Professional Documents
Culture Documents
Webpdf
Webpdf
Consolidated Democracies
This book examines why people vote in the newly consolidated democra-
cies of Africa, Latin America, East Asia, and Central and Eastern European
countries.
It addresses the question of how well models or theories of electoral par-
ticipation, initially developed in established democracies, “travel” to new
democracies. Based on recent cross-national survey data, it provides the first
systematic and comparative evaluation of this topic. Drawing on political
science, sociology, and psychology approaches, it reveals what is distinctive
about voting in new democracies and how they compare between themselves
and with more established democracies.
This book will be of key interest to scholars and students of political par-
ticipation, public opinion, voting behaviour, electoral politics, and political
parties as well as to international organisations and NGOs working in the
field of democracy promotion and in emerging democracies.
List of tables vi
List of figures viii
Acknowledgements ix
1 Introduction 1
Conclusion 176
Appendices 181
References 209
Index 233
Tables
xii Acknowledgements
family is large. Without putting them in any order, I would like to thank the
following: Taku Samuel Bisong, Ma Comfort, Taboko Eveline, Andreas Tarh,
Melvis Tambe, Serge Ndi, Daniel Tanyi, Efu, Moses, Gertrude, Corazon,
Alex, Yvonne, Armstrong, Martha, Laetitia. Lastly, special thanks to my
partner, Margarete Anyi, for her constant love and support and our sons
who have been wonderful, understanding, and always supportive through the
writing process.
Finally, significant gratitude goes to the four anonymous referees who
provided constructive feedback on the original book proposal submitted to
Routledge. Without their insightful and constructive feedbacks and suggestions
notably on theory, research design, and structure, this book would have been
a less well conceived undertaking. Also, the analysis in this book was based
on data collected by the Afrobarometer, AsianBarometer, European Social
Survey, and AmericasBarometer surveys (the AmericasBarometer surveys
has a subscription fee, so I only had access to the free or reduced version of
the data). So many thanks to the Afrobarometer, AsianBarometer, European
Social Survey, and the Latin American Public Opinion Project (LAPOP) and
its various partners, sponsor, and supporters for making these data available.
Lastly, special thanks to the editor Andrew Taylor and Sophie Iddamalgoda
production editor and copyeditor Ting Baker at Routledge for being very
patient, friendly, and accommodating through this project.
1
Introduction
Did you vote in the last election? The question of why people turn out on
election day has been a longstanding issue of debate among political scientists
(Geys, 2006, p. 637) but the focus has typically been on the established dem-
ocracies of Western Europe and North America.1 This book examines the
question of why people vote in the new democracies of Africa, Latin America,
East Asia, and Central and Eastern European countries. Relatedly, it also
addresses the question of how well models or theories of electoral participa-
tion that were initially developed in established democracies “travel” to new
democracies.
These objectives are important because, over the past century, several waves
of democracy have swept across the globe, bringing about representative dem-
ocracy in places where it seemed inconceivable 50 years ago. The most recent
tide of democratisation was initiated in the mid-1970s, and, ever since, we
have witnessed a global spread of democracy. It began in Southern Europe in
the 1970s with the fall of the Portuguese dictatorship, then extended to Latin
America2 in the 1980s and culminated in the 1990s with the first democratic
elections for national representative institutions taking place in 44 countries
across Africa (22), Eastern Europe (19), and Asia (3) (International IDEA,
2002, p. 14).3 As a result of this change, the number of regimes in the world that
are presently referred to as liberal democracies has increased such that it has
become common to speak of a “third wave” of democratisation (Huntington,
1991). In this context, the question of electoral turnout is crucial as the future
of democracy in both established and emerging systems depends to a signifi-
cant extent on the electoral process. The existence of free and fair competitive
elections is invariably considered one of the critical features that defines a
nation as “democratic” (LeDuc et al., 1996; Katz, 1997). Indeed, Schumpeter
(1942, p. 265) famously offered a strictly procedural definition of democ-
racy: the democratic method is that institutional arrangement for arriving at
political decisions in which individual leaders acquire the power to decide by
means of a competitive struggle for the people’s vote. Huntington (1991, p. 7),
on the other hand, views democracy as a system in which decision-makers
are selected through fair, honest, and periodic elections in which candidates
freely compete for people’s votes. Similarly, Diamond (1999, p. 10) considers
2 Introduction
electoral democracy as a system characterised by regular, competitive, multi-
party elections with universal suffrage. Although it is possible to argue for
more stringent tests4 of a “true” democracy, it is, however, clear that elections
are a central institution of democratic governance because they reflect the
condition of democracy in as many ways as they define it (LeDuc et al.,
1996, p. 4).
Moreover, modern-day formulations of democratic theory recognise the
irreplaceable significance of the ballot box for the potential control of elites by
ordinary citizens by arguing that voting is democracy’s most fundamental act.
Bennett (1986) suggests that voting is the only form of political participation
undertaken by half the adult populace. One might rebuff this view by arguing
this rather depends on the country and election. For example, there are many
examples of elections that do not achieve a majority turnout, including US
presidential elections, European parliamentary elections in several European
countries, and local elections in countries such as the UK. Nevertheless,
an important point to note, as Roberts (2008, p. 3) argues, is that voting is
believed to keep community life vital and public institutions accountable. It
resolves conflict through “a participatory process of ongoing, proximate self-
legislation and the creation of a political community capable of transforming
dependent private individuals into free citizens and the partial and private
interest into public goods”. Also, voting is the main instrument by which citi-
zens can express their choices and influence policy in democracies. As a result,
Lanning (2008) argues that voter turnout is an exact and key indicator for a
functioning democracy.
With elections broadly regarded as the heartbeat of democracy, it is under-
standable that electoral participation has been a widely studied phenomenon
in the literature of comparative politics (Diwakar, 2008, p. 75). Nonetheless,
the question of why people vote has received greater attention among scholars
of Western democracies. These studies have not just been limited to voting at
the individual level but have also incorporated voting at the aggregate5 level
with a special focus on declining turnout and cross-national variations in
turnout (Kostelka, 2017; Blais and Carty, 1990; Blais and Dobrzynska, 1998;
Franklin, 2004; Jackman, 1987; Stockemer, 2015; 2016; Powell, 1980; Fornos,
Power, and Garand, 2004; Pérez-Liñán, 2001; Kostadinova, 2003; Kostadinova
and Power, 2007; Boulding and Brown, 2013; Pacek, Pop-Eleches, and Tucker,
2009; Lehoucq and Wall, 2004; Kuenzi and Lambright, 2007; Remmer, 2010).
There is certainly much of value in these studies and many interesting results
have been produced. However, the question of why people vote in newly emer-
ging democracies has yet to receive significant attention.
While, several studies have examined electoral participation within a single
country or region6 (Kuenzi and Lambright, 2011; Bratton et al., 2012; Bratton
et al., 2005; Bratton, 2013; Tambe, 2016; 2017a; 2018a; Orvista et al., 2005;
Flanagan et al., 1991; Fuh-Sheng Hsieh and Newman, 2002; Bernhagen and
March 2007; Barnes, 2006; Carlin, Singer, and Zechmeister, 2015; Carreras
and Castañeda-Angarita, 2014; Nadeau et al., 2017; Achen and Wang, 2017;
Introduction 3
Tsai Chia-Hung, 2001; Denton, 2016), little attention has been given to
cross-regional comparative research7 that explores and compares the nature
of electoral participation across new democracies after more than 20 years
of democratic transition and experience. In this volume, I examine turnout
at the individual level by asking: Why do people vote in newly consolidated
democracies? That is, what influences people’s decisions to vote? Relatedly it
also addresses the question of how well do models that were developed in
established democracies “travel” to new democracies. In order to furnish an
answer to this puzzle, I rely on recent cross-national public opinion surveys
from the Afrobarometer, AmericasBarometer, AsianBarometer, and the
European Social Survey to examine the factors that affect the individual
propensity to vote in eight countries in sub-Saharan Africa, nine countries
in Latin America, three in East Asia and six countries in Central/Eastern
Europe giving a grand total of 26 newly consolidated democracies across the
four geo-political regions.8 Thus, with the unique resources of cross-national
survey data, this book provides the first systematic and comparative evalu-
ation of this topic.
Although the focus of this study is on newly consolidated democracies, it is
equally important to examine what is happening in older established democra-
cies (i.e. Western European democracies) in relation as well, as this provides an
empirical benchmark against which to compare the voting behaviour in sub-
Saharan Africa, Latin America, East Asia, and Central/Eastern Europe: is
the process by which voters turn out in elections in the newly consolidated
democracies broadly similar to that we find in the established democracies of
Western Europe? The established democracies provide both theoretical and
empirical benchmarks against which to compare the new democracies.
Having established the research question that this book seeks to answer,
the rest of this introductory chapter deals with the expected contribution of
this study and the structure of the book. In addition, it is important to clarify
the term new democracies or newly consolidated democracies as it is applied in
this book. The terms new democracies, transitional democracies, emerging,
and young democracies tend to be used interchangeably by scholars and most
of the time these scholars are inclined to assume that readers understand their
usage. I shall therefore also dedicate a few words to this issue in this introduc-
tory chapter.
New democracies
What is the meaning of new-democracies? To address this question, the best
place to begin is with Samuel Huntington’s path-breaking study The Third
Wave: Democratisation in the Late Twentieth Century. Based on this study,
my operationalisation of countries I deem to be newly consolidated democra-
cies11 are defined by three broad criteria:
Notes
1 See Campbell et al. (1960); Campbell et al. (1954); Wolfinger and Rosenstone
(1980); Miller and Shanks (1996); Franklin (2004); Thomassen (2005); Leighley
and Nagler (2014); Lazarsfeld, Berelson, and Gaudet (1948); Berelson, Lazarsfeld,
and McPhee (1954); Blais and Daoust (2020); Heath, A. et al. (1993); Marsh et al.
(2008).
2 Cohen (2017, p. 1) observes that since the democratic transition in the 1980s in Latin
America, electoral democracy has been the status quo system of governments with
more than 100 (mostly) free, competitive, and fair elections for executive positions
having been held across the region.
3 Also see Carothers (2002) for a discussion on the transition paradigm.
4 In contrast to this minimalist definition of democracy, “maximalist” definitions
argue that the protection of civil liberties is necessary for democracy to flourish.
Dahl (1971, p. 7) theorised that inclusiveness, or public participation, and liberal-
isation, or public contestation, are key features of a democracy, or “polyarchy”.
5 For an overview of the current research on voter turnout at the aggregate level
see: Geys (2006); Blais (2006), Cancela and Geys (2016); Stockemer (2017).
6 While this single country or regional studies have produced excellent results,
their strength tend to lie in individual case studies rather than a truly thematic
comparison.
7 Bratton, Yun-han Chu, and Lagos’ (2010) article is one exception that examines
the “who votes” question across three regions of the global South (i.e. sub-Saharan
Africa, Latin America, and East Asia).
8 More on case selection and data justification and elaboration will be discussed in
Chapter 3.
9 See Thomassen (2000); Miller (1994) for an overview of some of the reasons for
doing comparative electoral research.
10 Two classical books that established the Columbia schools are: The People’s Votes
(Lazarsfeld, Berelson, and Gaudet, 1948) and Voting (Berelson, Lazarsfeld, and
McPhee, 1954). While The American Voter (Campbell et al. 1960) recognised the
primacy of the Michigan school.
11 In this book, although I tend to use the word new democracies, transitional dem-
ocracies, emerging or young democracies interchangeable, yet in this book, they
generally referred to countries that I classified as newly consolidated democracies.
Introduction 11
12 https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/
28468. [accessed 18/02/2019].
13 https:// s ites.google.com/ s ite/ j oseantoniocheibub/ d atasets/ d emocracy- a nd-
dictatorship-revisited.[accessed 18/02/2019].
14 Also, another shortcoming is that the data are limited to 2010.
15 Among comparativists, Freedom House scores or rating are generally considered
as the most common indicator to measure level of democracy.
16 Freedom House offers two indices of freedom: political rights and civil liberties,
which are then used to indicate “democracy”.
17 The criteria employed for selecting appropriate cases will be discuss in detail in
Chapter 3.
2
Why citizens vote
For many years, the question of why people vote has been of great interest
to political scientists but also to sociologists and political psychologists alike.
The need to provide an answer to this puzzle has generated the publication of
many articles, books, and handbooks,1 which are all aimed at throwing light
on this issue. This theoretical and empirical review of the existing literature
seeks to capture what has been written about the why and who votes questions
of electoral participation. As I show else where (see, Tambe, 2018a; 2017a),
the ‘’why vote?’’ question seeks to advance an explanation of people’s deci-
sion to vote, while the “who votes?” question deals with the characteristics
of those who vote and indeed, those who do not vote in elections. Though
these questions stand independently from each other, in an analytical sense,
in reality, they are closely dependent on each other: for instance, the social
characteristics approach that will be examined shortly, effectively addresses
both the why and the who questions at one and the same time. Therefore, in
this review, they will be dealt with simultaneously.
In reviewing the extensive literature on electoral participation, it can be seen
that Western scholars2 have dominated the debate so far in trying to explain
why people vote in elections. As Diwakar (2008, p. 76) rightly puts it, “empir-
ical literature on the determinants of turnout tends to focus on research in
Western democracies”. This chapter will therefore necessarily draw heavily on
previous studies of advanced democracies in Western Europe and the United
States of America; but, it also discusses relevant studies of electoral behaviour
in Africa, Latin America, East Asia, and Central/Eastern Europe. However,
before I move to these issues, the first part of this chapter is directed towards
the general features of the theories of voter turnout. Here, I discuss a number
of established theoretical and empirical approaches of turnout, summarising
the main assumptions and propositions of each, before moving on to state the
theoretical expectations that can be derived from these theories with regards
to electoral participation in newly consolidated democracies. In the second
part, I proceed by providing an alternative explanation of turnout in newer
democracies. This section also considers some of the findings from the dem-
ocratisation and social capital literature about electoral participation (i.e. how
central are elections to democracy in new democracies), followed with a brief
Why citizens vote 13
discussion on why and how newly consolidated democracies may differ from
established democracies in terms of individual-level voter turnout.
Where R is the total reward a citizen will gain from voting, B is the benefit
a person thinks he/she will accrue from having his/her preferred candidate
win, P is an individual perception of the probability that his/her single vote
will change the election outcome, C represents the cost of voting to the indi-
vidual, and D stands for the psychic satisfaction an individual would gain
from voting. The logical explanation of this equation is that if R is positive,
the citizen is assumed to gain a reward from voting that outweighs the cost
and will, therefore, participate in an election. As Harder and Krosnick (2008)
suggest, the more positive R is, the more likely an individual is to vote. A vari-
ation on this general approach has been proposed by Joshua Harder and Jon
A. Krosnick (2008) as follows:
Socio-demographic model
Demographics factors6
Beginning with the demographic determinants of vote choice, two classical
works by the Columbia school (e.g., The People’s Choice by Lazarsfeld,
Berelson, and Gaudet 1948; and Voting by Berelson, Lazarsfeld, and
Mcphee, 1954) established the supremacy of the demographic model in
explaining individual-level voting behaviour. According to these scholars,
citizens decisions of whether to participate or not are based on sociological
groups. Based on this line of arguments, the authors gave weight to demo-
graphic characteristics like age, gender, place of residence, and religion as
determinates of vote choice. In line with these studies, I start my examinations
of the effect of the demographic factors on electoral politics by looking at
five key indicators: age, gender, religion, ethnicity/race, and place of residence
(rural–urban divide).
Age
Beginning with age, theoretical, and empirical findings in Western democ-
racies suggest that age is an important explanatory variable that can predict
16 Why citizens vote
people’s voting participation at the individual level. Drawing on earlier
studies (see e.g. Lane, 1959; Braungart and Braungart, 1986), two types of
mechanism can be used to explain how age influence political behaviour.
First, age depicts a life-cycle development, with different stages of the life-
cycle (i.e. young, middle-aged, and elderly) staging different life experience
that can influence the way people see and think about politics. Second, age
also tells of generational or cohorts” effects that can affect participation rates
(Plutzer, 2002; Cutler and Bengtson, 1974, p. 163; O’Neill, 2003; Smets, 2012;
Blais et al., 2004; Franklin, 2004; Gallego, 2009; Persson et al., 2013; Wass,
2007). Contrary to the life-cycle approach, which places more emphases on
the dynamics of biological ageing that makes citizens gain political experi-
ence or grow into political life, the generational effects focus more on social,
cultural, and historical events that shape participatory patterns. For example,
military coup, the legacies of authoritarian rule, first democratic elections,
and or first multi-party elections can be viewed as important events that can
have effects on people’s political thinking, thus influencing the way they are
involved in electoral politics.7 However, I argue making generational effects
certainly will be more difficult or hard to identify in a comparative context
(i.e. most notably as this book focus on four geopolitical regions), given that
the regions and even countries have experienced different events at different
times. That said, in this book, my focus on age differences in participation
centres more on the life-cycle approach.8
According to the life-cycle effects, political participation, most especially
voting, is said to increase with the age of an individual. What this implies is
that young people turn out in lower numbers than older citizens. Kleimhenz
(1995), cited in Bühlmann and Freitag (2006), theorised that younger people
are more likely to abstain from voting as part of a typical life-cycle that predicts
that levels of participation tend to rise over an individual life span, as people
experience certain life events. The life-cycle theory is based on the idea that
young people have a low attachment to civic life since they either stay in edu-
cation or they are trying to establish a career for themselves after completing
their education, or because of higher mobility, which makes them participate
less (Strate et al., 1989, p. 443; Jankowski and Strate, 1995; Kinder, 2006). The
combination of these two traits deprives these young people of vital polit-
ical resources needed to be involved in politics; that is, they possess few rele-
vant political skills or knowledge. Jankowski and Strate (1995, p. 91) suggest
that the absence of these resources makes political participation difficult and
meaningless to young people. However, the life-cycle theory is not linear, for
it also predicts that participation rates of the oldest age group will tend to
be low, with a number of studies attesting to the existence of a curvilinear
relationship between age and turnout in Western European countries (Bhatti
and Hansen, 2012; Konzelmann et al., 2012; Wass, 2007; Blais et al., 2004;
Leighley and Nagler, 2014; Norris, 2003; Stolle and Hooghe, 2005; Al-bacete,
2014; Franklin 2004; Melo and Stockemer 2014). Cutler and Bengtson (1974,
p. 163) argue that not only does the social disengagement of older people
Why citizens vote 17
from social networks lead to a lower attachment to political life and voting
but the fact that older people are influenced by health problems and declining
income due to retirement all greatly hinder their involvement in voting. Thus,
while young individuals and older people tend to participate less, the life-cycle
theory suggests that the middle-aged have higher participation rates. This can
be explained by the fact that this age group tends to be more settled; they have
experience, knowledge, and responsibilities –that is, the resources and oppor-
tunities to participate in politics.
Turning to new or emerging democracies, the effect of age on voting
decisions has been shown by empirical research to adopt a familiar pattern
similar to countries in the global North (see Potgieter 2013). For example,
individual country studies in South Korea confirm age as a significant factor
influencing vote choice at the individual-level (Mo, Brady, and Ro, 1991).
Moreover, Park (2002, p. 140) argues that in Korea younger people (mostly
in their twenties) show more cynicism about politics and politicians due to
their keen sense of alienation from electoral politics and therefore tend to
participate at a lower rate. Furthermore, a striking characteristic of East
Asian countries is the high geographical mobility among young people, which
makes it hard for them to participate in electoral politics, as they are more
likely to fail to register to vote in the district to which they have recently
moved. Similarly, for sub-Saharan Africa and Latin America, several studies
also revealed a positive and significant relationship between age and voting
(Bratton, 1999, p. 565; Tambe, 2017a, 2018a; Kuenzi and Lambright, 2011;
Carreras and Castañeda-Angarita, 2014; Carlin et al., 2015; Seligson et al.,
1995). Furthermore, there is a connection between age and socioeconomic
status, given the high level of unemployment among young people in the con-
tinent; not only does this deprive them of a vital resource necessary for elect-
oral politics but it also makes them more dependent on older adults for their
basic subsistence. In fact, in this book, I argue that youth as a period of tran-
sition to adulthood has been prolonged to a period of waiting or “waithood”9
as younger people and millennials all over the world (i.e. especially those in
Africa, Latin America, East Asia, and Eastern Europe) are finding it harder
to advance through life, move away from home, and established themselves as
independent. I, therefore, expect young people to be less likely to vote than
older people.
