Voter turnout decline and stratification: Quasi-experimental and comparative evidence of a growing educational gap
| DOI | 10.1177/0263395716674370 |
| Published date | 01 May 2017 |
| Date | 01 May 2017 |
| Author | Marc Hooghe,Ruth Dassonneville |
| Subject Matter | Articles |
https://doi.org/10.1177/0263395716674370
Politics
2017, Vol. 37(2) 184 –200
© The Author(s) 2016
Reprints and permissions:
sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/0263395716674370
journals.sagepub.com/home/pol
Voter turnout decline
and stratification:
Quasi-experimental and
comparative evidence of a
growing educational gap
Ruth Dassonneville
Département de science politique, Université de Montréal, Montréal, QC, Canada
Marc Hooghe
Centre for Citizenship and Democracy, University of Leuven, Leuven, Belgium
Abstract
Turnout rates are in decline in advanced democracies, but the consequences of this trend for
inequalities in participation have not yet been investigated in a large, comparative study. In
this article, we investigate educational inequalities in six countries. First, we examine whether
Tingsten’s law of dispersion holds and whether decreasing turnout rates imply more unequal
participation. Subsequently, we examine whether the more gradual over-time process of declining
turnout rates leads to biased participation. We show that the participation gap between low- and
high-educated citizens is growing. This pattern is strongest in countries where turnout rates have
decreased most.
Keywords
education, longitudinal analysis, quasi-natural experiment, turnout, unequal participation
Received: 10th March 2016; Revised version received: 7th July 2016; Accepted: 22nd August 2016
The decline of electoral turnout levels that has been observed in numerous liberal democ-
racies constitutes an important challenge for representative democracy. In Lijphart’s
(1997) words, ‘the democratic goal should be […] universal or near-universal turnout’
(p. 2). Ideally, elections offer a reliable and representative view of the political preferences
of the population (Ansolabehere et al., 2002; Hill, 2016; Phillips, 1996; Schlozman et al.,
2012). It has been argued that lower levels of participation among specific groups of the
Corresponding author:
Ruth Dassonneville, Département de science politique, Université de Montréal, C.P. 6128, Succursale
Centre-ville, Montréal, QC H3C 3J7, Canada.
Email: ruth.dassonneville@umontreal.ca
674370POL0010.1177/0263395716674370PoliticsDassonneville and Hooghe
research-article2016
Article
Dassonneville and Hooghe 185
population entail the risk that the interests and preferences of this group will not be ade-
quately represented within the political decision-making process. In order to ensure the
principle of equality, turnout rates should hence be maximized, as declining turnout rates
entail the risk that democracies are less likely to achieve the ideal of equal representation.
Non-universal turnout is problematic because it is often associated with unequal turn-
out. As a result, some groups of citizens are less likely to be represented in the policy
process. Time and again it has been shown that some groups of citizens are more likely to
turn out on Election Day than others. Those with a higher socio-economic status, the
higher educated and the higher politically sophisticated are more likely to turn out com-
pared to citizens without those resources (Smets and Van Ham, 2013; Verba and Nie,
1972). As these inequalities in turnout also imply unequal possibilities to influence gov-
ernment policy (Lijphart, 1997), there is a clear risk that some voices will not be repre-
sented. Furthermore, if higher turnout implies less inequality, the decline of turnout rates
across advanced democracies could be thought to result in increasing inequalities in who
turns out to vote and who does not.
In this article, we investigate the evolution of inequalities in turnout in advanced democ-
racies. The theoretical and normative relevance of our analysis lies in the fact that we can
shed light on the consequences of this observed decline for representative democracy. If
the decline in turnout would be random across all population groups, lower turnout levels
could still lead to an equally representative electoral outcome. If the decline, however, is
associated with a stronger pattern of stratification, this means that some groups of the
population will be more strongly represented in the electoral outcome than others.1
We focus on education as a stratification indicator because previous research shows
that within contemporary liberal democracies, educational attainment serves much
more strongly as a mechanism of stratification than other indicators that are focused
on income and professional position (Nie et al., 1996). Indirectly, this also renders the
research more relevant for the broader debate on representative democracy because we
know that the education level of citizens also has a strong effect on their political prefer-
ences (Dalton and Welzel, 2014). As a result, we can be quite confident that inequalities
based on education levels, in practice, will also imply that the input the political system
receives is not representative for the preferences of the population as a whole. On a more
pragmatic note, we like to point out that education level is a rather reliable indicator for
comparative research. This is not necessarily the case for income measurements (which
are routinely plagued by a high level of item non-response) or measurements of profes-
sional status (which are very difficult to standardize across societies).
Does declining turnout lead to growing disparities?
Across liberal democracies, we observe a decline with regard to electoral turnout rates
(Gray and Caul, 2000; Hooghe, 2014). Blais and Rubenson (2013) note a sharp drop in
turnout rates in advanced democracies, from about 80% in the 1980s to about 70% in
most recent elections. The observation of this decrease has instigated a fierce scholarly
debate on the reasons for this decline (Hooghe and Kern, 2016). Thus far, various possible
elements have been identified, ranging from the advent of post-materialist values (Dalton,
2007b), a waning of citizens’ sense of civic duty (Blais, 2000; Blais and Rubenson, 2013),
changing electoral rules, to the fact that elections are becoming less competitive (Franklin,
2004; Franklin et al., 2004). In this article, however, our focus is not on the determinants
of the decline but only on the potential consequences of this process: if turnout levels
To continue reading
Request your trial