification of the voting population and (3) Internet voting should trigger a greater diversification of the type of elected representatives.⁴ The empirical evidence more and more gravitates to the refusal of all three stipulations. When it comes to turnout, almost every study finds that Internet voting’s influence on electoral participation is non-existent or limited at best. For instance, neither studies of the longest lasting jurisdiction with e-voting, Estonia (Alvarez et al., 2009), nor of trials in the United Kingdom (Norris, 2004) and Switzerland (Germann and Serdült, 2017) have found a significant increase in aggregate turnout. In Canada, which is the country with the most positive reported effect, the evidence of a sustained turnout increase triggered by e-voting is lukewarm at best (see Goodman, 2014; Goodman and Stokes, 2020).

Studies testing the stipulation that Internet voting should diversify the electorate are smaller in number, but the few articles that exist also do not find evidence in favour of a diversification of the electorate. For example, Vassil et al. (2016) report in their study covering the period 2005–2015 that e-voting in Estonia started as a voting tool for the resourceful and privileged, with those aged 50–60 years old most likely to use this tool, and later diversified to other groups. Yet, they found no evidence that even after several online elections, it has made the electoral process more interesting for youth and other outgroups. Even more directly, Petitpas et al. (2021) find that the introduction of e-voting in Switzerland led to an increase in voting of the older rather than the younger cohorts. They also find that particularly for the group of occasional voters, Internet voting increases rather than decreases the gender gap in voting.

Our study adds a third layer; Internet voting does not make the pool of elected representatives more diverse. In Ontario, mayors elected through Internet elections are no different than mayors elected in non-Internet elections. They continue to be on average in their late 50s and overwhelmingly men. Given the slightly lower number of women mayors and the slightly older age of majors in online elections, if at all, our results point towards an increase in inequity in representation rather than a decrease.

Conclusion

In particular, during the current COVID-19 pandemic, Internet voting seems to be an attractive option. People can vote from any place of their convenience without actually needing to go to a polling station. At least from a theoretical point of view, Internet voting might also have the potential to render elections more diverse. Yet, all empirical evidence we gathered refutes these claims. With or without online voting, the municipalities in our sample elect roughly five men for every two women, and mayors of similar ages. On average, elected mayors in our sample are approximately 57 years old at the time of the election, and even slightly older in our online sample (59 vs 56) at the time of the election. This means that they are above 60 years at the end of their term 4 years later. Internet voting does not seem to change this tendency; if at all, it rather worsens it.

This leaves us with a dilemma; there is a marked disconnect between theory and empirics. In particular, a rational choice model of voting would predict higher electoral participation, a more diverse electorate and greater chances for outgroup candidates to get elected. Based on almost all empirical evidence available, however, none of these claims seem true. Future research should tackle delve into determining the root of this disconnect: do voters not find online voting to present a significant cost reduction? Or does the process of voting follow other criteria than a pure cost benefit analysis?

Regardless of the disconnection between theory and empirics, our study has strong repercussions for policy. In particular, in a (post)-pandemic situation, there are many good reasons to adopt Internet voting, but increasing turnout and diversifying the voter and candidate pool, as well as the pool of elected representatives is not one of them. This also implies that any political unit that decides to adopt Internet voting should temper its expectations concerning the possible effects this electoral reform might have.

Footnotes

Appendix 1

Table 5.

Multivariate Regression Models Including Turnout.