Gender
An evaluation of previous studies demonstrates gender to be a variable that
affects people’s voting participation. As Beauregard (2014, p. 617) explains,
gender differences in political participation are found throughout the litera-
ture on gender and political behaviour. To be precise, the difference in polit-
ical participation between men and women first received much attention in the
United States in the 1980s, with Wirls (1986) identifying four gender-gaps10
in political engagement: the participation gap, the policy opinion gap, the
18 Why citizens vote
electoral gap, and the partisan gap. Specifically, empirical studies all tend to
show men participate more than women. Blais, Gidengil, Nevitte, and Nadeau
(2004) suggest that men are more likely to vote compared to women, while
Inglehart and Norris (2003) state that women continue to lag behind men in
many forms of participation. According to Burns et al. (2001), this disparity
in voter turnout is explained by the fact that men tend to have a higher desire
to vote than women. However, this merely begs the question, why? Their
explanation is therefore incomplete and unsatisfactory. Schlozman, Burns,
and Verba (1994) move away from the simple assumption of women having
less desire to participate and argue instead that the gender gap in participa-
tion can be explained by availability and access to political resources.
First, women are considered to participate at a lesser rate because they
are disadvantaged or lack the resources that are needed to facilitate political
participation. Second, another line of argument consistent with women’s lack
of resources is based on the logic of social learning and social roles attached
to both men and women. Duverger (1955) argues that adult roles assigned
to gender lead to a situation where women are more likely to be placed in
the private milieu; that is, at home. The challenge of the overwhelming share
of the housework, coupled with childcare, which may be exacerbated when
combined with the need to undertake paid employment, gives women less
free time, which is translated into a reluctance to participate in politics. By
contrast, he argues that men are often placed in the world outside the home.
Though, concerning gender gap in electoral participation, it is essen-
tial to note that in established democracies, no gender gap in voter turnout
has been observed at least since the 1990s. For example, Poole and Zeigler
(1985) drawing on the US (i.e. for a similar result for Australia and Britain
see Studlar, McAllister, and Hayes, 1998) showed that women have closed
the gender gap in voting and are now actually voting in higher numbers.
Also, recent studies indicate that the gender gap seems to have reduced or
even reversed, with women consistently being more likely to vote than men
(Carreras, 2018; Norris, 2002; Kostelka, Blais, and Gidengil, 2019; Alexander
and Coffé, 2018). Several explanations have been advanced to explain the
important change in gender difference. Among them, the most prominent
has been the development theory of gender gap that postulates that signifi-
cant changes in the sex role in post-industrial societies has had a substantial
effect on women values (Inglehart and Norris, 2000). By this, I mean that
the younger generation of women are more likely to be educated and in paid
employment, which makes their situation more like that of men than is true
of older generations of women. This argument is supported by Norris (2002)
who shows that the gender gap in voting has closed in established democra-
cies, thanks to this interaction between gender and generations.
Having established the expected direction of the effect of gender on voting
in advanced industrial democracies, the question is, will we find a gender
gap in participation across our new democracies of interest? Analysis of the
gender gap in East Asian and Latin America countries (Bratton et al., 2010)
Why citizens vote 19
reveals that the applicability of gender-sensitive laws (i.e. compulsory pri-
mary education) and the introduction of inclusive institutions (i.e. compul-
sory voting, proportional electoral system, gender quotas) encourage female
political participation and thus has enabled women to attain parity with men,
thus narrowing the gender gap (Córdova and Rangel, 2017; Kittilson and
Schwindt-Bayer, 2012; Jones, 2009; Schwindt-Bayer, 2009; Tripp and Kang,
2008; Barnes and Burchard, 2013; Burnet, 2011; Desposato and Norrander,
2009). Given this, I, therefore, might not expect men to participate at a higher
rate than women in East Asia and Latin America. However, for Africa, the
adult roles attached to gender (which may be changing) generally lead to a
situation where it is quite common to see men dominating the political sphere
than women. Based on this theoretical and empirical discussion, I would
hypothesise overall that the propensity for men to participate in voting is
higher than that of women.
Religion
Another demographic factor that has been theorised to influence people’s
decision whether or not to vote is religion. Although the world has become
more secularised over the decades, Chesnut (2003) and Broughton and Napel
(2000) argue that religion has always played an essential role in shaping elect-
oral politics both in established democracies and other parts of the world (i.e.
Latin America). Nedeau et al. (2017, p. 30) cite in many European countries
the link between religion and voting is direct and is driven from the establish-
ment of parties based on religious principles (i.e. many Christian democratic
parties). However, as pointed out by other studies from North America (the
US and Canada) the relationship between religion and voting is rather indirect
(e.g. for studies in Canada see Blais et al., 2002; for the US see Lewis-Beck
et al., 2008) where some parties do tend to attract more religious voters than
others. To be precise, relating to the two cases from North America, although
these parties are not created based on religious principles, yet they do tend to
attract voters or members due to the position that they take concerning reli-
gious incline issues such as abortion or same-sex marriage.
Moving away from established democracies, Bratton (2003) has advanced
the view that people of faith are globally more likely to vote than those who
say they are non-believers. The obvious explanation of this phenomenon
could be that churches, mosques, and temples act as agents of mobilisation
(e.g. see Segura and Bowler, 2012; Harris, 1994; Jones-Correa and Leal, 2001).
While I would expect the mobilisation effect of religious attendance to come
true in Eastern Europe, Latin America (i.e. Latin Americans and Eastern
Europeans have historically been heavily if not exclusively Catholic11) and to
some extent Africa countries where religiosity in these regions often affect
voters through parties emphasising the importance of issues that are often
debated in the public sphere.12 However, in East Asia, there is a mixed result
with regards to the role of religion in influencing citizens’ voting decisions.
20 Why citizens vote
Rood Steven (2002, pp. 157–158) shows that, surprisingly, religion had little
effect on who voters preferred in the Philippines (rather than whether or not
they actually voted). For example, Fidel Ramos won the 1992 elections despite
been a Protestant in a profoundly Catholic country, while Joseph Estrada won
the 1998 election despite being heavily criticised by the Catholic hierarchy on
his alleged immoral lifestyle. On the other hand, Chin (2002) reveals that in
Malaysia religion is an important factor in influencing how people vote, with
vote choice being between a secular modernist Islam and the more orthodox,
fundamentalist brand of Islam. Overall, I can plausibly expect individuals
who are more religious to have a higher propensity to vote.
Ethnicity or race
Ethnic identity is the inclination of an individual to define themselves in
terms of their cultural origin and descent-based traits (Olorunsola, 1972;
Horowitz, 1991; Berman et al., 2004). In line with this argument, Wolfinger
(1965, p. 896) suggests that ethnic identity or origin represents one of the
strongest dimensions of people’s perception of themselves and of others
and is therefore considered to play a very important role in politics. Based
on these findings, I move forward with the argument that ethnicity is one of
the most essential variables in the demographic model, but the question that
arises is how does ethnic identity/race influence people’s propensity to vote in
an election? When ethnic identity becomes sufficiently strong it can consti-
tute the basis for forming political opinion and in stimulating people’s voting
decisions; this, of course, usually leads to what some experts term “ethnic
voting”. Wolfinger (1965) explains ethnic voting as a situation where members
of a cultural group tend to show disproportionate affinity at the polls for a
particular political party. Because of the strong affinity that people share with
others, parties, interest groups, or other social groups may take advantage of
this and therefore mobilise individuals who then tend to vote as a bloc.
Posner (2005) stipulates that the logic of ethnic voting can be explained by
the fact that when people from the same descent are mobilised, they tend to
turn out to vote since they believe they are expressing their solidarity with their
group. This act of group solidarity might elevate leaders from their own ethnic
background to positions of power, most especially in the political executive,
thereby gaining collective representation. As some scholars have highlighted,
this can lead to election outcomes that are mere headcounts of ethnic groups,
that is, if turnout is high and if all voters chose parties associated with their
own communal identities then an election can resemble an ethnic “census”
(Lever, 1979; Horowitz, 1985; Ferree, 2006; McLaughlin, 2008). In many
African, Latin America, and East Asian countries, it is common for people
to show a strong affinity with a particular ethnic or tribal group. Because
of this strong affinity that people share with others, parties may take advan-
tage of this and therefore mobilise individuals who may then tend to vote as
a bloc. In sum, drawing these studies, it might seem that ethnicity has two
Why citizens vote 21
dimensions relevant to turnout.13 One is the nature of the particular ethnic
group (i.e. whether they have a distinct propensity to participate in the elect-
oral process). The other is the degree to which ethnicity is a political cleavage
in a given country. So, people from group A may vote more because people of
group B are mobilised to vote more. In new democracies, particularly those of
sub-Saharan Africa, this is likely to be an interesting line of analysis. Based on
this, I would, therefore, expect people who have strong ethnic attachments to
have a high propensity to vote.14
Place of residence
Looking at the last demographic variable, I consider the individual place of
residence (i.e. the distinction between rural and urban voters). Earlier studies,
at least in the context of established democracies, suggest that citizens in
rural areas were historically more likely to engage in electoral politics as a
result of the higher level of associational or informal network.15 However, a
survey of recent studies, in Western democracies, indicates that scholars are
still undecided if rural or urban voters tend to participate more. For example,
Karp et al. (2007) argue voters in urban or cites will be more likely to engage
since they are more attractive locations for parties to canvass. On the other
hand, Hoffmann-Martinot (1994) refutes the idea that turnout will be higher
in urban or cities centres by arguing that urbanisation tends to reduce inter-
personal bond and social networks thus making it less likely for people to par-
ticipate. To my knowledge, although the effect of rural and urban divide has
not been thoroughly examined in the context of newer consolidated democra-
cies, yet in a study that focuses on the determinants of non-voting across new
democracies in Central/Eastern European countries and sub-Saharan Africa,
Tambe (2018b) provides evidence that depicts non-voters were mostly those
in urban or cities areas. Similarly, in a stimulating study of individual-level
voting in Latin America, Carreras and Castañeda-Angarita (2014, p. 1092)
reveal that place of residence tends to influence electoral behaviour, with Latin
Americans who live in urban areas being less likely to vote compared to rural
dwellers.16 Turning our focus on the East Asia region, individual studies (i.e.
Mo, Brady, and Ru, 1991) and comparative studies (Bratton et al., 2010) con-
clude that rural voters are more likely to participate compared to the urban
dwellers. Based on this initial finding, I would expect voters in rural areas will
be more likely to participate than urban/city dwellers.
Socioeconomic factors
Having concluded with the demographic predictors of turnout, I now turn to
the socioeconomic dimension. This model centres on individual resources as
a potential explanation of citizens’ voting propensity. That said, researchers
have bundled together groups of variable relevant to personal resources,
which are said to influence people’s decision to vote at the individual level.
22 Why citizens vote
These variables consist of income, education, and/or social status. From its
inception, the socioeconomic/resource model has stood the test of time in
predicting voters’ decisions in Advanced Western democracies, in that each of
these factors has widely been found to be significantly associated with voting.17
How do these socioeconomic/resource variables dictate why and who votes?
To answer this question, I begin with the classic study of Sidney Verba and
Norman Nie. In their seminal work Participation in America, Verba and Nie
(1972) built their explanation of electoral participation on what they called
a “Baseline Model” consisting of income, occupation, and education. Later
renamed the Resource Model, this approach came to dominate explanations
of the individual turnout decision in the United States and elsewhere. They
suggest that specific resources are vital for there to be effective participation
of the citizenry in the political life of their societies: time, money, and civic
skills. First, two prime resources for investment in political participation are
time and money. A simple baseline condition for political involvement, espe-
cially more involved modes of participation, is the availability of time (see
Caramani, 2008). In democratic societies, individuals use time in the service
of political action in many ways: working in a campaign, writing a letter to a
political official, attending a community or party meeting, and, finally, during
voting. It is often argued that people who do not work, or who do not work full
time, tend to display higher levels of political participation (Brady et al., 1995,
p. 273). On the other hand, money can, of course, be donated to candidates,
parties or innumerable political organisations or causes. The major critique
here is that time and money differs in significant ways as resources. In com-
parison with money, time is both constrained and more equally distributed –
everyone has only 24 hours in a day. The upper limit on money is, of course,
much less constrained and the differences among individuals can be much
larger (Brady et al., 1995, p. 273).
Second, the next resource is civic skills. Civic skills here refer to those
communication and organisational capacities that are so essential to pol-
itical activity. Citizens who can speak, write well, or who are comfortable
organising and taking part in meetings are likely to be more effective when
they get involved in politics (Brady et al., 1995). The acquisition of civic skills
begins early in life, at home, and especially in school. This process can con-
tinue throughout adulthood. Adult civic skills relevant for politics can be
acquired and honed in the non-political institutions of adult life: the work-
place, voluntary associations, and churches. All these undertakings represent
opportunities in non-political settings to learn, maintain, and improve civic
skills. Drawing on the work of Verba and Nie, recent scholarship on electoral
participation has tended to group several variables that jointly correspond to
socioeconomic status (SES). As noted by Plutzer (2018, p. 78) socioeconomic
status, or SES, is not a variable but, rather, a catch-all term that encompasses
a wide range of potential characteristics. SES consists of an individual’s edu-
cational level, income, and/or occupational status. These variables are said to
play an important role in understanding people’s political behaviour, most
Why citizens vote 23
notably as it concerns their voting decisions. From a theoretical perspec-
tive, a high level of SES is considered to lead to a higher propensity to be
involved in electoral politics. This is because people with a higher level of
education and income are believed not just to possess more time, but also to
have the capacity to understand and comprehend the daily political debates
and discussion. Various scholars (Verba, Nie, and Kim, 1978; Wolfinger and
Rosenstone, 1980; Leighley and Nagler, 1992; Buhlmann and Freitag, 2006)
have empirically demonstrated that the socioeconomic status of an individual
is positively associated with electoral participation at the individual level.
Looking at the relationship between education18 and turnout separately, the
weight of evidence shows that in advanced democracies those with a higher
level of education are more likely to turnout (see e.g. Schlozman, Verba, and
Brady 2012; Verba, Schlozman, and Brady 1995; Wolfinger and Rosenstone
1980) with La Due Lake and Huckfeldt (1998) attesting the positive link
between education and political participation as “one of the most reliable
results in empirical social science”. Hillygus (2005) but also Gallego (2010)
hint about the fact that education is a primary driver of increased political
participation is “largely uncontested”. Putnam (1995), on the other hand,
posits education is the best individual predictor of participation. Also, in their
meta-analysis of individual-level research on voter turnout that drew mostly
from studies of Western democracies, Smets and van Ham (2013) show that
education is indeed positively related to individual-level turnout with most
studies having a success rate of 67 to 72 per cent. Relatedly, in his review of
education and political participation, Persson (2015) reaffirms that education
is of central importance in most studies of political behaviour, with studies
finding individuals with higher education participating at a higher rate. In
discussing the role of education on political participation, Persson (2015,
p. 690) demonstrates at length three main linkages that explain how educa-
tion is related to political participation: absolute education model, the pre-
adult socialisation model, and the relative education model. Beginning with
the absolute education model, Persson argues education has a direct causal
effect on political engagement. What this implies is that education tends to
increase civic skills, political knowledge, and/or political efficacy, which acts
as the causal mechanism triggering participation.
Next, concerning the pre-adult socialisation model, Persson contends edu-
cation works as a proxy for variables such as family socioeconomic status, pol-
itical socialisation that takes place in the home, and/or personal characteristics
such as cognitive ability, which are considered to affect participation rather
than education itself. Finally, the last linkage that exists between education
and political participation centres on the relative education model or sorting
model. According to this perspective, there is an indirect effect of educa-
tion on political participation, which can be explained via social status19 (see
e.g. Nie, Junn, and Stehlik-Barry, 1996). Within the literature, social status
is defined as having a central social network position in society, with Nie,
Junn, and Stehlik-Barry (1996) operationalising social network centrality as
24 Why citizens vote
proximity to governmental incumbents and political actors who make public
policy and to those in the mass media who disseminate. Drawing from this
perspective, it is, therefore maintained it is social status and not education that
increases participation.
Next, if I turn to the relationship between income and turnout, findings
by Teixiera (1992) suggest individuals with a low level of income are less
likely to vote compared to their wealthier counterparts, which also raise the
question why? Drawing on their now seminal work: Who Votes? Wolfinger and
Rosenstone (1980, p. 20–22) outline a theoretical and empirical connection of
why income tends to be associated with higher turnout, which is summarised
as follows:
• The poor have less time to devote to matters not essential to everyday
existence.
• Wealthy people have jobs that tend to increase their political engagement.
• Income affects one’s social context; thus, the wealthier are more likely to
be exposed to social networks with norms of civic duty.
• Lastly, wealthy individuals have a more significant stake in the political
system.
Socio-demographic Age +
model Gender (male) +
Religion +
Ethnicity/race +
Residence (rural) +
Education +
Note: Explains what variables within the four regions that are expected to take a positive/negative
sign. * AF (Africa), * LEC (Latin America), * EA (East Asia), and *CEEC (Central and Eastern
Europe).
26 Why citizens vote
Table 2.1 summarises this part of the discussion so far. Studies in advanced
Western democracies have so far dealt with six main socio-demographic
characteristics that explain people’s decision whether to vote or not to
vote: age, gender, religion, ethnicity/race, residence, and socioeconomic status
(i.e. education or income). Based on these studies, the standard hypothesises
that can be derived with regards to each of these variables are as follows:
H1a: Beginning with age, the older an individual is, the higher is his or her
propensity to turnout.
H1b: As concerns gender, I would expect that men will have a higher pro-
pensity to vote than women.
H1c: Individuals who are more religious will have a higher propensity
to vote.
H1d: Concerning ethnic identity, people who have strong ethnic
attachments will have a higher propensity to vote.
H1e: I would expect voters in rural areas will be more likely to participate
than the city or urban dwellers.
H1f: Moving to the socioeconomic status of an individual, the higher an
individual’s socioeconomic status (i.e. education and or income) the
higher is his or her propensity to vote.
Mobilisation model
The mobilisation model stipulates that the only way to explain political
participation is to move beyond the world of individuals to include social
networks, politicians, parties, activists, and interests groups (Rosenstone and
Hansen, 1993, p. 23). I have already discussed a number of demographic
influences that pertain to the socio-demographic model (age and gender), but
other sociological factors relate to the idea that people’s propensity to vote is
influenced by their associational life and this is what the mobilisation model is
concerned with. The central logic here is that social networks such as unions,
pressure/interest groups, and parties exert pressures on people to behave as
members of a group rather than as isolated individuals. In making people
act as groups rather as individuals, not only are these social networks able
to mount pressure on their members that enable them to comply, but these
networks have the potential to reduce the cost of political participation by
providing group members with political information and resources that are
necessary for electoral participation.
Apart from Rosenstone and Hansen, several other works have highlighted
the importance of the mobilisation model in getting out the vote. Uhlaner
(1989) found that group identification can make it more rational for par-
ticular segments of the electorate to show up at the polls even when the
cost-benefit incentive structure clearly predicts that they should not do so.
Further, the politicisation that occurs within parties, interest groups, and
social groups might increase the likelihood that an individual will vote
Why citizens vote 27
(Brady, Verba, and Schlozman, 1995; Leighley and Nagler, 2007). Based
on the theoretical literature, I can highlight the following variables that
will be empirically tested in this book. The variables are grouped into three
main categories: party identification, associational networks/civil society
organisations, and clientelism.
Party identification
Campbell, Gurin, and Miller (1954, pp. 88–89) define party identification as a
“sense of personal attachment which the individual feels towards the party of
his/her (sic) choice”. Previous studies have demonstrated the utility of party
identification as an important variable in understanding and explaining pol-
itical behaviour, especially voting in Western electoral politics (Dalton, 1984,
p. 264). According to this view, elections in most democracies are almost always
partisan contests, which therefore means that partisan attachments guide citi-
zens through the complexities of politics, providing vital information that
is necessary for making an informed electoral decision (see Brader, Tucker,
and Duell, 2013). Strong partisan identifiers are generally more involved in
electoral politics than non-identifiers (Dalton, 1984, p. 267). Partisanship,
therefore, helps in performing a mobilisational function;24 it is this mobilising
function that I am interested in, on the assumption that it might influence
individuals to vote at the individual level. King (1969, pp. 123–128) but also
Verba et al. (1978) have previously linked party identification to turnout.
These studies assume that attachment to a political party draws an individual
into the political process. In short, it has been shown that people with weak
or no party identification at all will show a lower propensity to participate in
elections, while those individuals with a strong party attachment or identifica-
tion possess a higher propensity to vote in elections.
Yet, concerning East Asia, studies of voter behaviour conducted in Hong
Kong and South Korea found party identification to be very weak (Scott,
1996; Park, 2002). Newman (2002, p. 57) argues that weak party identifica-
tion in Hong Kong could be explained by the fact that political parties are
all relatively new and subject to considerable change and weak consolidation
over a short period of time, and also because the Hong Kong population
now includes a large component of immigrants from the Mainland where
the cultural norm is to avoid politics. Given that parties in East Asia, Africa,
Latin America, and Eastern Europe are new and fluid, it may be that I will
not find a positive relationship between party identification and propensity to
vote at the individual level, as the comparative literature on Western democ-
racies would suggest. At the aggregate level, it is certainly possible that par-
tisan identification will not drive turnout strongly. It should also be said that
the aggregate-level mobilisational impact of partisan commitment may be
diminishing even in established democracies. There is now extensive empir-
ical literature that tends to show declining party memberships in Western
democracies (Van Biezen, Mair and Poguntke, 2012). For example, tracing
28 Why citizens vote
party membership and the extent of election organisation across UK con-
stituencies, Fisher et al. (2006, pp. 4–6) indicate a steady decline in member-
ship within the Conservative, Labour, Liberal Democrats, Plaid Cymru, and
SNP from 1992 to 2001.
Similarly, there is ample evidence of the weakening of partisan identifi-
cation across the established democracies (Dalton, 2000). This declining
membership and identification, coupled with the declining cost of obtaining
political information thanks to an information explosion that has been made
possible through the expansion and growth of mass and electronic media,
gives one cause to question whether party identification is still a viable explan-
ation of the voting decision at the individual level. That is, as party mem-
bership declined, so has partisan identification –and this should mean that
a significant source of electoral mobilisation has been eroded. From this
perspective, it is no surprise that electoral turnout has declined as well in
Western democracies. These developments are all consistent with one other.
Nevertheless, at the individual level, where people retain a strong partisan
identity, this may be expected to boost the propensity to vote. Thus, I hope to
find a positive association between party identification and propensity to vote.
That is, the more strongly an individual develops party ties, the higher will be
his or her propensity to vote.
Clientelism
The practice of clientelism is relatively frequent in emerging democracies
(i.e. particularly those of Africa, Latin America, and East Asia) and is gen-
erally used by politicians and political entrepreneurs seeking for public office
to attract voters. Clientelism typically revolves around a network of direct
relationships between citizens and the patron (i.e., politicians) involving the
exchange of targeted goods or gifts in exchange for their support. Or as cited
by Kramon (2017, p. 7), clientelism tends to take on many different forms
ranging from politicians targeting resources directly to voters to sometimes
through direct vote-buying. Hicken (2011) argues the practice of clientelism
appears to be one of the main ailments affecting the quality of democracy
or democratic governance. First, clientelism tends to run counter to democ-
ratisation given that it does impede the development of political institutions,
undermines democratic consolidation by limiting the exercise of citizenship
(see e.g. Graziano, 1973; Fox, 1994) and as cited by Stokes (2005) but also Lyne
(2007) it endangers the secrecy of the vote. Second, Keefer (2007); Kitschelt
et al. (2010); Singer (2009); and Persson, Tabellini, and Trebbi (2003) argue
clientelism tends to be associated with corruption as it implies illegal practices
of vote-buying and encourages politicians to adopt illegal fundraising strat-
egies. In this volume, although the core objective of clientelism is generally to
alter the electoral results in a way that suits the patron, I argue that clientelism
does work as a tool of electoral mobilisation.
For instance, as cited by Tambe (forthcoming) in new democracies, particu-
larly those in Africa, the resource-poor compared to rich tend to have little
or no contact with the state or political parties except for the period leading
to elections. During these election periods, it is not uncommon to find polit-
ical parties, party heads, representative, and members of parliaments going
30 Why citizens vote
Table 2.2 Overview of the expectation of the mobilisation model on vote choice
Note: Explains what variables within the four regions that are expected to take a positive/negative
sign. * AF (Africa), * LEC (Latin America), * EA (East Asia), and *CEEC (Central and Eastern
Europe).
Political efficacy
Political efficacy is described as the perception citizens have of being cap-
able of acting effectively in the political arena. Hadjar and Beck (2010) define
34 Why citizens vote
political efficacy as the degree to which a person believes in his or her ability,
first to understand politics and second to influence the political processes
through action, including voting. Political efficacy comes in two parts: internal
efficacy generally refers to an individual’s competences, skills and resources to
deal with politics, while external efficacy is defined as an individual’s percep-
tion that political institutions are responsive to people’s attempts to exert pol-
itical influence (Clarke and Acock, 1989, p. 552). Studies by Campbell et al.
(1954), Abramson and Aldrich (1982), and Becker (2004) have shown that
political efficacy is the single most important variable in the psychological
model that tends to influence the decision whether to vote or not to vote. This,
therefore, means that widespread political efficacy throughout the electorate
will lead to a higher voter turnout, while the lack of political efficacy among
voters will lead to lower voter turnout. This line of argument is supported by
empirical studies by Craig and Maggiotto (1982), Karp and Banducci (2008),
and Norris (2002) who demonstrate that more efficacious voters are, the more
likely they are to vote. Consequently, I would hypothesise that voters in young
democracies who are politically efficacious will reveal a greater propensity to
participate in elections.
Note: Explains what variables within the four regions that are expected to take a positive/negative
sign. * AF (Africa), * LEC (Latin America), * EA (East Asia), and *CEEC (Central and Eastern
Europe).
H3a: The higher an individual’s political interest, the higher is his or her
propensity to vote in an election.
H3b: The higher an individual’s sense of internal or external efficacy, the
higher will be his or her propensity to participate in electoral politics.
H3c: The greater an individual’s trust in political institutions, the higher
will be his or her propensity to vote.
Institutional factors
Having surveyed some of the relevant literature that focuses on the polit-
ical and institutional factors, it is possible to identify a set of variables that
I intend to test empirically. They are as follows:
Electoral system
According to Farrell (2001, p. 4), the electoral system is defined as “the means
by which votes are translated into seats in the process of electing politicians
into office”. A good number of comparative studies have succeeded in
showing that the electoral system impacts on political participation, particu-
larly as it relates to turnout (Franklin, 2002; Norris, 2002). The argument here
is that cross-national differences in turnout can be explained by the type of
electoral system that is adopted by a country. What kind of electoral system
is theorised to positively affect citizens’ desire to participate? The literature
suggests that proportional electoral systems (i.e. PR systems) are most likely
to have a positive effect on voters’ turnout, as opposed to majoritarian or
plurality systems (Blais and Aarts, 2006), with Blais and Carty (1990) pro-
posing this is because PR systems tend to reduce distortion in the conversion
of ballots into seats. Essentially, proportional systems tend to enhance the
likelihood that a single vote might contribute to the election of a winning
candidate, so there is a greater incentive to vote. Similarly, Ladner and Milner
(1999) argue that majoritarian electoral systems tend to hinder electoral par-
ticipation for two reasons: First, supporters or voters of a previously defeated
party may either stay away from the election since they consider their vote to
have no effect on the election.
Second, they argue that majoritarian electoral systems tend to hinder voting
because previously defeated parties may reduce their mobilisational efforts.
With regards to these considerations, therefore, the theoretical proposition
which I will make is that we expect proportional representation to enhance
people’s propensity to vote. This is supported by the fact that PR systems
encourage citizens to vote since they know that votes transfer more directly into
party representation and they thus perceive the system as fair (Endersby and
Krieckhaus, 2008, p. 602; Stockemer, 2015). Based on this reasoning, I hypothe-
sise a positive relationship between the level of proportionality of an electoral
system and the propensity of voters to participate in elections. However, taking
into consideration that a bulk of these studies were conducted in Western
democracies, we cannot be sure that in new democracies we are still going to
find such as strong correlation between the electoral system variable and voter
turnout, given the newness of the electoral systems in these countries.27
Compulsory voting
Birch (2009, p. 2) defines compulsory voting as the legal obligation to
attend the polls at election time/day. Studies in Western democracies have
Why citizens vote 39
categorically established compulsory voting as having a positive effect on
people’s voting decisions. Blais (2000) suggest the effect of compulsory voting
on turnout is one of the robust findings in studies that study cross-national
variation in voter turnout.28 Though compulsory voting clearly increases
voter turnout, most especially in countries where it is enforced, little is known
with regards to the type of sanction that is applied to non-voters. The uncer-
tainty about the exact influence of compulsory voting exists because research
findings on the contribution of this variable usually have mixed results. One
brand of research reveals turnout to be high in countries where compulsory
voting exists and is complemented with sanctions. However, another group of
studies shows that in countries where voting is compulsory by law, but where
the law is not enforced by sanctions, voter turnout is equally high since the
duty to vote is internalised by the citizens (Hadjar and Beck, 2010).
Concurrent elections
Concurrent elections is another contextual factor that will foster people’s pro-
pensity to vote. Cox and Munger (1989) contend that the possibility of many
elections taking place at the same time will arouse greater media coverage in at
least one of the elections, which invariably boosts the amount of money spent
on campaigning and as such increases the amount of information and public
awareness about the issues at stake in these elections. Additionally, Filer and
Kenny (1980) and Carter (1984) suggest concurrent elections tend to increase
people’s propensity to vote given that the cost of going to the polling booth is
fixed and unrelated to the number of elections in which a voter needs to cast
a vote. I, therefore, expect people would participate more in countries where
multiple elections are held concurrently than in countries where they are not.
Political/institutional Disproportionality -
model Compulsory voting +
Closeness of the election +
Concurrent elections +
Note: Explains what variables within the four regions that are expected to take a positive/negative
sign. * AF (Africa), * LEC (Latin America), * EA (East Asia), and *CEEC (Central and Eastern
Europe).
scholars such as Cox and Munger (1989) and Shachar and Nalebuff (1999)
have succeeded in showing empirically that campaign efforts by parties and
candidates are usually focused more on districts where outcomes are expected
to be close. This means that there is an added mobilisational effect in such
contests as well. Each of the variables discussed in the previous paragraph
might be considered a factor relating to the institutional context of the polit-
ical system in question: proportional electoral systems increase the propensity
to vote; the more competitive an election, the higher the propensity to vote
in such countries; countries having compulsory voting will produce a higher
propensity for people to vote and individuals have a higher propensity to vote
when legislative and presidential elections are held concurrently. I would,
therefore, expect such factors to come into play in Africa, Latin America,
East Asia, and Eastern Europe. Table 2.4 summarises the variables and the
expected effect across the four regions.
Note: Explains what variables within the four regions that are expected to take a positive/negative
sign. * AF (Africa), * LEC (Latin America), * EA (East Asia), and *CEEC (Central and Eastern
Europe).
Notes
1 For an overview of recent handbooks on voting, electoral behaviour or public
opinion see: Arzheimer, Evans, and Lewis-Beck (2017); Fisher et al. (2018).
2 For a good review of this literature, see Blais (2006).
3 For a meta-analysis of individual-level research on voter turnout: see Smets and
van Ham (2013); but also Evans (2004).
56 Why citizens vote
4 The majority of studies of voter turnout take the rational choice theory or Down’s
economic model as a starting point (e.g. see Blais 2000; Franklin 2004; Aldrich,
1993; Bufacchi, 2001).
5 It is a meta-theory in which most of the models are nested. For example, the
importance of mobilisation and/or institutions can be interpreted through the
lenses of the rational choice model.
6 As cited by Plutzer (2018, p. 69), the term “demographics” conjures up the idea
of personal characteristics whose categories are typically reported in national
censuses and government reports.
7 Nadeau et al. (2017, p. 27) note events of this nature are not likely to affect
everyone equally as they can leave more impressionable and long-standing effects
on some age groups than others.
8 As cited by Smet (2010, 2012), making a theoretical distinction between the life-
cycle and generational approach is easier than disentangling them empirically.
9 For an overview of the concept of waithood and political participation see: Dawson
(2014); Honwana (2012)
10 In this volume my interest is more on the gender gap related to electoral
participation.
11 However, in Latin America, the Catholic church has lost its monopolist positions
with significant rise of protestants and individuals with no religion.
12 Also, for a more recent and complete account of religiosity on vote choice in Latin
America see Boas and Smith (2015).
13 I thank one reviewer for pointing out this point during the initial review of this
book project.
14 However, it is equally possible that turnout among ethnic minorities would be
lower than among ethnic majorities, which may be partly explained by their
limited access to resources and skills.
15 In their meta-analysis of individual level turnout, Smet and van Ham (2013, p. 7)
show that this line of argument or hypothesis is updated as almost all the test and
studies find insignificant effects.
16 Bratton, Chu, and Lagos (2010) explain this by arguing social pressure to partici-
pate may be felt much more strongly in rural areas, which house more tightly knit
communities. In contrast, Kuenzi and Lambright (2011) advance the conditions
for political mobilisation through patronage networks are more favourable in rural
areas, as a recent study of electoral participation in Africa has demonstrated.
17 Jennings (1979) contends an individual’s socioeconomic status has a known posi-
tive relationship with most form of political participation.
18 Apart from age, no other socio- demographic variable has received as much
attention as education (see Burden 2009).
19 In the literature, high social status is defined as having a central social network
position in society
20 SES seems an important determinant of turnout in the context of Western democ-
racies amid a few mixed results. For an overview of these studies see Norris (2002);
Pierce (1995); Teorell et al. (2007); Topf (1995).
21 It can be argued that in Asia, poor voters are often tied into clientelistic networks
(which are very efficient in mobilising voters), while middle-class voters are often
disillusioned with politics and hence more likely to abstain.
22 This result is perhaps driven by a high proportion of older people in countries
such as Mongolia, Taiwan, and China who lack any kind of formal schooling but
who, nevertheless, are mobilised to become habitual voters.
Why citizens vote 57
23 Blaydes (2011) attributes the high rate of participation of those with low educa-
tion to vote buying.
24 See e.g. Karp et al. (2008).
25 I find it important to note Almond and Verba referred to Italy, Germany, and
Mexico of the 1950s/1960s
26 For instance, Franklin (1996, p. 219) has outlined a number of instrumental motiv-
ations that are largely determined by the institutional context in which elections
are held, such as the salience of elections, the use of compulsory voting, highly
proportional election systems, postal voting, weekend polling, electoral formula,
and effective number of parties.
27 For example, in their study of voter turnout at the aggregate in Latin America,
Fornos et al. (2004) do not detect any association between PR and turnout.
28 For individual-level evidence, see Perea, 2002 and Brockington, 2004.
29 Moreover, since scholars tend to view electoral choice as been contingent on other
freedoms such as: expression, assembly and association, the safeguarding of these
liberties also constitutes at least an implicit element of most definitions.
30 Studies in the context of sub-Saharan Africa and East Asia reveals holding
regular elections does not necessarily leads to democracy (e.g. see Bleck and van
de Walle 2018; Rodrigues Sanches, José Jaime Macuane, and Chipo Dendere 2020;
Morgenbesser and Pepinsky (2019).
31 See Svolik (2015).
32 Others such as clientelism, corruption, and authoritarism are discussed in more
detail along the core theoretical and alternative models or explanations of elect-
oral participation.
33 Pop-Eleches (2010); Rose and Mishler (1998) show anti-party sentiment has been
relatively high in Eastern Europe.
34 With a few exceptions like Gabon, Algeria and Mauritius, South Africa (Africa);
Brazil, Argentina (Latin America); South Korea (East Asia); Croatia, Czech
Republic, and Hungary (Eastern European).
35 Exclusion by socioeconomic group occurs when individuals are denied access to
services or participation in governed spaces based on their identity or belonging to
a particular group.
36 For detail of diversity and social cleavages in the Asia-Pacific region, see Reilly
(2006); Hyug Baeg Im (2004); Montinola (1999); Albritton (2004); Surin
Maisrikrod (2002).
37 However, it is important to note the time of democracy in Latin America and
citizens experience with elections and voting does vary across countries ranging
between 1980 and 1996 as this mark the breakdown of authoritarism in the region
except for Cuba. See Nohlen (2005, p. 9).
38 Pioneering studies of vote choice included those of: Dow’s (1998), which focuses
on the analysis of policy relevant voting in Chile; Domínguez and McCann’s
(1998) which dealt with a two-step model of vote choice in Mexico and, finally,
Mainwaring’s (1999), which focuses on the weak party attachment in Brazil.
39 Carlin, Singer, and Zechmeister (2015) cite analysts tended to focus either on
Brazil or Mexico because of the development and production vote- oriented
survey project. Also, for details on studies Brazil and Mexico see: Ames, Baker,
and Rennó (2008); Domínguez and Lawson (2003); Moreno (2003, 2009b); Telles
and Moreno (2013).
40 However, a select number of small states studies include those conducted in
Venezuela by Weyland (2003) and Nadeau, Belanger, and Didier (2013).
58 Why citizens vote
41 As noted by Kostelka (2014) but also Roberts (2018, p. 10) before 1989, Central
and Eastern European countries, which used to be known as the world harshest
dictatorship, now comprise countries that are ranked among the global demo-
cratic front-runners.
42 For a discussion of this regime change during this period see: Fagan and
Kopecký (2018)
43 Fakuyama further notes that human civilisation might be witnessing a final stage
of institutional evolution, represented by the universal spread of liberal democra-
cies. (Also, for more detail on this historical development see Herron (2009); Coffé
and van der Lippe (2010).
44 For details, see studies by Turner (1993), and Ishiyama (1996).
45 More details on the operationalisation of the dependent variable will be given in
Chapter 3.
46 Corruption is conventionally defined as the abuse of public office for private gain
(see e.g. Shleifer and Vishny, 1993)
47 Apart from voting, Monyake and Hough (2019) find that in sub-Saharan Africa,
people’s experience with corruption or bribery tends to increase the likelihood of
engaging in protest activities.
48 In a more recent of economic voting in major low-income regions, Gélineau (2013)
finds a significant relationship in the expected direction.
3
The comparative framework
This chapter deals with the research design and data upon which the entire
study rests. I begin with a discussion of the methodological issues and problems
associated with research on electoral studies, before proceeding to the next
section of the chapter that centres on the research method and design upon
which this book is based. Here, I describe the methods to be used and high-
light the methodological challenges associated with studies of electoral behav-
iour. This includes a discussion of case selection and the operationalisation
and measurement of our dependent and independent variables. But before
I move to these issues, it is essential to clarify one thing: why focus on the
newly consolidated democracies of sub-Saharan Africa, Latin America, East
Asia, and Central/Eastern Europe?
Selection of cases
One of the most durable conventions about what constitutes an appropriate
research strategy in comparative politics is the selection of cases (Geddes,
1990, p. 131). The cases in this volume consist of newly consolidated democ-
racies of sub-Saharan Africa, Latin America, East Asia, and Central/Eastern
European states. As previously mentioned in the introductory chapter, the
cases selected for empirical analysis in this book will be defined by adherence
to these three criteria: countries that have experienced the end of authoritarian
rule, countries that have experienced the installation of a democratic regime
and countries that are experiencing the consolidation of their democratic
regimes (i.e. democratic consolidation can be expected to take about 20 years,
that is, approximately five or six democratic elections). Beyond this, however,
I rely on Freedom House’s concept of “electoral democracy”. Specifically,
Freedom House’s rating judges which countries are considered democratic
and which are not. The evaluation of African, Latin America, East Asian,
and Central/Eastern European democratic experiences are made by using the
index of democratisation, which is based on an annual assessment of polit-
ical rights and civil liberties monitored by Freedom House. Freedom House
has been rating political rights and civil liberties in each country of the world
since the 1970s. It is the most systematic evaluation of the degree of democ-
racy, an evaluation whose validity is judged to be reputable (Bollen, 1993).
Freedom House’s index of democracy has been extensively used in cross-
national research (Burkhart and Lewis-Beck, 1994; Helliwell, 1994). Freedom
House gives every country a rating ranging from 1 to 7 on political rights and
civil liberties, a score of 1 corresponding to a maximum degree of freedom,
and a score of 7 to the minimum degree.5
My interest is directed mainly at the political rights scale, which focuses on
the fairness of elections. I construe a rating of 1 and 2 as reflecting a satisfac-
tory level of democracy; Freedom House itself collapses scores of 1 and 2 into
a single category indicating a free country (Gastil, 1979, p. 24). It should be
noted that in drawing on the Freedom House ratings, I do not make a decision
to include a case solely based on a single election or year, but rather look sys-
tematically at every national election covered by the surveys and then decide
Comparative framework 63
if the country can be judged democratic overall. At the time of writing, out
of 54 countries in Africa, ten countries6 met these criteria. Concerning Latin
America, nine countries could also be classified as newly consolidated dem-
ocracies based on my three criteria mentioned above and Freedom House’s
ratings. Turning to East Asian countries, based on this criterion, out of the
18 East Asian countries, only four countries (Japan,7 South Korea, Mongolia,
and Taiwan) meet the requirement of being fully free and fair, while the
Philippines, Singapore, Indonesia, and Malaysia are only considered partly
free and as such are excluded from this study. Finally, rounding up with post-
communist states, I focus on six countries. Countries like Belarus, Russia,
Serbia, Albania, Kosovo and Bosnia, and Herzegovina were all eliminated
because they were not considered as free by Freedom House ratings, while
Croatia, Slovakia, and Latvia etc., although democratic, were excluded from
our study because of the absence of any survey for this country. The full set
of countries included in this study are reported in Table 3.1.