	(1a)Gender	(2a)Gender	(3a)Age	(4a)Age	(5a)#Women candidates
Online voting	−0.413 (0.613)	−0.205 (0.458)	−1.262 (2.114)	−1.043 (1.730)	0.505 (0.333)
Mayor acclaimed	−0.231 (0.880)	−0.824 (0.733)	−0.307 (2.867)	−0.135 (2.345)	−0.927 (1.027)
Population	−0.0000765 (0.0000453)	−0.00000147 (0.00000167)	−3.94e–08 (0.0000350)	−0.000000169 (0.00000382)	−0.00000158 (0.000000993)
Population density	−0.000475 (0.00633)	0.000168 (0.000637)	−0.0170 (0.0116)	−0.00167 (0.00223)	0.000197 (0.000288)
Median age	0.0145 (0.219)	0.123 (0.0805)	1.402* (0.596)	1.085*** (0.295)	0.0252 (0.0581)
Median income	−0.000189* (0.0000819)	0.00000249 (0.0000325)	0.000210 (0.000247)	0.0000932 (0.000124)	0.0000121 (0.0000223)
Percentage with university degree	0.00339 (0.117)	0.0191 (0.0574)	0.106 (0.519)	−0.156 (0.232)	−0.0238 (0.0352)
Percentage immigrants	0.0130 (0.138)	0.0483 (0.0404)	1.208*** (0.352)	0.364* (0.142)	0.0224 (0.0198)
Employment rate	0.236 (0.148)	0.0685 (0.0578)	0.0616 (0.341)	0.255 (0.203)	0.00150 (0.0517)
Turnout (%)	0.107** (0.0398)	0.0443* (0.0215)	−0.114 (0.0912)	−0.0113 (0.0742)	−0.00530 (0.0223)
Total candidates					0.190* (0.0757)
Constant		−13.29* (5.995)	−23.77(42.98)	−13.23(20.69)	−3.264(4.882)
Observations	225	595	277	277	106
Groups	45	151	79	79
(Pseudo) R²			0.078	0.164	0.1358
Log likelihood	−67.89	−241.48			−97.97
Municipal fixed effects	Yes	No	Yes	No	No
Election-year dummies	Yes	Yes	Yes	Yes	No

Standard errors in parentheses.

p < 0.05; **p < 0.01; ***p < 0.001.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

This research was supported by the Konrad Adenauer Stiftung through the Konrad Adenauer Research Chair in Empirical Democracy Studies at the University of Ottawa.

ORCID iDs

Michael J Wigginton

Daniel Stockemer

Notes

Author Biographies

Michael J Wigginton is a PhD candidate in the School of Political Studies at the University of Ottawa. Michael’s research focuses on political parties and elections, with a particular interest in representation. His work has been published in Electoral Studies, PS: Political Science and Politics and Parliamentary Affairs, among others. His doctoral dissertation examines women’s participation in party nomination contests in Canada.

Daniel Stockemer is Full Professor in the School of Political Studies at the University of Ottawa. His main research interests are political participation, political representation, right-wing extremism in Europe, as well as quantitative and qualitative research methods. Since becoming a professor in 2010, Dr Stockemer has published four single authored books, one edited volume, one textbook and more than 120 articles in peer-reviewed journals. He is also editor of European Political Science and the International Political Science Review.

References

Allison

(2009) Fixed Effects Regression Models. Thousand Oaks, CA: SAGE Publications, Inc.

Alvarez

Nagler

(2000) The Likely Consequences of Internet Voting for Political Representation. Loyola of Los Angeles Law Review 34 (3): 1115–1154.

Alvarez

Hall

Trechsel

(2009) Internet Voting in Comparative Perspective: The Case of Estonia. PS: Political Science & Politics 42 (3): 497–505.

Badas

Stauffer

(2019) Voting for Women in Nonpartisan and Partisan Elections. Electoral Studies 57: 245–255.

Blais

(2000) To Vote or Not to Vote? The Merits and Limits of Rational Choice Theory. Pittsburgh, PA: University of Pittsburgh Press.

Bird

Jackson

McGregor

, et al. (2016) Sex (and Ethnicity) in the City: Affinity Voting in the 2014 Toronto Mayoral Election. Canadian Journal of Political Science/Revue Canadienne de Science Politique 49 (2): 359–383.

Brady

Verba

Schlozman

(1995) Beyond SES: A Resource Model of Political Participation. American Political Science Review 89 (2): 271–294.

Castells

(2002) The Internet Galaxy: Reflections on the Internet, Business, and Society. Oxford: Oxford University Press.

Crowder-Meyer

(2013) Gendered Recruitment without Trying: How Local Party Recruiters Affect Women’s Representation. Politics & Gender 9 (4): 390–413.

10.

Dahlgren

(2005) The Internet, Public Spheres, and Political Communication: Dispersion and Deliberation. Political Communication 22 (2): 147–162.

11.

De Wet

Koekemoer

Nel

(2016) Exploring the Impact of Information and Communication Technology on Employees’ Work and Personal Lives. SA Journal of Industrial Psychology 42 (1): 1–11.

12.

Gabel

Goodman

Bird

, et al. (2016) Indigenous Adoption of Internet Voting: A Case Study of Whitefish River First Nation. International Indigenous Policy Journal 7 (3).

13.

Germann

Serdült

(2017) Internet Voting and Turnout: Evidence from Switzerland. Electoral Studies 47: 1–12.