Also, Table 3.2 displays the political rights for these countries for a period
of 20 years (1999–2019) and the number of elections8 conducted since the
very first elections that were held following the third wave of democratisation.
Of the 26 countries from the four-geo political regions included in this book,
most of the countries (most notably those in post-communist Europe, Latin
America, and, to a lesser extent, those in Africa) have experienced no change
or stable trajectory in their political rights index and only Botswana,9 Poland,
Hungary,10 and Namibia experienced a slight decline.
Regions Countries
Note: Countries selected are based on the adherence to three criteria: countries that have
experienced the end of authoritarian rule; countries that have experienced the installation of
a democratic regime, and countries that are experiencing the consolidation of their democratic
regimes.
Source: Freedom House. The index ranges from 1 (best) to 7 (worst) referring to the degree of
political rights in a country. The number of elections: These are based on the most important or
national election (i.e. for some countries these are parliamentary elections, while for others they
are presidential elections) that have been conducted since the first elections marking the third
wave and are derived from www.idea.int/data-tools/data/voter-turnout. * The number elections
relate to the current wave of democratisation across each region. I focus on the most important
national election. In Latin America, the focus was on presidential elections; in post-Communist
Europe, the focus was on legislative/parliamentary elections. In sub-Saharan Africa, the focus
was on presidential elections and parliamentary elections for South Africa and Botswana cases.
In East Asia, the focus was on presidential/or legislative election.
66 Comparative framework
A and B varies according to the context, C. The method I, therefore, rely on
for this book must take account of the impact of context.
Data
A decade ago, cross-regional comparison of voting behaviour would not
have been possible, as multinational efforts at generating comparable survey
projects are quite recent. (For an overview of cross-national survey projects see
de Vries, 2017; Heath et al., 2005; Kittilson 2007; Norris 2009). Among these
initiatives, I count the Comparative Study of Electoral Systems (CSES), the
World Values Survey (WVS) and the International Social Survey Programme
(ISSP). Unfortunately, I do not use any of these survey data because of the
following reasons.
First, beginning with the Comparative Study of Electoral System
(CSES),11 Module 1 (1996–2001) only included three countries from Latin
America (i.e. Peru, Mexico, and Chile), five countries from East Asia (i.e.
Hong Kong, Japan, South Korea, Taiwan, and Thailand), nine countries in
post-communist Europe (i.e. Ukraine, Russia, Slovenia, Romania, Poland,
Lithuania, Hungary, Czech Republic, and Belarus), with no sub-Saharan
Africa country. CSES’ latest Modules 4 and 5 only saw the addition of two
more countries in Latin America, one East Asia country, and two coun-
tries in sub-Saharan Africa (i.e. South Africa and Kenya). Not only are we
confronted with a limited number of cases, which makes it hard to draw any
kind of inferences across these regions but most of the countries included
(e.g., Thailand, Philippians, Hong Kong, Kenya, Belarus, Ukraine, Russia,
Serbia, Montenegro, Mexico, etc.) cannot be regarded as consolidated dem-
ocracies according to my logic of case selection outlined above (i.e. also see
Chapter 1).
Second, the World Value Survey (WVS)12 is another multinational survey
initiative that was launched in the 1980s and, so far, seven waves have been
conducted, which covers most of the countries from all the four regions.
However, the question about electoral participation is not ideal since it does
not ask directly whether respondents actually voted in the most recent national
elections, but rather asks “When elections take place, do you vote always, usu-
ally or never?”. I argue this question tends to capture the frequency of voting
rather than whether the respondent truly voted in the most recent national
election or not.
Third, the International Social Survey Programme (ISSP)13 is a cross-
national programme established in 1984 and has a common set of questions
for established social survey in Australia, Germany, Great Britain, and the
US. It now covers 42 countries and has collected data annually, through a
core questionnaire. Although, the survey capture attitudes towards economy,
gender, the legal system, socio-demographic as well as variables needed to
Comparative framework 67
study voting behaviour, in terms of geographical coverage, the survey focuses
mainly on Western democracies. For instance, only two sub-Saharan African
and three Latin American countries are included (i.e. South Africa, Tunisia,
Chile, Mexico, and Venezuela).
Based on these shortcomings of the CSES, WVS, and ISSP data, I, there-
fore, rely on four regional datasets across the four regions, which incorporate
almost similar questions wordings for some variables and are therefore appro-
priate for comparing with each other. The Afrobarometer for sub-Saharan
African cases, AmericasBarommeter or Latin American Public Opinion
Project for Latin America sample, the AsianBarometer for East Asian cases,
and the European Social Survey for Central/Eastern European countries.14
These data thereby permit me to see how Africans, Latin Americans, East
Asians, and East European voters have behaved over recent decades with
regards to their decisions to vote or not to vote in elections.
First, the Afrobarometer is an independent, non-partisan research pro-
ject that measures social, political, and economic attitudes and behaviour in
Africa. The surveys have so far been conducted in 36 African countries and
are repeated on a regular cycle. They ask a standard set of questions so that
countries can be systematically compared since trends in public attitudes are
tracked over time. They are designed to generate samples that are nationally
representative across all citizens of voting age and has a sample size ranging
between 1200 and 2400. Second, the AmericasBarommeter15 conducted by
the Latin American Public Opinion Project is a periodic study of 34 countries
in the Western Hemisphere. It consists of a country survey based on national
probability samples of voting-aged adults. The principal objective of the
survey is to advance the accurate description and understanding of public
opinion and behaviour across the Americas. Third, the AsianBarometer is
an applied research program that aims to gauge public opinion on issues
such as political values, democracy, and governance across Asia. The survey
was launched in 2000, and so far four survey rounds have been completed
including 13 East Asian and five South Asian countries and the aim is a
sample size of 1200 per country. Finally, the European Social Survey (ESV) is
an academically driven cross-national survey that has been conducted across
Europe since its establishment in 2001. The survey measures the attitudes,
beliefs, and behaviour patterns of the diverse population in more than 30
nations.16
The beauty of using these four datasets is that most of the survey taps
questions pertaining to electoral or voting behaviour in the various regions.
Thus, these surveys are more appropriate than national election studies for
this book because they provide for the possibility of cross-regional com-
parison of people living across diverse political, economic, and social contexts.
As Curtice (2007) puts it, the advantage of a cross-national survey is that it
provides or enable researchers to assess the empirical generalisability of their
theoretical claims about the causes and consequences of political attitudes
and behaviour. Similarly, it permits one to examine the robustness of their
68 Comparative framework
research findings by replicating their analyses using different independently
collected data sources.
Although the focus of this book is on newly consolidated democracies,
I nevertheless argue that it will be important to examine what is happening in
Western European democracies in order to establish a benchmark for com-
paring newly consolidated democracies. I therefore once again rely on Round
8 data from the European Social Survey, which was conducted in ten different
European countries: Austria, Belgium, Finland, Germany, Ireland, Italy,
Netherlands, Norway, United Kingdom, and Sweden. This offers one signifi-
cant advantage. First, as I have pointed above, it covers ten Western European
countries, and sets a baseline for comparing newly consolidated democracies
and old democracies (i.e. this allow me not only to determine whether but also
by how much new democracies differ from older democracies). Tables A1, A2,
A3, A4 (e.g. see Appendix 1) contains detailed information about the datasets.
First, Table A1 consists of eight sub-Saharan African countries for which
we are interested in exploring which factors determine an individual’s propen-
sity to vote in national elections. The Xs in the table show which Afrobarometer
survey was conducted in a country. This, therefore, means that the absence
of an X signifies that no survey was conducted in a country during that
round. From Table A1, we see that the survey is currently on its sixth round.17
Second, Table A2 shows the number of countries and their distribution in the
different round of the AmericasBarommeter that I am interested in studying.
The first set of the survey was conducted in 2004 in 11 countries; the second
took place in 2006 (consisting of 22 countries across the region); in 2008,
the survey grew and included 24 countries, and in 2010 and 2012, it included
26 countries. Finally, Round 6, the 2014 survey that this book focuses on,
consists of a total of 28 countries.18 Third, Table A3 shows the number of
countries and their distribution in the different wave of the AsianBarometer
that I am interested in studying. All of the cases were included across all the
four waves. Lastly, rounding up, with post-communist European countries,
Table A4 shows the number of countries and their distribution in the different
rounds of the European Social Survey, which I am interested in studying. In
summary, having outlined the datasets that this study will rely on, it is essen-
tial to note that I focus on the most recent rounds or waves (i.e. as it falls
approximately between five and six national elections conducted across these
countries) thus providing valuable perspective on voting behaviour across the
four geo-political regions.
Notes: a: Survey year for the different regions are as follows: 2014–2016 for the Africa cases; 2014
for the Latin America, 2014–2016 for East Asia and 2016 for the post-communist states/Western
European countries. b: Source of official turnout figures: International Institute for Democracy
and Electoral Assistance. c: Source of self-reported turnout figures: Afrobarometer (Round 6,
2016), AmericasBarometer (Wave 6, 2014), AsianBarometer (Wave 4, 2016) and European Social
Survey (Round 8, 2016).
Socio-demographic variables
The main variables from this theoretical perspective include age, gender, reli-
gion, ethnicity/race, place of residence and education.
Demographic variables
First, beginning with demographic variables, age is included in the analysis.
Given that the relationship age and vote choice is non-linear, I find it effi-
cient to keep age as a continuous variable and entered it as a quadratic term.
I expect the variable coefficient to show a positive sign, as older people/citi-
zens should be more likely to vote.
Second, looking at gender, all four surveys have standard questions that
reflect the gender of respondents. For the sake of clarity and ensuring com-
parability across the regions, I recoded gender into a dummy variable, where
men are coded as 1 and women coded as 0 (i.e. women being our reference
category). Again, I expect the effect of the dummy variable to be positive,
indicating men are more likely to participate.
Third, looking at religion, across three regions, this variable is measured by
a question that asks about the frequency of attendance at religious services. For
example, how often do you personally engage in religious practices (Africa),
how often you practise religious services or rituals (East Asia), and how often
do you attend religious services apart from special occasions (post-communist
European)? Considering that I propose that voters who attend religious ser-
vices more frequently/or who are more religious should be more likely to vote,
to maintain comparability across the regions, I included a five-point scale for
Comparative framework 73
church attendance. This variable runs from never or almost never attend a
religious service (0) to attending services more than once a week/every day (5).
However, for the Latin America regions, AmericasBarometer (Wave 6, 2014)26
I did not have any question measuring religion attendance. Thus to account
for the role of religion, I rely on a question that asks respondents: what is your
religion? I created a dichotomous variable, indicating whether a respondent
is Catholic (1) or not (0). Considering that the Latin America region is pre-
dominately Catholic, I, therefore, expect this variable to show a positive coef-
ficient, as Catholic voters should be more likely to vote.
Fourth, another demographic variable I consider is ethnic identity/race.
For the Africa region, I rely on the salience of ethnicity or the weight that
respondents or citizens place on ethnicity as the core identity, usually in com-
parison to other identities.27 The salience of ethnicity is measured by asking
respondents to choose between ethnic identity and national identity. In the
Afrobarometer the question reads as follows: Let us suppose that you had
to choose between being a (Ghanaian/etc.) and being a (respondent’s iden-
tity group), which of these two groups do you feel most strongly attached
to? The response options range across ethnic identity only, ethnic identity
more than national, national/ethnic identity equal, national identity more
than ethnic, and national identity only. To capture the salience of ethnicity
in Africa, I created a three-category variable that identifies those with Only
Ethnic Identity (coded 1), those with Equal Identity (coded 0.5) and those with
Only National Identity (coded 0). Based on the hypothesis, we would expect
voters with ethnic identity to vote more frequently.
For the East Asia cases, I rely on a question in the AsianBarometer that
taps or measure nominal ethnic identity and reads as follows: what is your
racial or ethnic background? Focusing on three East Asian countries (South
Korea, Taiwan, and Mongolia), I have a list of 14 groups across South Korea,
Taiwan, and Mongolia. I, therefore, classify these groups into two comparable
groups coded 1 for the largest ethnic group and 0 for the minority ethnic group.
The majority ethnic groups constituted 20 per cent or more of the national
population (i.e. Korean 33 per cent belonging to South Korea; Khalkh 27.7
per cent belonging to Mongolia and; Minan 24.6 per cent belong to Taiwan).
The minority ethnic group constitute 5 per cent or less of the population
(i.e. this included the following groups: Korean Chinese, Vietnamese, Kazak,
Hakka, Mainlander, Aboriginal, Vietnamese, Indonesian, Overseas Chinese,
Filipino, and others).
Turning to Latin America, relying on the AmericasBarometer, the question
that captures ethnicity reads as follows: do you consider yourself white, mes-
tizo, indigenous, black, mulatto, or of another race? To capture ethnicity,
I created a four-category variable where other ethnic groups are coded 0;
blacks/indigenous people are coded 0.5, mestizos/mulattos coded 1, and whites
coded as 2. Based on the theoretical discussion, I would expect white voters
to be more likely to participate. Rounding up with Central/Eastern European
countries to capture race or ethnicity, I include a dichotomous variable that
74 Comparative framework
measured whether a voter belongs to a minority ethnic group coded 0 and 1
if the respondent is member of the largest ethnic group. Remember, I expect
that individuals from the largest ethnic group would tend to vote more.
Fifth, the last demographic variable I consider is the place of residence
(i.e. urban and rural divide). All four surveys have approximately the same
question words that reflect an individual place of residence. Thus, to account
for the difference in voting behaviour between urban and rural voters, I con-
struct a dummy variable for voter’s place of residence, with voters living in
urban or large cities coded 0 and those living in rural or small towns coded 1.
Socioeconomic variables
Drawing on our previous discussion (i.e. theoretical chapter), two standard
indicators that capture socioeconomic status are income and education.
Unfortunately, in this study, I do not consider respondents’ income as a
measure of voter’s socioeconomic status because of a large number of missing
values and, in addition, because of its problematic measurement in Africa.28
Thus, for the sake of ensuring comparability across the four geo-political
regions, I have therefore omitted income and focused more directly on edu-
cation. In all four surveys, citizens are asked about the highest level of edu-
cation they have completed or achieved. Thus, the different educational
systems and degrees allow me to create three educational categories: pri-
mary/low education (coded 0), secondary/medium education (coded 0.5),
and higher/tertiary education (coded 1). By recoding a respondent’s level of
education into these three categories, a higher value reflects a higher level of
education, which I would expect to be associated with a higher propensity
to vote in elections.
Mobilisation variables
For the mobilisation model, I consider three variables, which are all geared
towards testing this theoretical perspective. These variables include party
identification, civil society organisations/associational membership, and clien-
telism. First, beginning with party identification, I focus on a classic question
that looks at respondent’s closeness to political parties. In the Afrobarometer,
AmericasBarometer, AsianBarometer, and European Social Survey, party
identification is captured by a survey question that asks the respondents if
they feel close to any political party. I collapse the responses into a dummy
variable: no, not close to any party is coded as 0 (reference category), while
yes, feel close to a party is coded as 1. Once more, I expect the effect of the
dummy variable to be positive, indicating party identifiers are more likely to
participate.
Second, civil society organisation/ associational membership is also
included in our model. Concerning the African cases, this variable is measured
Comparative framework 75
by a question that asks about membership in voluntary association or com-
munity group. I recoded this variable into three categories: 0 = Not a member,
0.5 = Inactive member, while Active member/Official leader29 was coded
1. For the Latin America cases, I combine the different question measuring
associational membership in four types of social groups, namely: religious
organisations, parent association, community improvement groups, and pol-
itical movements/parties. Consequently, I construct an additive scale of mem-
bership index that runs from no participation (0) to weekly participation in all
social groups (1). Turning our attention to the AsianBarometer, civil society
organisational membership is measured by a question that asks respondents
if respondents are members of any organisation. I, therefore, construct a
dummy variable with non-members (coded 0) and members (coded 1). For
post-communist countries, membership in voluntary or informal networks is
measured by evaluating a respondent’s membership in trade unions or similar
organisations. Thus, a code of 1 is assigned for voters who declared to be
members and 0 for otherwise. Here again, I expect the coefficient for asso-
ciational membership to be positive as those who declared to be members
would be more likely to participate.
Third, the last mobilisation variable that I examine is clientelism. I rely on
the surveys and include a measure of whether the respondent has received
any gift or promise from a political party to turn out. In the Afrobarometer,
the question measuring clientelism reads as follows: in your opinion, how
often do the following things occur in this country’s elections: Voters are
bribed (see Table B1 in appendix on the variable is coded). While in the
AmericasBarometer, the question measures clientelism through vote-buying
reads as follows: thinking about the last presidential elections, did someone
offer you something, like a favour, gift or any other benefit in return for your
vote or support? Based on these questions, I, therefore, created a dummy
variable with 1 assigned for voters who have been offered a gift in return for
their votes and 0 people with no experience (who never experience receiving
gifts/brides). I expect this variable to take a positive coefficient, suggesting
that voters who were offered gifts/brides would be more likely to participate
However, in the East Asian and post-communist cases, no question focuses
on the clientelism.
Political-psychological variables
A review of the existing literature on electoral participation pinpointed sev-
eral political-psychological variables that I intend to test concerning newly
consolidated democracies. These variables consist of political interest, polit-
ical trust, and political efficacy (internal/external efficacy).
First, starting with political interest, I measure this following the clas-
sical way of asking people how interested they are in politics or public
affairs (Gabriel and Van Deth, 1995, p. 395). In all four regions, political
76 Comparative framework
interest is captured with an almost identical question. Accordingly, polit-
ical interest is measured in a four-point scale that runs from not interested
(0) to very interested (2). Second, turning to political trust, once again, I rely
on a standard survey question that tap respondents’ trust in several political
institutions (president, national assembly, political parties, national police,
etc.). Thus, for the sake of comparability across all regions, I created a polit-
ical trust index that runs from low/no trust to great/high trust.
Third, political efficacy is measured by items intended to capture internal
and external efficacy. Beginning with the Africa region, external efficacy
is measured by a question that taps individuals’ subjective assessments of
whether elections allow voters to remove from office leaders that people
do not want. The external efficacy variable within the African region is
therefore measured using a four-point scale that runs from 0 to 2, with a
low value meaning no or little efficacy and a high value indicating voter
is efficacious. While, regarding internal efficacy, this is measured by a
question which reads as follows: elections ensure voters’ views are reflected
and like the external efficacy indicator, it is measured with a four-point
scale running from 0 to 2. Next for the Latin America cases, I rely on the
question of whether respondents believe those who govern their country
are interested in what people like you think as a measure of external effi-
cacy. The variable is measured on a seven-point scale (1 = strongly dis-
agree to 7 = strongly agree) so that those who declared to strongly disagree
are considered as those with little or no efficacy. While those who strongly
agree are considered as efficacious individuals. Concerning internal effi-
cacy, this variable is measured by a question that taps respondents’ sub-
jective assessment of their understanding of the most important political
issues (i.e. measured on a seven-point scale).
In the East Asian region, external efficacy is measured by a question that
reads as follows: how well do you think the government responds to what
people want? I recoded the different responses into a four-point scale that
runs from not at all responsive (0) to very responsive (2), with a lower score
indicating a voter less efficacious and higher range meaning the voters is effi-
cacious. On the other hand, internal efficacy is captured by a question that
asks respondents if politics/government is so complicated that a person like
me can’t understand. This is equally coded into a four-point scale running
from 0 (strongly agrees politics is complicated) meaning the voter is less effi-
cacious to 2 (strongly disagrees politics is too complicated), which implies the
voter is more efficacious.
Lastly, in post- communist European countries, internal efficacy is
measured by the following questions: how confident are you in your ability to
participate in politics? I maintain the original categories but recoded: Not at
all confident30 coded as 0, A little confident coded 0.5, Quite confident coded
1, Very confident coded as 2 and Completely confident coded 3.31 Rounding
up with external efficacy, I yet again rely on the following question: and how
Comparative framework 77
much would you say that the political system in your country allows people
like you to have an influence on politics? Recoded 0 = Not at all; 0.5 = Very
little; 1 = Some; 2 = A lot; 3 = A great deal.
Alternatives variables
To test an alternative model of vote choice in newly consolidated democracies,
I operationalise a handful of variables, including the economy, corruption,
and authoritarianism. First, the country’s economic variable is captured in
the African, Latin American, and East Asian regions as follows: looking
back, how do you rate the following compared to 12 months ago: economic
conditions in this country? Or, how would you describe the change in the
economic condition of our country over the last years? I, therefore, collapse
the different response options into three categories: worse (0), same (0.5), and
better (1). Turning to the Eastern European cases, this question is captured
as follows: how satisfied are you with the present state of the economy in-
country? This is measured in a ten-point scale from extremely dissatisfied with
the country’s economic situation (0) to extremely satisfied with one’s country
economic situation (10).32
Second, I include a corruption variable that taps people’s perception or
experience with corruption in the different regions. Beginning with sub-
Saharan Africa, corruption is measured based on an additive index of seven
similarly worded questions about the respondents’ perception of corruption
in their country. The question reads as follows: how many of the following
people do you think are involved in corruption, or haven’t you heard enough
about them to say: (a) office of the presidency, (b) members of parlia-
ment, (c) government officials, (d) local government councillors, (e) police,
(f) judges and magistrates, and (g) business executives. The response options
were: 0 = None, 1 = Some of them, 2 = Most of them, 3 = All of them.
Moving to the Latin America region, the question about corruption reads
as follows: is corruption among public officials very common? The response
options are 1 = very common, 2 = common, 3 = uncommon, 4 = very
common. Due to the small response rate across the four categories, I combine
the different categories and therefore created a dummy variable with 3 and 4
coded as 0 (corruption uncommon), while 1 and 2 were coded as 1 (corruption
common). For the East Asia cases, the question that measures corruption
reads as follows: Have you or anyone you know personally witnessed an act
of corruption or bribe-taking by a politician or government official in the past
year? 1 = Personally witnessed, 2 = Told about it by a family member who
personally witnessed, 3 = Told about it by a friend who personally witnessed,
4 = Personally never witnessed, 5 = No one I know has personally witnessed.
Relying on these different categories, I recoded people’s experience with
corruption into a dummy variable coded 1 = having witness corruption and
0 = never witness corruption. Rounding up with post-communist European
78 Comparative framework
countries, our measure of corruption is derived from the Transparency
International corruption perceptions index (i.e. as no question in the survey
measure people’s experience/perception of corruption at the individual-level).
The index measured the perceived level of public/political corruption on a
scale of 0 (highly corrupt) to 100 (very clean).33
Third, our measure of authoritarianism is derived from the democratic
erosion event dataset34 that captures the symptoms of democratic erosion (i.e.
media repressions, curtailed of civil liberties, repression of the opposition,
reduction in judicial independence, etc.) across countries/regions over time.
Thus, relying on expert coding, the question measuring democratic erosion
asks: what is the overall erosion of a country’s democracy? Based on a four-
point scale the response ranges from 0 = No backsliding, and weak threat
of future backsliding; 1 = There are precursors to backsliding (e.g. the rise
of extremist parties, but erosion of democratic institutions has not yet taken
place); 2 = There is weak erosion of democratic institutions; 3 = There is
moderate erosion of democratic institutions; 4 = There is a severe erosion
of democratic institutions. Drawing on this data, I recoded authoritarianism
into a dummy variable where 0 and 1 are combined and coded as 0 (i.e. coun-
tries not experiencing democratic erosion). Note, for the countries we analyse
here no country was placed under the 4 category (i.e. 4 = there is a severe
erosion of democratic institutions). So, 2 and 3 are combined and coded as 1
(i.e. country experiencing democratic erosion).
Notes
1 This period can be described to have ushered in an unprecedented political change
and reforms across these regions.
2 As cited by Brigevich (2008, p. 1) the collapse of the Soviet Union in the 1990s
discredited the viability of authoritarian regimes and one-party systems in Central
and Eastern Europe, Latin America, parts of Asia and sub-Saharan Africa, and
replacing them with governments determined to enact pro- democratic, liberal
reforms.
Comparative framework 81
3 This book draws upon four different cross-national public surveys.
4 Harkness et al. (2003) outline two strategies to established equivalence by
adocating first whether constructs are, in fact, comparable and second whether the
instruments that are useful in one context are equally so in another.
5 https://freedomhouse.org/report/methodology-freedom-world-2018 [Accessed 18/
06/2019].
6 Tunisia is, however, excluded since data for this country were only collected three
years after Tunisia democratised. Also, although Senegal does qualify as a newly
consolidated democracies, it is, however, excluded as no question relating to voter
turnout was ask in the country.
7 I do not considered Japan as a new democracy and, therefore, my analyses in the
subsequent chapter will exclude Japan.
8 Here, I refer to only national elections that could be presidential or parliamentary
elections depending on the country.
9 Though Botswana is considered a stable democracy that has been dominated by
a single party since independence. The decline in its political rights index should
probably be down to tight control over media freedom and the discrimination of
migrants, refugees, and LGBT groups.
10 Hungary experienced a decline in its political right index due to increasing intimi-
dation of civil society groups and the opposition that has largely being orchestrated
since the Prime Minister Viktor Orbán’s Alliance of Young Democrats–Hungarian
Civic Union (Fidesz) took over power.
11 www.cses.org/datacenter/download.htm [Accessed 18/06/2019].
12 www.worldvaluessurvey.org/WVSContents.jsp [Accessed 18/06/2019].
13 http://w.issp.org/about-issp/ [Accessed 27/03/2019].
14 The four-survey data across the regions do not contain the same question wording
for all independent variables, and so they cannot be pooled into a single dataset
for analysis. Rather, I analyse each dataset separately, while trying to keep the
variables as similar as possible given that some variables have different questions
wordings.
15 The AmericasBarometer is coordinated by a team at Vanderbilt University, which
hosts the Latin American Public Opinion Project (LAPOP).
16 More technical information on the Afrobarometer, Latinobarometer,
AsianBarometer, and European Social Survey, can be obtained from their
websites: www.afrobarometer.org/survey-and-methods/samplingprinciples, www.
vanderbilt.edu/lapop/about-americasbarometer.php, www.asianbarometer.org/
survey/survey-countries, www.europeansocialsurvey.org/methodology/ess_meth-
odology/sampling. [Accessed 16/03/2019].
17 Important to note though, Round 7 (2016/2017) was completed and released
probably by the end 2020.
18 However, as previously specified in our case selection, we focus only on Latin
America countries that classify as free or as liberal democracies.
19 By summarising these materials here, it becomes possible to reduce the amount of
methodological justification offered in each of the individual chapter.
20 This problem arises as people tend to overestimate the turnout in surveys.
21 Turnout intention is likely to be even more prone to social desirability bias and is,
therefore, not often used.
82 Comparative framework
22 In their meta-analysis of individual level research on voter turnout, Smets
and van Ham (2013) state that 82 per cent of the studies measured turnout as
reported turnout, 11 per cent as validated turnout and 7 per cent as turnout
intention.
23 The question wording from the four surveys is reported in Appendix 2 (Table
B1 and B2), while the description of variables is reported in Table B3 (i.e.
Appendix 2).
24 Wolfinger and Rosenstone (1980, p. 115) argue that the discrepancy between offi-
cial turnout and reported turnout in surveys can be explained by misreporting
of respondents who are reluctant to admit they did not vote because it is more
socially acceptable to be fulfilling one’s duty as a citizen by voting. Also, it may be
that the surveys are not randomly conducted and were therefore biased towards
voters. Curtice (2007, p. 902) has strongly criticised the survey (i.e. CSES survey)
for not implementing random sampling. Lastly, spoiled ballots may not have
been counted in the official totals. Finally, with respect to sub-Saharan Africa,
Isaksson (2010, p. 6) argues that the discrepancy between reported turnout in
survey and official/actual turnout percentages may arise because voting survey
questions simply ask respondents whether they voted in the last election or
not without taking into consideration whether the elections were concurrently
staged or not. For example, in a few cases like Ghana where the presidential and
parliamentary elections were held concurrently, the Afrobarometer does not tell
us with precision which of the elections the respondents voted for in the last
election.
25 The precise questions wording in the surveys across the four regions are presented
in Appendix 2 (e.g. Table B2).
26 This is because I have access only to the free or reduced version of the
AmericasBarometer data.
27 For an overview see; Bhavnani and Miodownik, 2009; Eiffert et al., 2010.
28 In the African region, in the Afrobarometer there is no question that captures
a respondent’s income, unfortunately. Previous studies (Bratton et al., 2005;
Isaksson, 2014; Kuenzi and Lambright, 2011; Tambe, 2017a), have simply used a
question in the Afrobarometer about access to food as a proxy for income (Over
the past years, how often, if ever, have you or your family gone without food?).
However, Maydom (2017) caution about the use of access to food (e.g. how often
respondents have gone without food) as a proxy to income as it might be capturing
other factors other than income. Relatedly, I would argue that trying to measure
voters’ socioeconomic status using access to food is very unconvincing given that
almost 80 per cent of individuals in these countries depend on agriculture or
farming for their basic subsistence. Moreover, in these countries it is a common
tendency for people to borrow from their neighbours if they cannot afford food
for themselves.
29 The response rate for both categories was small, thus enabling us to combine both.
30 “Not at all confident” in the sense of “having no confidence at all in your own
ability”.
31 A low value or score means little or efficacy, while a higher value means more
efficacy.
Comparative framework 83
32 Since the mid-point of the scale lies between 5 and 6 and for the sake of compar-
ability across other regions, we, therefore, collapse the response options into three
categories: 0–4 on the scale was assigned as worse (0); 5–6 was coded as same (0.5);
while those in the top half (7 to 10 on the scale) was assigned better (1).
33 To ensure comparability with the other regions, the score was rescaled from 0 to 1
34 www.democratic-erosion.com/ [Accessed 13/04/2019].
35 For details see Gallagher (1991); Gallagher and Mitchell (2005).
4
Socio-demographics and the vote
Building on the very first studies of voting and or electoral behaviour (i.e.
Lazarsfeld, Berelson and Gaudet 1948; Berelson, Lazarsfeld and Mcphee,
1954; Campbell et al., 1960), scholarly literature on voting from older dem-
ocracies from the global North1 (i.e. Western Europe and North America)
gave enormous importance to voters’ socio- demographic or descriptive
characteristics in predicting individual-level turnout. Thus, in this chapter,
my examination of why people vote in newly consolidated democracies starts
by looking at the effect of six key socio-demographic indicators (e.g. age,
gender, ethnicity/race, religion, place of residence, and education) on vote
choice. Recall that while the primary interest lies in explaining the voting deci-
sion in newly consolidated democracies, I include Western European cases so
as to compare the experiences of people’s voting decisions in sub-Saharan
Africa, Latin America, East Asia, and Eastern European countries. That is,
are the socio-demographic factors that influence voters’ decisions to turn
out in elections in these newly consolidated democracies broadly similar to
those that we find in the established democracies of Western Europe? Having
identified the appropriate empirical indicators, I now move on to empirical
analysis, where I report bivariate and multivariate associations between the
independent variables (i.e. socio-demographics) and the dependent variable
(voter turnout).
Notes: Bivariate results between voting and the socio-demographic variables across the regions
are derived from the European Social Survey, Afrobarometer, AmericasBarometer, and the
AsianBarometer. The observations are all weighted using the different surveys weights. Notes:
**Correlation is significant at the 0.01 level (2-tailed); *Correlation is significant at the 0.05 level
(2-tailed).
Socio-demographics and the vote 87
likely to turn out. For East Asia, the relationship between ethnicity and vote
choice is non-significant. Rounding up with sub-Saharan African democra-
cies, our data for the African cases show that there is a relationship between
ethnic identity and voting, but the result is surprising, revealing that voting in
Africa is not based on ethnic considerations since respondents or citizens with
ethnic identities were less likely to have voted when compared to those with
simple national identities. In the face of these findings, I, therefore, conclude
post-communist, and Latin America democracies do resemble older democ-
racies, but East Asia and African democracies are different.
Fifth, if we turn to the effect of an individual place of residence, the prop-
osition was that citizens living in rural areas would be more likely to partici-
pate. That said, looking at the preliminary bivariate results suggest this is true
only for African cases while for the remaining regions (Latin America, Eastern
Europe, and East Asia) the relationship between voting and whether an indi-
vidual resides in a rural or urban area is non-significant. Finally, concluding
with socioeconomic variables (i.e. education), while the result tells of a signifi-
cant relationship between voting and education in Eastern European democ-
racies (i.e. similar in Western Europe), for East Asia and Latin America our
data indicate a non-effect between education and turnout. Rounding up with
the Africa region, the results are even more astonishing given that those with a
higher level of education were less likely to vote.
So far, using bivariate statistics, I have now tested six different socio-
demographic groups of factors that I theorise will influence people’s decision
to vote across newly consolidated democracies in four geopolitical regions.
The bivariate analysis suggests some relationship (although very limited in
strength at least at the bivariate level) and quite a lot of agreement across the
regions but with some inconsistency too. Thus, to be more confident of this,
and to explore the sources of variation more rigorously, one needs to engage
with multivariate analysis of vote choice, including all six socio-demographic
characteristics.
Table 4.2 reports the multivariate analyses in which all of the socio-
demographic variables are added simultaneously. This multivariate result is
remarkable because it will enable us to re-establish which of the variables will
continue to exert a positive and significant relationship with voting while con-
trolling for all of the others. The first model (Model 1) is the baseline model,
which includes a month’s variable (month since the last election) and country
dummies for each region.4 Thus, as a first step, I start by regressing vote
choice on months since the previous election and the country’s dichotomous
variables. This result shows that months since the last election has a negative
or no effect on vote choice in newer democracies of Eastern Europe, East
Asia, and Latin America except for sub-Saharan Africa (as similar in Western
European democracies). Next, the second column (Model 2, Table 4.2) shows
the socio-demographic variables of greater interest. This model presents the
standardised coefficient from the multivariate analysis of vote choice, with
standard errors in parentheses. In addition to the socio-demographic variables,
it includes the month variable and the country dummies for each region. Also,
the inclusion of the socio-demographic variables shows a marked improve-
ment in the model fit from the baseline model (i.e. Model 1). For example,
in Latin America, East Asia, and post-Communist European countries as
in Western European democracies the Nagelkerke pseudo R2 increases from
0.172, 0.136, and 0.146 respectively except for the sub-Saharan African region
that shows a low Nagelkerke pseudo R2 of 0.068. The logit models for each of
the geographical regions are interested in many regards:
First, starting with age, as expected, there is a highly significant overall
effect of age on voting in newly consolidated democracies of Africa, Latin
America, East Asia, and post-Communist European countries (as is gener-
ally true of established Western European democracies). Compared to young
people, older voters are significantly more likely to vote.5 These results do
not only strongly collaborate single country studies from the different regions
(Tambe, 2017a; Resnick and Casales, 2014; Kuenzi and Lambright, 2011;
Isaksson, 2010; 2014 (for Africa); Carrera and Castañeda-Angarita, 2014;
Seligson et al., 1995 (Latin America); Bernhagen and March 2007; Barnes,
2006 (post-communist Europe); Chung-li Wu and Tzu-Ping Liu, 2017; Mo,
Socio-demographics and the vote 89
Table 4.2 Logistic regression of socio-
demographic model and voting in newly
consolidated democracies
Regions/variables
Sub-Saharan Africa
Months since last election 0.01*** (0.00) 0.02*** (0.00)
Age 0.38*** (0.04)
Gender (1 = men) -0.02 (0.06)
Religion (2 = Everyday/once a week; 0 = Never) 0.08*** (0.03)
Salience of ethnicity (1 = Only ethnic ID; Equal -0.24** (0.10)
national/ethnic ID = 0.5; Only national
ID = 0)
Residence (1 = Rural) 0.28*** (0.07)
Education -0.07 (0.12)
Constant 1.62*** (0.35) 0.45*** (0.16)
Nagelkerke pseudo R2 0.029 0.068
% correctly classified 80% 82%
Number of observations/cases 10969/08 10023/08
Latin America
Months since last election -0.05*** (0.01) -0.03*** (0.01)
Age 0.05*** (0.00)
Gender (1 = men) -0.12*** (0.05)
Religion (Catholics = 1; others 0) 0.23*** (0.05)
Ethnicity/race (Whites = 2; Mestizo/ -0.02 (0.04)
Mulattos = 1; Blacks/Indigenous = 0.5;
others = 0)
Residence (1 = Rural) 0.19*** (0.05)
Education 0.73*** (0.07)
Constant 3.18*** (0.33) -0.04 (0.37)
Nagelkerke pseudo R2 0.026 0.172
% correctly classified 76% 78%
Number of observations/cases 13609/09 12978/09
East Asia
Months since last election -0.04***(0.01) -0.09*** (0.01)
Age 0.06*** (0.00)
Gender -0.07 (0.09)
(continued)
90 Socio-demographics and the vote
Table 4.2 Cont.
Regions/variables
Source: European Social Survey (Western and Eastern European countries); Afrobarometer (Sub-
Saharan Africa); AmericasBarometer (Latin America); AsianBarometer (East Asia). The data
are weighted using the weights provided for each survey data. Country dummies are included for
each region but not shown. For Western Europe ten country election dummies included with the
United Kingdom as the reference category; for Africa 08 included with South Africa as the refer-
ence category; for Latin America, 08 included with Brazil as the reference case; For East Asia, 03
included with South Korea as the reference category; For Eastern European countries, 06 country
dummies are included with Poland as the reference case. Significance: ***p < 0.001; **p < 0.01;
*p < 0.05. Entries are standardized logistic regression coefficients, standard errors in parentheses.
Brady and Ro, 1991; Park, 2002; Denton, 2016 (East Asia)) but even so con-
firm the assertion that in Western democracies, most of the tests between age
and turnout are successful/indicating a positive effect of age (Smets and van
Ham, 2013).
Second, concerning gender, our results show gender is not a significant pre-
dictor of turnout in new democracies of sub-Saharan Africa and East Asia
(as in Western European democracies) except for post-Communist European
states and Latin America. In post-Communist European democracies, men
are 24 percentage points more likely to vote than women. However, a closer
look at the parameter estimates across the other newer democracies tells of
negative sign, and when gender is found to be significant (e.g. Latin America),
it is usually women who turn to vote at higher rates than men. (for similar
Socio-demographics and the vote 91
results see e.g. Only the work of: Carreras and Castañeda-Angarita (2014) as
it focus just on Latin America).
Third, looking at religion, there is an overall significant and positive rela-
tionship (i.e. in the direction expected by our hypothesis) between being reli-
gious and voting in newly consolidated democracies of Africa, Latin America,
East Asia, and Eastern Europe (which is similar to Western European dem-
ocracies). Given this result is remarkably consistent across all newer democ-
racies, it, therefore, confirms Smets and van Ham’s (2013) meta-analyses in
Western established democracies that attendance of religious services is found
to affect individual turnout roughly half of the time, with most studies having
a success rate that lies between 50 and 57 per cent.
Fourth, looking at ethnicity/race, just like the bivariate analysis, the result
from the multivariate analyses suggest the findings are quite nuanced across
the different regions. First, the results between ethnicity and voting across
two newer democracies (East Europe and Africa) tells of a significant rela-
tionship that is similar to those of Western Europe. However, the parameters
estimate reveals that voters in Eastern Europe (similar to Western Europe)
who belong to the majority ethnic group tend to vote at higher rates than
those of ethnic minorities. In sub-Saharan Africa, the result is significant but
negative revealing that voting in Africa is not based on ethnic considerations
as citizens with ethnic identities were less likely to have voted when compared
to those with simple national identities. Concerning Latin America and East
Asia cases, surprisingly, our multivariate results disconfirm the bivariate
results in that the relationship between ethnicity/race is non-existent across
both regions.
Fifth, the effect of an individual place of residence (i.e. Urban vs Rural)
equally produces some mixed results. For example, the results for sub-Saharan
Africa, Latin America, and East Asia (i.e. which is similar to those of Western
European democracies) are in line with those of Bratton, Chu, and Lagos
(2010); Carreras and Castañeda-Angarita (2014); Resnick and Casale (2014)
indicating voters in rural areas are significantly more likely to turn out at the
polls compared to their urban dwellers. However, concerning newer democra-
cies of post-communist European countries, the results suggest the relation-
ship between individuals’ place of residence and whether they voted in the last
national election is non-significant.