14.

Goodman

(2014) Internet Voting in a Local Election in Canada. In: Grofman

Trechsel

Franklin

(eds) The Internet and Democracy in Global Perspective. Cham: Springer International Publishing, pp.7–24.

15.

Goodman

Stokes

(2020) Reducing the Cost of Voting: An Evaluation of Internet Voting’s Effect on Turnout. British Journal of Political Science 50 (3): 1155–1167.

16.

Goodman

Pammett Jon

DeBardeleben

(2010b) Internet Voting: The Canadian Municipal Experience. Canadian Parliamentary Review 33 (3): 13–21. Available at: http://www.revparlcan.ca/en/vol33-no3-internet-voting-the-canadian-municipal-experience-2/ (accessed 26 November 2021).

17.

Heisz

Schellenberg

(2004) Public Transit Use among Immigrants. Ottawa, ON, Canada: Statistics Canada. Available at: https://www150.statcan.gc.ca/n1/en/pub/11f0019m/11f0019m2004224-eng.pdf (accessed 26 November 2021).

18.

Hooghe

Reeskens

Stolle

, et al. (2009) Ethnic Diversity and Generalized Trust in Europe: A Cross-National Multilevel Study. Comparative Political Studies 42 (2): 198–223.

19.

Lucas

Merrill

Blidook

, et al. (2021a) Canadian Municipal Elections Database. Scholars Portal Dataverse. DOI: 10.5683/SP2/4MZJPQ.

20.

Lucas

Merrill

Blidook

, et al. (2021b) Women’s Municipal Electoral Performance: An Introduction to the Canadian Municipal Elections Database. Canadian Journal of Political Science 54 (1): 125–133.

21.

Miller

Powell

(2016) Overcoming Voting Obstacles: The Use of Convenience Voting by Voters with Disabilities. American Politics Research 44 (1): 28–55.

22.

Norris

(2004) Will New Technology Boost Turnout? Evaluating Experiments in UK Local Elections. In: Kersting

Baldersheim

(eds) Electronic Voting and Democracy: A Comparative Analysis. London: Palgrave Macmillan, pp.193–225.

23.

Oostveen

A-M

Van Den Besselaar

(2004) Internet Voting Technologies and Civic Participation: The Users’ Perspective. Javnost: The Public 11 (1): 61–78.

24.

Petitpas

Jaquet

Sciarini

(2021) Does E-Voting Matter for Turnout, and to Whom? Electoral Studies 71: 102245.

25.

Rubin

(2002) Security Considerations for Remote Electronic Voting. Communications of the ACM 45 (12): 39–44.

26.

Schwindt-Bayer

(2005) The Incumbency Disadvantage and Women’s Election to Legislative Office. Electoral Studies 24 (2): 227–244.

27.

Smith

Reingold

Owens

(2012) The Political Determinants of Women’s Descriptive Representation in Cities. Political Research Quarterly 65 (2): 315–329.

28.

Solvak

Vassil

(2018) Could Internet Voting Halt Declining Electoral Turnout? New Evidence That E-Voting Is Habit Forming. Policy & Internet 10 (1): 4–21.

29.

Stockemer

Sundström

(2016) Modernization Theory: How to Measure and Operationalize It When Gauging Variation in Women’s Representation? Social Indicators Research 125 (2): 695–712.

30.

Sullivan

Bélanger

(2016) La cyberdémocratie québécoise: Twitter bashing, #VoteCampus et selfies. Politique et Sociétés 35 (2–3): 239–258.

31.

Tolley

(2011) Do Women ‘Do Better’ in Municipal Politics? Electoral Representation across Three Levels of Government. Canadian Journal of Political Science 44 (3): 573–594.

32.

Vassil

Solvak

Vinkel

, et al. (2016) The diffusion of internet voting. Usage patterns of internet voting in Estonia between 2005 and 2015. Government Information Quarterly 33(3): 453–459. DOI: 10.1016/j.giq.2016.06.007.

33.

Webster

Pierce

(2019) Older, Younger, or More Similar? The Use of Age as a Voting Heuristic. Social Science Quarterly 100 (3): 635–652.

34.

Wolchok

Wustrow

Halderman

, et al. (2010) Security Analysis of India’s Electronic Voting Machines. In: CCS ’10: Proceedings of the 17th ACM Conference on Computer and Communications Security, New York, 4 October 2010, pp.1–14. New York: Association for Computing Machinery.