Sixth and finally, concluding with the core socioeconomic variable (educa-
tion), as expected education has a significant and positive relationship with
vote choice in three newer democracies: Latin America, East Asia,6 and post-
communist European countries, with the exception being sub-Saharan Africa,
where the link appears to be reversed.7 To be precise, the positive and signifi-
cant effect of education on vote choice is more in line with those of Western
European democracies. From this, one can infer considerable support for
the fundamental proposition that a higher level of education does help to
reduce the cognitive cost of voting. These cognitive costs such as registration
procedures, the ballot structure, filling out forms and even meeting deadlines
92 Socio-demographics and the vote
may depress the participation rate of poorly educated citizens who may find it
hard to understand the bureaucratic hurdles of voting, while the participation
rates of citizens who are highly educated may not be affected as they tend to
understand voting procedures more easily (see, Gallego, 2015).
Having concluded with the multivariate analyses, and recall the central
objective lies in knowing how far Western generated theories could travel to
newer consolidated democracies of Africa, Latin America, East Asia, and
post-communist countries, Table 4.3 provides a tabular presentation of a the-
oretical matching of the socio-demographic model on vote choice based on
the results obtained from the regions.
Thus far, the multivariate analyses presented in Table 4.2 provide us with
the real test of which socio-demographic variables matter for vote choice in
newly consolidated democracies. Undoubtedly, no matter how exciting these
results are, I argued the pooled analyses for each of the regions presented
might tend to hide potentially essential differences between the different coun-
tries considered in the analyses. To be accurate, it can be possible that across
the different regions, individual countries may turn to vary with regards to
different socio-demographic characteristics. For example, in sub- Saharan
Africa and or Latin America, some countries might have a larger rural popu-
lation, while others may have a large urban population. Relatedly, the ethnic
composition in some countries/regions (i.e. East Asia and Latin America)
Socio-demographic model
Age + Applies*** Applies*** Applies*** Applies*** Applies***
Gender (male) + Does not Does not Applies** Does not Applies***
apply apply apply
Religion + Applies*** Applies*** Applies*** Applies*** Applies***
Ethnicity/race + Applies*** Applies** Does not Does not Applies***
apply apply
Residence (rural) + Applies*** Applies*** Applies*** Applies*** Does not
apply
Education + Applies*** Does not Applies*** Applies*** Applies***
apply
Conclusion
In this chapter, using bivariate and multivariate statistics, I have tested six
socio-demographics or descriptive characteristics on vote choice across four
geopolitical regions of the globe. Overall these results suggest quite a lot of
agreement across newly consolidated democracies but some inconsistency too.
To be specific, while the effect for some socio-demographics predictors of vote
choice across the four regions is more uniform (i.e. age, religion, education),
for others (gender, ethnicity/race, place of residence) there is less consistency.
Also, based on the region by region analysis derived from the pooled data, and
as reported in Table 4.3 (i.e. theoretical matching of the socio-demographic
model across newer democracies), one can make two inferences:
First, the analysis finds or suggests the transcendent importance of three
socio-demographic variables (age, religion, and education) on vote choice. To
be precise, the effect of age, religion, and education are not only very strong in
newer consolidated democracies but are largely similar in older democracies
of Western Europe. This would suggest that the socio-demographic models
that have been developed in the first of these types of countries (older democ-
racies) to explain why people choose to vote are general in many instances and
therefore have wide applications in democracies. Second, based on the value
of Nagelkerke pseudo R2 that shows the overall model fit when the socio-
demographic variables are added to the vote choice model, it would suffice to
say that socio-demographic model are strong determinants of vote choice in
Latin America, post-Communist European countries, and to a lesser extent
in East Asia democracies. In these three regions (as in Western democracies)
age, religiosity, and education are the most influential variables on electoral
participation. However, for sub-Saharan Africa, I cannot say that the socio-
demographic predictors are generally powerful determinants of vote choice.
Notes
1 See Lewis-Beck et al., (2008); Leighley and Nagler (2014) for a recent analysis on
how socio-demographics shapes individuals voting decisions.
2 Pearson’s P is not a proper correlation measure given that the main dependent vari-
able (voter turnout) is not continuous. In this context Spearman’s rho or Kendall
tau are more appropriate choices.
3 I thank an anonymous reviewer for pointing this point out as ethnicity tends to
interact with urban–rural residence in many countries while religiosity interacts
with gender.
96 Socio-demographics and the vote
4 Although I include country dummies across all regions, except for models 6 and 7,
we do not report estimated coefficient and standard errors in order to save space.
5 The odds ratios indicate that older voters, when compared to younger voters, are
3 per cent more likely to vote in Eastern and Western European democracies, 6
per cent more likely to vote in East Asian, 7 per cent more likely to vote in Latin
America and almost 46 per cent more likely to vote in sub-Saharan Africa.
6 This finding is in line with a more recent country study that focuses on Taiwan (e.g.
see Achen and Wang, 2017). However, it goes against previous studies by Tambe
(2016) and Bratton et al. (2010) that are more comparative (countries covered ranges
between 7 and 8) and includes both liberal and electoral authoritarian regimes.
7 See for example see Mattes and Mughogho (2010) for similar results that indicates
educated Africans are not the most likely to vote.
8 Although some of these contextual or country characteristics will be addressed
more fully in Chapter 8, it is still useful to highlight some of the difference here.
5
Mobilisation and the vote
Sub-Saharan Africa
Benin N/A 0.41
Botswana 0.75 0.73
Cape Verde N/A 0.40
Ghana 0.67 0.66
Mauritius N/A 0.23
Namibia 0.71 0.79
Sao Tome and Principe N/A 0.63
South Africa 0.45 0.76
Latin America
Argentina 0.24 0.25
Brazil 0.25 0.22
Chile 0.21 0.12
Costa Rica 0.30 0.67
El-Salvador 0.40 0.44
Panama 0.32 0.45
Peru 0.19 0.19
Guyana 0.17 0.14
Uruguay 0.50 0.60
East Asia
Taiwan 0.54 0.48
Mongolia 0.89 0.81
South Korea 0.73 0.45
Post-communist Europe
Czech Republic 0.47 0.36
Estonia N/A 0.40
Hungary 0.59 0.43
Lithuania N/A 0.26
Poland 0.29 0.30
Slovenia 0.36 0.35
Western Europe
Austria 0.55 0.47
Belgium 0.48 0.48
Germany 0.46 0.55
Finland 0.55 0.56
United Kingdom 0.51 0.54
Ireland 0.47 0.32
Italy 0.44 0.37
Netherlands 0.59 0.54
Norway 0.60 0.69
Sweden 0.68 0.64
Regional total
Sub-Saharan Africa 0.54 0.59
Latin America 0.31 0.35
(continued)
100 Mobilisation and the vote
Table 5.1 Cont.
Notes: Entries display the mean percentage of party identifiers in individual countries and overall
regional average. This is based on the earliest and most recent survey data across the different
regions. (i.e. Appendix 1 for information on earliest/most recent survey data). N/A: indicates the
country was a participating nation during the particular survey wave/round.
This result confirms party identification has been quite stable in established
Western democracies (i.e. amidst only a 2 per cent decline), which is not so
dramatic as proponents of party decline would want to believe. Second, for
post-Communist countries, using the same data and focusing on only four
countries (Czechia, Hungary, Poland, and Slovenia) included in the 2002
survey, our data suggest some 42 per cent of the citizenry reported being close
to a political party. Comparing the result with the 2016 survey (ESS, Round
8) indicates that only 36 per cent of citizens were revealed to be party identifiers
(i.e. 6 per cent decline party ID within 14 years). Overall, for post-communist
European countries, I can confirm party identification has remained relatively
low. Third, moving to sub-Saharan African, relying on the Afrobarometer
data (AB, Round 1) conducted between 1999 and 2001, across 12 countries4 54
per cent of Africans reported being party identifiers. Comparing party identi-
fication levels using a recent round of the Afrobaromter (Round 6, 2016) for
36 countries5 reports 59 per cent of Africans were declared to be members of
political parties. Fourth, looking at Latin America and relying on the second
wave of AmericasBarometer (LAPOP, 2008) for 16 Latin American coun-
tries6 reveals that 32 per cent of people were declared to be party identifiers.
Comparing this party identification level with a more recent survey conducted
in 2014 (Wave 6) reveals 36 per cent of respondents were declared to be iden-
tified with a political party, which is quite lower than what has been observed
in older established democracies of Western Europe. Rounding up with East
Asian cases, data from the first and most recent of the AsianBarometer survey
(i.e. Wave 1 and 4) conducted in 2001 and 2014 composed of 7 and 13 East
Asian countries suggest that 56 and 58 per cent of people report themselves
to be close to political parties.
Drawing from these party identification trends across all regions, we can
make three concluding remarks. First, the global trend in party identification
suggests that across the old and new democracies, the percentage of people
reporting ties to political parties is higher in Western Europe but surprising
too in sub-Saharan Africa, and East Asia. At the same time, it is lower in
Mobilisation and the vote 101
post-communist Europe and Latin America. Second, while party identifica-
tion in Western Europe has been stable and declining for post-communist
Europe and Latin America, this has not been the case in other new democra-
cies, especially sub-Saharan Africa and East Asia. In these two regions, our
data suggest a surge in the number of people who report to be identified with
or close to political parties. Third, there seems to be a considerable variation
in levels of party identification across the different countries in each region.
That is, while some countries have experienced an increase in party identifiers,
others are experiencing a decline (i.e. this is even more evident across the Latin
America cases).
Source: Bivariate results between voting and the mobilisation variables across the regions are derived
from the European Social Survey, Afrobarometer, AmericasBarometer, and the AsianBarometer.
The observations are all weighted using the different surveys weights. Notes: **Correlation is sig-
nificant at the 0.01 level (2-tailed); *Correlation is significant at the 0.05 level (2-tailed).
Regions/variables
Months since last election 0.20*** (0.02) 0.19*** (0.02) 0.15*** (0.02)
Age 0.03*** (0.00) 0.02*** (0.00)
Gender (1 = men) 0.07 (0.04) -0.02 (0.04)
Religion (5 = everyday/once a 0.12*** (0.01) 0.10*** (0.01)
week; 0 = never)
Nominal ethnic identity 0.96*** (0.09) 0.91*** (0.09)
(Largest ethnic group = 1;
Minority ethnic group = 0)
Residence (1 = Rural) 0.15*** (0.04) 0.16*** (0.04)
Education 0.56*** (0.03) 0.49*** (0.03)
Party Identification (1 = party 1.13*** (0.05)
ties)
Associational membership 0.65*** (0.06)
(1 = member currently;
0.5 = member previously;
0 = not a member)
Clientelism N/A
Constant -2.96*** (0.38) -5.74*** (0.42) -5.11*** (0.44)
Nagelkerke pseudo R2 0.044 0.146 0.209
% correctly classified 83% 83% 84%
Number of observations/cases 18657/10 18332/10 17962/10
Sub-Saharan Africa
Months since last election 0.01*** (0.00) 0.02*** (0.00) 0.02*** (0.01)
Age 0.38*** (0.04) 0.35*** (0.05)
Gender (1 = men) -0.02 (0.06) -0.11 (0.07)
Religion (5 = more than once 0.08*** (0.03) 0.07** (0.03)
a day; 0 = never)
Salience of ethnicity (1 = only -0.24** (0.10) -0.18 (0.11)
ethnic ID; Equal national/
ethnic ID = 0.5; Only
national ID = 0)
Residence (1 = rural) 0.28*** (0.07) 0.20** (0.08)
Education -0.07 (0.12) 0.01 (0.13)
Party Identification (1 = party 0.76*** (0.08)
ties)
Associational membership 0.56*** (0.11)
(1 = active member/official
leader; 0.5 = inactive
member; 0 = not a member)
Clientelism (1 = receive gifts/ -0.24** (0.09)
bribes)
Constant 1.62*** (0.35) 0.45*** (0.16) -0.43 (0.65)
Nagelkerke pseudo R2 0.029 0.068 0.098
% correctly classified 80% 82% 82%
Number of observations/cases 10969/08 10423/08 10023/08
(continued)
104 Mobilisation and the vote
Table 5.3 Cont.
Regions/variables
Latin America
Months since last election -0.05*** (0.01) -0.03*** (0.01) -0.03*** (0.01)
Age 0.05*** (0.00) 0.05*** (0.00)
Gender (1 = men) -0.12** (0.05) -0.11** (0.05)
Religion (Catholics = 1; 0.23*** (0.05) 0.25*** (0.05)
others 0)
Ethnicity (Whites = 2; -0.02 (0.04) 0.01 (0.04)
Mestizo/Mulattos = 1;
Blacks/Indigenous = 0.5;
Others = 0)
Residence (1 = rural) 0.19*** (0.05) 0.15** (0.06)
Education 0.73*** (0.07) 0.68*** (0.07)
Party Identification (1 = party 0.95*** (0.06)
ties)
Associational membership 0.22*** (0.03)
(1 = member; 0 = not a
member)
Clientelism (1 = receive gifts/ 0.18 (0.13)
bribes)
Constant 3.18*** (0.33) -0.04 (0.37) 0.21 (0.38)
Nagelkerke pseudo R2 0.026 0.172 0.213
% correctly classified 76% 78% 80%
Number of observations/cases 13609/09 12978/09 12469/09
East Asia
Months since last election -0.04*** (0.01) -0.09*** (0.01) -0.04*** (0.01)
Age 0.06*** (0.00) 0.06*** (0.00)
Gender -0.07 (0.09) -0.06 (0.09)
Religion (5 = several times a 0.12*** (0.04) 0.11** (0.04)
day; 0 = never)
Nominal Ethnic ID (Largest -0.15 (0.15) -0.08 (0.15)
ethnic group = 1; Minority
ethnic group = 0)
Residence (1 = Rural) 0.25* (0.13) 0.35** (0.14)
Education 0.44*** (0.06) 0.42*** (0.07)
Party identification (1 = party 1.05*** (0.11)
ties)
Associational membership -0.18 (0.10)
(1 = member; 0 = not a
member)
Clientelism N/A
Constant 2.65*** (0.24) -0.65 (0.36) -0.66** (0.39)
Nagelkerke pseudo R2 0.009 0.136 0.175
% correctly classified 86% 86% 86%
Number of observations/cases 3956/03 3728/03 3639/03
Post-communist Europe
Months since last election -0.01 (0.01) 0.00 (0.01) -0.01 (0.01)
Age 0.03*** (0.00) 0.02*** (0.00)
Gender (1 = men) 0.22*** (0.05) 0.09* (0.05)
Mobilisation and the vote 105
Table 5.3 Cont.
Regions/variables
Source: European Social Survey (Western and Eastern European countries); Afrobarometer (Sub-
Saharan Africa); AmericasBarometer (Latin America); AsianBarometer (East Asia). The data
are weighted using the weights provided for each survey data. Country dummies are included for
each region but not shown. For Western Europe, ten country election dummies are included with
the United Kingdom as the reference category; for Africa, 08 countries are included with South
Africa as the reference category; for Latin America, 03 countries are included with Brazil as the
reference case; for East Asia, 09 countries are included with South Korea as the reference case;
for Eastern European countries, 06 country dummies are included with Poland as the reference
case. Significance: ***p < 0.001; **p < 0.01; *p < 0.05. Entries are standardized logistic regres-
sion coefficients, standard errors in parentheses.
Recall that column 3; Model II takes into consideration only the effect of
the socio-demographic variables (age, gender, religion, ethnicity, residence,
and education) on voting across the four regions. That said, this socio-
demographic model is, of course, effectively the same as the one previously
described in Table 4.2 (see, Chapter 4). In column 4, Model III, I proceed by
adding mobilisation variables into the model that already consists of socio-
demographic variables (see, Table C3, Appendix 3, for descriptive results for
the mobilisation variables).
First, I examine the contribution of the three mobilising variables to the
explanatory power of these models, which is quite an improvement across
the different geo-political regions except for sub-Saharan African coun-
tries. In sub-Saharan Africa, the Nagelkerke pseudo R2 moves from 0.068
to 0.098, which is quite an improvement but it is still rather low compared
with other regions. In contrast, for three newer democracies (i.e. as in
Western European democracies) adding the mobilisation variables in Model
106 Mobilisation and the vote
III moves the Nagelkerke pseudo R2 values substantially. To 0.213 in Latin
America, 0.175 in East Asia, and 0.273 in post-Communists European states,
with little changes in the values or estimates in the previous model, which
only included the block of socio-demographic predictors. In fact, for some
regions, Model III suggests the mobilisation variables mediate some of the
socio- demographic factors. For instance, for post- Communist European
states, adding the mobilisation variables reduced the impact of gender on
voting, as men are only now modestly (i.e. at the 0.05 level) more likely to
vote. For East Asian democracies, place of residence shows a more sub-
stantial influence on vote choice once I include the mobilisation variables.
Finally, for sub-Saharan Africa, except for age, we observe a decrease in the
effect estimate for the socio-demographic variables, notably religion and
one’s place of residence.
Second, I examine which of the three mobilisation indicators seems to
matter the most to vote choice in new democracies. Beginning with party
identification, in all newer democracies, I observe a statistically significant
effect of party identification on voting.7 This validates our expectation that
those who are party identifiers tend to participate at higher rates than non-
party members. Overall, this result is not only in line with that of Western
democracies (Smets and van Ham, 2013), and they are quite an interesting
finding given the weak roots that parties tend to have new democracies.
Also, a look at the effect of associational membership on electoral partici-
pation across the different geographical regions (i.e. new and older democra-
cies) reveals almost similar findings except for East Asian democracies. In new
democracies of sub-Saharan, Latin America, and post-communist European
countries as in Western European democracies, our data shows that there is
a statistically significant relationship between associational membership and
voting. In these regions, voters who declared themselves to be members of
informal networks and or unions tend to participate at higher rates than non-
members. However, for East Asian cases, this proposition is not supported
as the relationship between associational membership and voting is non-
significant and negatively signed.
Although this finding from the East Asia region seems inconsistent with the
conclusions of the other regions, it might not be too surprising given that there
is little agreement in the comparative literature on the types or dimensions
of social capital that matters for political engagement. For example, the
AsianBarometer asked respondents whether they are members of a group in 20
different areas. Thus, I argue it might be that not all associational memberships
may tend to be important for voting. To substantial, in their country study on
South Korea, Lee and Glasure (2007, pp. 109–111) show that while partici-
pation in some types of associations (i.e. trade unions) leads to higher rates
of political participation. In contrast, for other kinds of associations, such
as environmental groups, their findings reveal a reverse relationship with par-
ticipation as members were less likely to engage than non-members. Overall,
Mobilisation and the vote 107
in the face of this finding, I, therefore, conclude that African, Latin America,
and Central/Eastern European democracies do resemble older democracies of
Western Europe, but East Asian democracies are different.
Finally, regarding clientelism, recall, this variable was subjected to empir-
ical testing only in sub-Saharan Africa and Latin America. As reported
in Table 5.3, the result across the two regions are quite revealing. In Latin
America, our data suggest a non- significant (i.e. although being in the
expected direction) relationship between clientelism and voting. This result
for the Latin American countries might not be too surprising given Carlin
et al.’s (2015) and Nadeau et al.’s (2017) studies that include a sample of 18
Latin American countries depict only a modest or limited relationship of the
impact of clientelism on vote choice/presidential vote.
For sub-Saharan Africa cases, the relationship between clientelism and
vote choice is significant and negatively signed, which is directly opposite to
what I have expected. First, for one reason, this result might be very incon-
clusive since our measure of clientelism or vote-buying in sub-Saharan Africa
does not capture precisely whether voters were offered material benefit or not
in exchange for this vote. But rather the question measuring vote-buying only
captures people’s perception of the present of vote-buying in their country
(i.e. in your opinion, how often do the following things occur in this country’s
elections: Voters are bribed). Second, even though we lack an appropriate
question for vote-buying, still this result might not be too surprising since evi-
dence convened by other scholars in the broader sub-Saharan African region
shows mixed results concerning the influence of vote buying and vote choice
(see Lindberg and Morrison, 2008; Young, 2009; Letsa, 2019) For instance,
in their country study of what shapes voting decision in Ghana, Lindberg
and Morrison (2008) concluded the effect of clientelism was relatively unim-
portant in influencing Ghanaians voting decision during the 2000 election.
Similarly, in a recent study of expressive voting in Cameroon, Letsa (2019,
p. 11) shows vote-buying does not necessarily predict voting.
Overall, irrespective of the lack of an appropriate measure for clientelism in
sub-Saharan African, the result in Latin American may as well be very incon-
clusive for the reason that the measures of clientelism refer to outright vote-
buying, which is typically illegal and underreported in the AmericasBarometer
survey. I consider the issue of clientelism in more detail towards the end of
this chapter, trying to assess its effect across all four regions. In sum, having
concluded with the logistic regression results for each of the geo-political
regions, Table 5.4 enables us to gauge how far the mobilisation model suits
newly consolidated democracies. In this regard, I can confirm the transcendent
importance of party identification and association memberships (albeit of a
stronger impact of party identification) in new democracies, with the result
very much mirroring those of older democracies of Western Europe.
Having concluded with the analyses of which mobilisation variables matter
for predicting individual voting propensity in the multivariate context, I now
108 Mobilisation and the vote
Table 5.4 Theoretical matching of mobilisation model on vote choice: based on the
multivariate results
Mobilisation model
Party + Applies*** Applies*** Applies*** Applies*** Applies***
identification
Associational + Applies*** Applies*** Applies*** Does not Applies***
membership apply
Clientelism + N/A Applies* Does not N/A N/A
apply
Mean
Sub-Saharan Africa
Benin 0.52
Botswana 0.23
Cape Verde 0.26
Ghana 0.50
Mauritius 0.33
Namibia 0.31
Sao Tome and Principe 0.45
South Africa 0.33
Latin America
Argentina 0.50
Brazil 0.35
Chile 0.10
Costa Rica 0.15
El-Salvador 0.46
Guyana 0.48
Panama 0.46
Peru 0.55
Uruguay 0.05
East Asia
Taiwan 0.29
Mongolia 0.36
South Korea 0.18
Post-communist Europe
Czech Republic 0.14
Estonia 0.06
Hungary 0.29
Lithuania 0.16
Poland 0.12
Slovenia 0.09
Western Europe
Austria 0.07
Belgium 0.04
Germany 0.04
Finland 0.06
United Kingdom 0.06
Ireland 0.13
Italy 0.25
Netherlands 0.03
Norway 0.03
Sweden 0.03
Regions
Sub-Saharan Africa 0.37
(continued)
112 Mobilisation and the vote
Table 5.5 Cont.
Mean
Latin America 0.34
East Asia 0.28
Post-Communist Europe 0.15
Western Europe 0.07
Source: V-Dem data (2019), index of clientelism are based on 2011–2016. Index based on
three dimensions: election vote buying (C) (v2elvotbuy), particularistic or public goods
(C) (v2dlencmps) and party linkages (C) (v2psprlnks).
Sub-Saharan Africa
Clientelism index 3.49*** (0.26) 5.90*** (1.29) 7.73*** (1.41)
Country dummies Yes Yes Yes
Nagelkerke’s R2 0.029 0.068 0.104
Number of observations/cases 10969/08 10423/08 10023/08
Latin America
Clientelism index 2.59*** (0.47) 1.47*** (0.51) 1.69*** (0.53)
Country dummies Yes Yes Yes
Nagelkerke’s R2 0.026 0.172 0.215
Number of observations/cases 13609/09 12978/09 12571/09
East Asia
Clientelism index 2.03*** (0.62) 5.05*** (0.71) 2.23*** (0.83)
Country Dummies Yes Yes Yes
Nagelkerke’s R2 0.009 0.136 0.175
Number of observations/cases 3956/03 3728/03 3639/03
Post-communist Europe
Clientelism index 4.26 (0.39) -2.01 (0.79) 7.66 (5.08)
Country dummies Yes Yes Yes
Nagelkerke’s R2 0.036 0.146 0.273
Number of observations/cases 10060/06 9877/06 9544/06
Notes: These results are based on a measure of clientelism that is derived from V-Dem data
(2019), the index of clientelism is based on 2011–2016. Significance: ***p < 0.001; **p < 0.01;
*p < 0.05. Entries are standardized logistic regression coefficients, standard errors in parentheses.
Model I has no Controls; Model II controls for socio-demographic variables; Model III controls
for socio-demographic and two mobilisation variables (i.e. party ID and associational networks).
The data are weighted using the weights provided for each survey data.
Conclusion
In this chapter, I have discussed and tested the effect of three mobilisa-
tion variables on electoral participation: party identification, associational
networks, and clientelism. Having concluded with the multivariate ana-
lyses across the four regions, the results obtained are not only clear but
very interesting in several ways. The standout result is that party identifi-
cation plays a more dominant role in explaining voting behaviour in newly
consolidated democracies in sub-Saharan Africa, Latin America, East Asia,
and post-communist Europe as the pooled results revealed. Similarly, the
examination of the impact of party identification across individual-countries
re-attests the overriding effect of party identification as its relationship with
voting proves to be statistically significant and in the expected direction in
22 of the 26 countries in all four regions, which is a little surprising given
their limited experience with democracy, and the instability of their party
systems. Second, another mobilisation variable that seems to matter for vote
choice is associational membership, although not so dominant as the role
of party identification. To be precise, not only does the multivariate analysis
suggest the importance of associational membership on vote choice, but the
country by country analysis also reveals this variable tends to be essential
for electoral participation in 16 out of 26 newly consolidated democracies.
Finally, our examination of clientelism using a measure from Varieties of
114 Mobilisation and the vote
Democracies (V-Dem data) indicates that even if the practice is relatively fre-
quent in emerging democracies and is generally used by politicians and pol-
itical entrepreneurs seeking to attract voters, the presence of clientelism does
not seem to diminish the value of other predictors of vote choice, notably the
socio-demographic and mobilisation variables.
Notes
1 However, the act of canvassing can be equally considered as a vital mobilisation
factor. Canvassing is seen as an act of requesting or inspiring people to partici-
pate, and this can definitely have a significant effect on people’s voting decisions.
According to scholars such as Gerber and Green (2000a, 2000b), the act of
canvassing and knocking on voters’ doors and reminding them to vote seems to
be a most effective tool in increasing citizens’ propensity to turn out in the polls.
Similarly, Johnston, Cutts, Pattie, and Fisher (2012), focusing on the British gen-
eral election in 2010, provide evidence that people who are contacted by a party
are more likely to vote than those who are not based on these studies, although
canvassing seems an important mobilisation factor. Yet, because of the absence of
data availability, I am unable to examine its impact on vote choice across newer
democracies.
2 In new democracies, it is not uncommon for politicians and or political
entrepreneurs to attempt to secure the popular vote or support of the voters by
offering gifts, money, or material benefits.
3 Countries surveyed include: Austria, Belgium, Switzerland, Germany, Denmark,
Spain, Finland, France, United Kingdom, Greece, Ireland, Italy, Luxembourg,
Netherlands, Norway, Portugal, and Sweden.
4 The countries included in the survey were: Botswana, Ghana, Lesotho, Malawi,
Mali, Namibia, Nigeria, South Africa, Tanzania, Uganda, Zambia, and
Zimbabwe.
5 The countries include: Algeria, Benin, Botswana, Burkina Faso, Burundi,
Cameroon, Cape Verde, Cote d’Ivoire, Egypt, Gabon, Ghana, Guinea, Kenya,
Lesotho, Liberia, Madagascar, Malawi, Mali, Mauritius, Morocco, Mozambique,
Namibia, Niger, Nigeria, Sao Tome and Principe, Senegal, Sierra Leone, South
Africa, Sudan, Swaziland, Tanzania, Togo, Tunisia Uganda, Zambia, and
Zimbabwe.
6 The Latin America countries include: Bolivia, Mexico, Guatemala, El Salvador,
Honduras, Nicaragua, Costa Rica, Panama, Colombia, Peru, Paraguay, Chile,
Uruguay, Brazil, Argentina, and Dominican Republic.
7 For similar results see Carreras and Castañeda-Angarita (2014); Bratton, Chu,
and Lagos, (2010); Kuenzi and Lambright (2011).
8 For instance, in Latin America, our sample suggest only 4 per cent of the
respondents declared they had received bribes/gifts for their vote.
9 The question measuring vote buying reads as follows: in this national election, was
there evidence of vote and/or turnout buying?
10 A party-constituent linkage refers to the sort of good that the party offers in
exchange for political support and participation in party activities.
11 Considering the profile of social and infrastructural spending in the national
budget, how “particularistic” or “public goods” are most expenditures.
Mobilisation and the vote 115
12 Such spending may be referred to as pork, clientelistic, or private goods.
13 Note the index is formed across all three dimensions by taking the reversed point
estimates, so that higher scores equal more clientelism.
14 This time period was chosen since it coincided very closely to the time the elections
and survey data were run across the different regions.
15 I did not conduct any analysis for Western European democracies as the rate of
clientelism is low, with very little variation across the countries.
6
Political-psychological variables
and the vote
45
40
35
30
25
20
15
10
5
0
Western Europe Post-communist Sub-saharan Africa La n America East Asia
Europe
Very interested Somewhat interested Not very interested Not at all interested
50
45
40
35
30
25
20
15
10
5
0
Western Europe Post-communist Sub-saharan Africa Lan America East Asia
Europe
Trust a great deal Trust quite a lot Not very much None at all
East Asia
Lan America
Sub-saharan Africa
Post-Communist Europe
Western Europe
0 10 20 30 40 50 60 70 80 90 100
Need to be very careful Most people can be trusted
Source: Bivariate results between voting and the political-psychological variables across the
regions are derived from the European Social Survey, Afrobarometer, AmericasBarometer, and
the AsianBarometer. The observations are all weighted using the different surveys weights. Notes:
**Correlation is significant at the 0.01 level (2-tailed); *Correlation is significant at the 0.05 level
(2-tailed).
Political-psychological variables 121
However, the strength of the relationship is revealed to be quite weak in new
democracies of sub-Saharan Africa, Latin America and East Asia except for
post-communist European countries (as in Western European democracies).
Overall, given the significant and positive relationship across all regions, at
least in the bivariate context, I can tentatively conclude that new democra-
cies resemble Western European democracies. Moving to the last political-
psychological variables, external and internal efficacy, the proposition about
the effect of external and internal efficacy on voting produced varied findings.
The data from the two regions, Latin America and post-Communist states,
indicate that as in Western European democracies, the relationship between
electoral participation and external efficacy is positive and significant. In con-
trast, for sub-Saharan African and East Asian democracies, the relationship
between voting and external efficacy is non-significant. Rounding up with
internal efficacy, we observe a significant and positive relationship with vote
choice only in Latin America, sub-Saharan African, and post-communist
countries (i.e. as in Western European democracies) with the exception being
East Asia.
While these bivariate results tell us of the direction of the relationship
between political interest, political trust, external efficacy, and internal effi-
cacy and vote choice, to be more confident of these results, and to explore the
sources of variation more rigorously, one needs to engage with multivariate
analysis, which is the task of the next section.
Table 6.2 reports the logistic regression analysis for each of the geo-political
regions (i.e. Appendix 3, Table C5, provides summary statistics for the political-
psychological variables). Note, the model includes socio- demographic
and mobilisation variables to which I now add the political-psychological
indicators.
The first question to examine is the extent to which political-
psychological variables contribute to an overall explanation of people’s
newgenrtpdf
Table 6.2 Multivariate models of voter turnout in newly consolidated democracies: socio-demographic, mobilisation and political-psychology
122
variables
Political-psychological variables
Regions/variables
Months since last election 0.20*** (0.02) 0.19*** (0.02) 0.15*** (0.02) 0.14*** (0.02)
Age 0.03*** (0.00) 0.02*** (0.00) 0.02*** (0.00)
Gender (1 = men) 0.07 (0.04) -0.02 (0.04) -0.17*** (0.05)
Religion (2 = everyday/once a week; 0 = never) 0.12*** (0.01) 0.10*** (0.01) 0.08*** (0.01)
Nominal ethnic identity (Largest ethnic group = 1; Minority ethnic 0.96*** (0.09) 0.91*** (0.09) 0.89*** (0.10)
group = 0)
Residence (1 = rural) 0.15*** (0.04) 0.16*** (0.04) 0.20*** (0.05)
Education 0.56*** (0.03) 0.49*** (0.03) 0.31*** (0.03)
Party Identification (1 = party ties) 1.13*** (0.05) 0.85*** (0.05)
Associational membership (1 = member currently; 0.5 = member 0.65*** (0.06) 0.64*** (0.06)
previously; 0 = not a member)
Clientelism N/A N/A
Political interest 0.88*** (0.05)
Political trust 0.19*** (0.03)
Internal efficacy 0.09** (0.04)
External efficacy 0.11* (0.05)
Constant -2.96*** (0.38) -5.74*** (0.42) -5.11*** (0.44) -5.38*** (0.46)
Nagelkerke’s R2 0.044 0.146 0.209 0.244
% correctly classified 83% 83% 84% 85%
Number of observations/cases 18657/10 18332/10 17962/10 17318/10
Sub-Saharan Africa
Months since last election 0.01*** (0.00) 0.02*** (0.00) 0.02*** (0.01) 0.02*** (0.03)
Age 0.38*** (0.04) 0.35*** (0.05) 0.29*** (0.05)
Gender (1 = men) -0.02 (0.06) -0.11 (0.07) -0.16* (0.08)
Religion (2 = more than once a day; 0 = never) 0.08*** (0.03) 0.07** (0.03) 0.08*** (0.03)
Salience of ethnicity (1 = only ethnic ID; Equal national/ethnic -0.24** (0.10) -0.18 (0.11) -0.14 (0.12)
ID = 0.5; Only national ID = 0)
Residence (1 = rural) 0.28*** (0.07) 0.20** (0.08) 0.18** (0.08)
Education -0.07 (0.12) 0.01 (0.13) -0.09 (0.14)
Party identification (1 = party ties) 0.76*** (0.08) 0.61*** (0.09)
Associational membership (1 = active member/official leader; 0.56*** (0.11) 0.53*** (0.11)
0.5 = inactive member; 0 = not a member)
Clientelism (1 = receive gifts/bribes) -0.24** (0.09) -0.21* (0.11)
Political interest 0.23*** (0.06)
Political trust 0.08 (0.05)
Internal efficacy 0.06 (0.09)
External efficacy 0.05 (0.08)
Constant 1.62*** (0.35) 0.45*** (0.16) -0.43 (0.65) -0.17 (0.75)
Nagelkerke’s R2 0.029 0.068 0.098 0.102
% correctly classified 80% 82% 82% 83%
Number of observations/cases 10969/08 10423/08 10023/08 5920/08
Latin America
Months since last election -0.05*** (0.01) -0.03*** (0.01) -0.03*** (0.01) -0.03*** (0.01)
Age 0.05*** (0.00) 0.05*** (0.00) 0.05*** (0.00)
Gender (1 = men) -0.12** (0.05) -0.11** (0.05) -0.13*** (0.05)
124
Regions/variables
Political-psychological variables
Western Europe Model I Model II Model III Model IV
% correctly classified 76% 78% 80% 80%
Number of observations/cases 13609/09 12978/09 12469/09 11467/09
East Asia
Months since last election -0.04*** (0.01) -0.09*** (0.01) -0.04*** (0.01) -0.04** (0.01)
Age 0.06*** (0.00) 0.06*** (0.00) 0.05*** (0.01)
Gender (1 = men) -0.07 (0.09) -0.06 (0.09) -0.08 (0.11)
Religion (2 = several times a day; 0 = never) 0.12*** (0.04) 0.11** (0.04) 0.11** (0.05)
Nominal Ethnic ID (Largest ethnic group = 1; Minority ethnic -0.15 (0.15) -0.08 (0.15) 0.02 (0.16)
group = 0)
Residence (1 = rural) 0.25* (0.13) 0.35** (0.14) 0.30** (0.15)
Education 0.44*** (0.06) 0.42*** (0.07) 0.31*** (0.07)
Party Identification (1 = party ties) 1.05*** (0.11) 0.91*** (0.12)
Associational membership (1 = member; 0 = not a member) -0.18 (0.24) -0.25** (0.11)
Clientelism N/A N/A
Political interest 0.79*** (0.14)
Political trust 0.04 (0.07)
Internal efficacy -0.05 (0.10)
External efficacy 0.01 (0.16)
Constant 2.65*** (0.24) -0.65 (0.36) -0.66** (0.39) -0.96** (0.43)
Nagelkerke’s R2 0.009 0.136 0.175 0.189
% correctly classified 86% 86% 86% 86%
Number of observations/cases 3956/03 3724/03 3639/03 3194/03
Post-communist Europe
Months since last election -0.01 (0.00) 0.00 (0.01) -0.01 (0.01) -0.01 (0.01)
Age 0.03*** (0.00) 0.02*** (0.00) 0.02*** (0.00)
Gender (1 = men) 0.22*** (0.05) 0.09* (0.05) -0.08 (0.05)
Religion (1 = everyday/once a week; 0 = never) 0.16*** (0.02) 0.15*** (0.02) 0.13*** (0.02)
Nominal ethnic ID (Largest ethnic group = 1; Minority ethnic 0.45*** (0.09) 0.52*** (0.11) 0.44*** (0.12)
group = 0)
Residence (1 = rural) -0.05 (0.05) -0.00 (0.05) 0.03 (0.05)
Education 0.60*** (0.03) 0.53*** (0.06) 0.39*** (0.04)
Party Identification (1 = party ties) 1.73*** (0.06) 1.34****(0.06)
Associational membership (1 = member currently; 0.5 = member 0.54*** (0.09) 0.42*** (0.10)
previously; 0 = not a member)
Clientelism N/A N/A
Political interest 1.18*** (0.07)
Political trust 0.24*** (0.03)
Internal efficacy 0.20*** (0.06)
External efficacy 0.22** (0.07)
Constant 1.15*** (0.14) -1.66*** (0.21) -1.45*** (0.22) -1.63*** (0.25)
Nagelkerke’s R2 0.036 0.146 0.273 0.334
% correctly classified 67% 71% 74% 77%
Number of observations/cases 10060/06 9877/06 9544/06 8943/06
Political-psychological model
Political + Applies*** Applies*** Applies*** Applies*** Applies***
interest
Political + Applies*** Does not Does not Does not Applies***
trust apply apply apply
Internal + Applies*** Does not Does not Does not Applies***
efficacy apply apply apply
External + Applies*** Does not Does not Does not Applies***
efficacy apply apply apply
Conclusion
This chapter continues with our causal sequence of individual- level
explanations of vote choice in newer democracies by focusing on political-
psychological variables. It explores how political interest, political trust,
internal and external efficacy affect people’s voting decisions in four regions.
The overall contribution of this model to the vote choice model indicates it
is quite modest only in new democracies of post-communist European coun-
tries while for the other regions, sub-Saharan Africa, Latin America, and East
Asia, the general contribution of the model to vote choice is relatively weak or
tiny. To be precise, the surprise finding regarding the political-psychological
model that is tested is that all these variables except for political interest prove
to be non-significant for voting in sub-Saharan Africa, Latin America, and
East Asia. However, the clear message from this chapter is that among the
political-psychological variables, political interest constitutes the main driving
force behind people’s decision of whether to vote or not across all new dem-
ocracies. Consequently, the main implications of this result are that, given
that interest about politics positively inspires people to go to the polls, as indi-
viduals continue to gain political knowledge and experience with the further
consolidation of democracy across these emerging democracies, voters will
be more likely to hold their government, leaders, and policy-makers account-
able. This is positive for the overall quality of the democratic process in newer
democracies.
Notes
1 That is the feeling that political and social change is possible and that the individual
citizens can play a part in bringing about the change.
2 Internal efficacy refers to the belief that one can personally influence the political
process, whereas external efficacy refers to the belief that the political system is
responsive to one’s views (Balch 1974; Niemi, Craig and Mattei, 1991).
3 www.worldvaluessurvey.org/WVSDocumentationWV6.jsp.
4 Although I attempt to present a picture of people’s psychological orientation based
on political trust, as pointed out by Shi (2014), it might be argued whether political
trust has the same meaning in such diverse systems or regimes in the globe.
5 For established Western European democracies, the Nagelkerke pseudo R2 increases
by 0.035.
7
Alternative explanations and the vote
So far, to establish why people vote in the newly consolidated democracies, and
as illustrated from the previous empirical chapters (see e.g. Chapter 4, 5, and
6), I have relied heavily on theories of voting behaviour that were primarily
developed based on the experience of voters from older and economically
more developed democracies. However, considering the newness of elections
across emerging democracies, coupled with their cultural and historical leg-
acies, I contend new democracies could differ from established democracies
of the global North in terms of individual-level turnout.
To be precise, most countries in sub-Saharan Africa, Latin America,
East Asia, and post-Communist Europe continue to be plagued by wide-
spread corruption allegations, weak institutions, and socioeconomic
exclusion or inequality. Therefore, at the heart of this chapter is the
need to identify those factors that set newer democracies apart from the
established Western democracies. To do this, I examine what I deem to
be an alternative explanation1 of turnout that focuses on the impact of
corruption, authoritarianism, and the economy. In the first part of this
chapter, I explore and provide some descriptive and qualitative infor-
mation about the three core explanatory variables. In the second section
of the chapter, relying on inferential statistics, I estimate the effects of
corruption, authoritarianism, and the economy on vote choice across the
four areas with newer democracies.
Sub-Saharan Africa
Benin 41
Botswana 61
Cape Verde 58
Ghana 41
Mauritius 52
Namibia 52
Sao Tome and Principe 46
South Africa 44
Latin America
Argentina 45
Brazil 35
Chile 67
Costa Rica 56
El-Salvador 34
Guyana 40
Panama 36
Peru 36
Uruguay 71
East Asia
Taiwan 65
Mongolia 35
South Korea 59
Post-communist Europe
Czech Republic 56
Estonia 74
Hungary 44
Lithuania 60
Poland 58
Slovenia 60
Western Europe
Austria 77
Belgium 75
Germany 80
Finland 86
United Kingdom 77
Ireland 74
Italy 53
Netherlands 82
Norway 84
Sweden 85
Regional mean
Sub-Saharan Africa 49
Latin America 47
East Asia 53
Post-communist Europe 58
Western Europe 77
Notes: percentages are derived from the different cross-national survey data. European Social
Survey (Eastern European countries); Afrobarometer (Sub-Saharan Africa); AmericasBarometer
(Latin America); AsianBarometer (East Asia).
134 Alternative explanations and the vote
60
50
40
30
20
10
0
2009 2010 2011 2012 2013 2014 2015 2016 2017 2018
Africa Asia Americas Europe
Repression of opposion
Media repression
0 10 20 30 40 50 60 70 80 90
However, before estimating the effects of all four blocks of variables on vote
choice, the bivariate relationship between the three alternative variables and
voter turnout is first assessed across all four newer democracies.
Alternative explanations and the vote 137
Table 7.3 Corruption, authoritarianism and the economy and voting in new democra-
cies: bivariate relationships
Western Europe
Corruption perception -0.04**
Authoritarianism (democratic erosion = 1; no democratic erosion = 0) N/A
Perception of the economy (better = 1; same = 0.5; worse =0) 0.07**
Sub-Saharan Africa
Corruption perception -0.06**
Authoritarianism (democratic erosion = 1; no democratic erosion = 0) -0.07**
Perception of the economy (better = 1; same = 0.5; worse = 0) 0.01
Latin America
Corruption perception -0.01
Authoritarianism (democratic erosion = 1; no democratic erosion = 0) 0.01
Perception of the economy (better = 1; same = 0.5; worse = 0) -0.02*
East Asia
Corruption perception -0.03*
Authoritarianism (democratic erosion = 1; no democratic erosion = 0) 0.05**
Perception of the Economy (better = 1; same = 0.5; worse = 0) 0.04**
Central and Eastern Europe
Corruption perception -0.04**
Authoritarianism (democratic erosion = 1; no democratic erosion = 0) 0.12**
Perception of the economy (better = 1; same = 0.5; worse = 0) 0.09**
Source: Bivariate results between voting and the alternatives variables across the regions
are derived from the European Social Survey, Afrobarometer, AmericasBarometer, and the
AsianBarometer. The observations are all weighted using the different surveys weights. Notes:
**Correlation is significant at the 0.01 level (2-tailed); *Correlation is significant at the 0.05 level
(2-tailed).
138 Alternative explanations and the vote
a weak but significant relationship between voter’s perception of the economy
and voter turnout. Although, this association is quite weak, it is apparent
citizens who have a more positive evaluation of their country’s economic
conditions are more likely to participate than those who lack this positive
evaluation. However, the same cannot be said for new democracies of sub-
Saharan Africa and Latin America.
Finally, concluding with our last alternative variable, authoritarianism. It
is essential to note that because of data availability, I was unable to empirically
put this variable to test for the Western European countries.7 Notwithstanding
our inability to test this variable, the proposition vis-à-vis this variable was
that in countries experiencing a surge in authoritarian practices, voters would
be more likely to engage. Table 7.3 shows the results across the four regions
tend to be varied. For example, in East Asia and post-Communist European
cases, I can confirm voters are significantly more likely to vote in countries
experiencing the erosion of democratic principles. For sub-Saharan African
countries, I observe the opposite, in that a voter’s experience with authori-
tarianism or the erosion of democratic values significantly reduces the prob-
ability of voting. Concluding, with the Latin American region, I observe a
non-significant relationship between authoritarianism and voting. Although
our bivariate test of the impact of the alternative variables on vote choice in
new democracies does provide some clues about which factor is relevant or
not in which context/region, only a multivariate model would enable us to
be more confident of these results. Thus, the model that I build in the next
section takes into account all four blocks of variables (i.e. socio-demographic,
mobilisation, political-psychological, and alternative factors).
Regions/variables
Months since last election 0.20*** (0.02) 0.19*** (0.02) 0.15*** (0.02) 0.14*** (0.02) 0.001 (0.00)
Age 0.03*** (0.00) 0.02*** (0.00) 0.02*** (0.00) 0.02*** (0.00)
Gender (1 = men) 0.07 (0.04) -0.02 (0.04) -0.17*** (0.05) -0.16*** (0.05)
Religion (1 = everyday/once a week; 0 = never) 0.12*** (0.01) 0.10*** (0.01) 0.08*** (0.01) 0.07*** (0.01)
Nominal ethnic identity (Largest ethnic group = 1; 0.96*** (0.09) 0.91*** (0.09) 0.89*** (0.10) 0.90*** (0.09)
Minority ethnic group = 0)
Residence (1 = rural) 0.15*** (0.04) 0.16*** (0.04) 0.20*** (0.05) 0.23*** (0.05)
Education 0.56*** (0.03) 0.49*** (0.03) 0.31*** (0.03) 0.30*** (0.03)
140
Regions/variables
142
Regions/variables
Source: European Social Survey (Western and Eastern European countries); Afrobarometer (Sub-Saharan Africa); AmericasBarometer (Latin America);
AsianBarometer (East Asia). The data are weighted using the weights provided for each survey data. Significance: ***p < 0.001; **p < 0.01; *p < 0.05. Entries
are standardized logistic regression coefficients, standard errors in parentheses.
144 Alternative explanations and the vote
as the Nagelkerke pseudo R2 barely moves in some regions, while it reduces
in others. Second, I examine which of the socio-demographic, mobilisation,
and political-psychological indicators that were included in the third block of
our recursive casual system continue to matter for electoral participation even
after the alternative variables have now been included. The straight response
to this is that across the different regions, the introduction of the alternative
variables does not change the values of the socio-demographic, mobilisation,
or political, psychological variables very much, as the results remain mostly
unchanged.
Third, following up on the bivariate results (see Table 7.3), I probe
which of the alternative variables seems to matter for vote choice across the
different regions. As shown in Table 7.4, looking at the impact of corruption,
across across three newer democracies sub-Saharan Africa, East Asia, and
post-Communist European democracies (i.e. except for the Latin America
region8) the result reveals corruption has a significant and negative effect on
people's decision of whether to vote or not. To be precise, in sub-Saharan
Africa, East Asia, and post-Communist European democracies, this result
implies citizens’ perception of corruption in their countries tend to decrease
turnout significantly. These findings seem to corroborate those of Dahlberg
and Solevid (2016), Stockemer, LaMontagne, and Scruggs (2013), Sundström
and Stockemer (2015), Domínguez and McCann (1998), and Hooghe and
Quintelier (2014). All said and done, this finding gives weight to the fact that
corruption has a fundamentally different influence in newer democracies
when compared to established Western democracies.
Next, concerning authoritarianism, our data tells of a statistically sig-
nificant relationship between authoritarianism and vote choice only in
two newer democracies, notably sub-Saharan Africa and post-communist
European democracies. Yet, the direction of the coefficient does vary across
the two regions. On the one hand, the impact of authoritarianism on vote
choice is significant and positive for post-communist European countries
while it is significant and negative for sub-Saharan Africa. Consequently, we
can infer two things. (1) In post-communist European states, it is apparent
voters’ experience with the rise of authoritarian practices and or demo-
cratic erosion, which is characterised by the curtailed of civil liberties and
press freedom, leads to an increase in voter turnout. (2) In sub-Saharan
Africa, voters’ experience with democratic erosion or authoritarian tenden-
cies tends to decrease voters’ turnout. For the two remaining regions, East
Asia and Latin America, the effect between authoritarianism and voting is
non-significant.
Finally, looking at the economy, the expectation was that voters who had a
more positive evaluation of their country economy would be more likely to vote.
Our result was only confirmed in the post-communist European region. Apart
from this, in the remaining three regions notably sub-Saharan Africa, Latin
America and East Asia the result depicts a non-significant relationship with vote
choice. However, the coefficient across these regions was positively signed.
Alternative explanations and the vote 145
Conclusion
In this chapter, I proposed a framework to study voter participation that
focuses on the impact of corruption, authoritarianism, and the economy.
Overall, I find distinct patterns in the impact of these factors across the
different regions. On the one hand, my results indicate that the influence of
corruption on voting is quite similar in sub-Saharan Africa, East Asia, and
the post-communist European countries. In these three settings, perceptions
of corruption tend to significantly decrease turnout at the individual level.
However, the same cannot be said for the other two variables: the economy
and authoritarianism. Voters’ evaluations of the economy do not fully attain
significance across most of the regions except in post-communist European
countries, although the coefficient is positive, suggesting voters having a
positive evaluation of their national economy should be more likely to turn
out. The effect of authoritarianism on vote choice is significant only in sub-
Saharan Africa and post-communist European states but, even so, the result
across these regions reveals a mixed pattern. For sub-Saharan Africa, voters’
experience with authoritarian practices and undemocratic tendencies tends to
decrease turnout, whereas the opposite is true for post-communist countries.
Notes
1 Although, I am not the first person to examine the impact corruption, authoritar-
ianism, and the economy in explaining voting behaviour in new democracies. Yet,
except for Tambe (forthcoming), no cross-national study has been able to estimate
all these variables in a single model of vote choice in new democracies.
2 Rose and Peiffer (2019, p. 46) argue this categorisation permits for a greater degree
of differentiation than sorting countries into two categories, corrupt and high in
integrity.
3 Moreover, as Zakaria (1994) and Fukuyama (2001) argue, the so-called Asian
values tend to be conducive to corruption.
4 This was closely follow with corruption, with 8 per cent of the respondents
registering it as a crucial problem.
5 A cross-university collaboration that aims to help evaluate threats to democracy
by capturing the symptoms and precursors of democratic erosion across countries
over time www.democratic-erosion.com/event-dataset/project-summary/ [Accessed
01/03/2020].
6 However, democratic erosion has been so far less severe across East Asia liberal
democracies compared to those that are outright electoral authoritarianism or
authoritarian regimes.
7 In most Western European democracies there is either no democratic backsliding
and or weak threat of future backsliding. Similarly, although there might have been
some precursors to backsliding, such as the rise of extremist parties, erosion of
democratic institutions has not yet taken place.
8 In Latin America, corruption has no significant effect on whether people vote or
not. For a more recent similar result see Rose and Peiffer (2019, p. 134).
8
The electoral context and the vote
that have a low voter turnout and turnout remains low; (4) countries that have
a low voter turnout and turnout then increases; (5) countries that have a stable
level of turnout over time, enable us to make the following inferences. First,
Table 8.1 reveals, across the different geo-political regions, in some countries
The electoral context and the vote 149
mostly in Latin America and Western Europe (e.g. Argentina, Uruguay, Peru,
Sweden, Netherlands and Belgium) that have a high voter turnout in their first
election, turnout has remained high is most of the countries. Second, in some
countries that have a high turnout during their first elections (e.g. mostly post-
communist European states and Chile and Mongolia for Latin America and
East Asia), turnout has later declined. Third, in a few countries (i.e. Poland)
that have a low voter turnout, turnout has remained low. Fourth, in some
countries (Taiwan) that had a low voter turnout, turnout has later increased.
Fifth, for Western European democracies, although turnout as shown a slight
decline, yet most of the countries have a stable level of turnout overtime.
In general, Table 8.1 paints a picture of turnout variation across the different
regions, which therefore raises the question of whether these variations can be
in any way accounted for by changes in voters’ descriptive characteristics. That
is, could it be that the composition of the electorate in terms of their socio-
demographic characteristics (education levels) or political- psychological
factors (political interest) across these countries/regions might have changed
over time to explain the patterns depicted in Table 8.1? For instance, focusing
on post-communist European countries where the decline in voter turnout has
been more drastic, could it be that the high turnout rate in the 1990s can be
explained by the presence of highly educated and politically interested voters,
who, later on, disappear in elections that show lower turnout? However, this
should not be the case as studies have long suggested the impact of individ-
uals’ characteristics on vote choice tends to be rather stable over time. In his
study of the determinants of electoral participation in the Netherlands, Van
Egmond (2003) argues: “people tend to be rather stable in most of their indi-
vidual characteristics and behaviour. If they change, this is likely to be can-
celled out by another voter changing in another direction” (p. 5).
Thus, in this chapter, I argue to understand electoral behaviour fully,
it is essential to combine information on individual and institutional
characteristics. I further gain assurance in this line of argument given that some
authors like Coleman (1990) but also Carmines and Huckfeldt (1996) posit
that combining both levels (micro and macro) bring considerable benefits as it
offers explanations on how contextual effects translate into individual behav-
iour. Relatedly, uniting both approaches enables one to position individuals’
characteristics within the boundaries of the political context, which there-
fore helps displays the influences institutional factors can have on individual
behaviour. To illustrate this point, Van Egmond, De Graafi, and Van Der
Eijk (1998, p. 282) using political interest and closeness of elections show the
forms these influences can take. The authors posit during a competitiveness
election –contextual factors such as intensive media coverage may increase
political interest in the electorate and consequently aggregate turnout.
However, this relationship might equally operate in the opposite direction
as alluded to by Campbell (1960). Campbell argues extensive media coverage
of a close race might ensure that voters who are less interested in politics
become fully aware of the political situation, thereby reducing the impact of
150 The electoral context and the vote
political interest. Drawing from these two examples, it is clear the contextual
characteristics or electoral context might function in either impacting dir-
ectly or indirectly individual voting behaviour. Thus, placing the voters within
the context of the election gives me the leeway to test the hypothesis on the
influences of individual and institutional variables on vote choice. All said,
having outlined some of the merits for combining individual and contextual
factors to explain vote choice in new democracies, I round up this section
by presenting some qualitative information on the institutional variables that
will be subjected to empirical testing.
152
Regions/ Year*** Compulsory Concurrent Disproportionality Closeness of
Notes: The sources from which these institutional characteristics are calculated are: Africans Elections Database; Political Database of the Americas; Adam
Carr Elections Archives; Gallagher (2019). Election indices dataset; www.idea.int/data-tools/data/voter-turnout/compulsory-voting. * = for these countries
closeness of the election result derives from the second rounds of the presidential elections. ** = the gap in percentage between the Conservative and the Labour
party was 6.5. However, this was rounded as 7 thus not a close race. *** = this year reflects the calculation of the least square index of disproportionality. Where
in doubt consult Table 3.3. ****In Argentina, elections are partially concurrent with presidential elections. However, the country 2011 presidential and legisla-
tive elections were conducted on the same day.
The electoral context and the vote 153
Ghana and Namibia). For East Asia, of the three countries, elections are
concurrently held only in Taiwan. For post-communist Europe, concurrent
elections were non-existent in all six countries. Rounding up, with Western
European countries of the ten countries it was only in Sweden and the UK
where national and local elections tend to be held concurrently.
Third, for the electoral system, the expectation was that proportional
representation (PR) would enhance people’s propensity to vote. This prop-
osition is based on the fact that most cross-national studies, notably Blais
and Dobrzynska (1998) and van der Eijk and Franklin (1996), show that
disproportionality reduces turnout since the translation of votes into
seats tends to be unequal leading voters to abstain. Across the different
regions (old and new democracies), it is clear the electoral system across the
majority of Western European countries except that of the UK, Italy, and
Germany is predominately proportional. However, this pattern is entirely
the opposite when I focus on newer democracies.4 In new democracies,
except for a few cases in sub-Saharan Africa (e.g. South Africa and Cape
Verde) and Latin America (Guyana, Brazil, Uruguay, etc.) that tend out
to be a parliamentary or presidential system with a proportional represen-
tation system, the electoral system is highly disproportional as the figures
suggest.
Finally, our last institutional variable is the closeness of the election. The
expectation was that in elections where voters perceive it to be a close race
or competitive, turnout will be high. Recall, I operationalise closeness of
the election as the margin of victory for the winning candidate or over the
runner-up in presidential elections while, for parliamentary democracies,
I measure the closeness of election as the difference in seat shares between
the top two parties. For simplicity’s sake and to ensure comparability across
the different regions, I take a 6 per cent threshold as a difference in the seat
shares between the top two parties or between the runners-up in presidential
elections. That said, in Table 8.2, Yes is assigned when the winning margin
is 6 per cent or less (i.e. closed race) and No when the margin is greater than
6 per cent. Using this criterion, it appears East Asian elections were more
competitive in terms of the closeness of the results. This variable will not be
examined in this context because it is a constant for the election year/period
I analyse. On the other hand, across the different regions/countries, there is
a wide variation of the closeness of the election results. Having identified
the empirical indicators, the next section asks if these political-institutional
variables are related to electoral participation in newly consolidated
democracies.
Western Europe
Least squares index of disproportionality (0 = perfect -0.09**
proportionality to 100 maximum disproportionality)
Closeness of election -0.08**
Concurrent elections 0.05**
Compulsory voting 0.07**
Sub-Saharan Africa
Least squares index of disproportionality (0 = perfect 0.03**
proportionality to 100 maximum disproportionality)
Closeness of election 0.01
Concurrent elections 0.01
Compulsory voting N/A
Latin America
Least squares index of disproportionality (0 = perfect -0.02
proportionality to 100 maximum disproportionality)
Closeness of election 0.05**
Concurrent elections 0.02*
Compulsory voting 0.11**
East Asia
Least squares index of disproportionality (0 = perfect 0.05**
proportionality to 100 maximum disproportionality)
Closeness of election N/A
Concurrent elections -0.05**
Compulsory voting N/A
Central and Eastern Europe
Least squares index of disproportionality (0 = perfect 0.04**
proportionality to 100 maximum disproportionality)
Closeness of election -0.12**
Concurrent elections N/A
Compulsory voting N/A
Source: Bivariate results between voting and the Institutional variables across the regions
are derived from the European Social Survey, Afrobarometer, AmericasBarometer, and the
AsianBarometer. The observations are all weighted using the different surveys weights. Notes:
**Correlation is significant at the 0.01 level (2-tailed); *Correlation is significant at the 0.05 level
(2-tailed).
In Table 8.4 (Model VI, column 7), the political-institutional factors are now
added into a model that already consists of all the individual-level explana-
tory variables. Accordingly, it is of foremost importance to examine if the
explanatory power of each of the individual-level blocks of factors from the
previous chapter somehow vary once these political-institutional factors are
included. We observe two things though: first, the addition of the political-
institutional predictors increases the Nagelkerke’s pseudo R2 value only min-
imally in sub-Saharan Africa and Latin America (i.e. as in Western European
democracies) except for East Asia and post-communist European democ-
racies.5 In fact, in East Asia and post-Communist European countries, the
Nagelkerke’s pseudo R2 barely moves or changes, revealing the electoral con-
text variables tend to be less influential than the individual-level indicators.
Second, across the different regions, we observe most of the individual-level
estimates show little or no change when the contextual factors are added to
our vote choice model. To be precise, controlling for the political-institutional
effects on voting, the socio-demographic variables, namely age, education,
and religiosity, continue to retain a highly significant and positive impact on
electoral participation. Similarly, a look at the mobilisation variables suggests
party identification and associational membership continues to emerge as sig-
nificant predictors of voting at the individual level. Also, political interest
remains the most crucial political-psychological predictor of vote choice, even
when political-institutional variables are taken into account. Consequently,
considering the political-institutional variables that are entered in Model VI
(Table 8.4, column 7), although the absence of data availability has hidden the
empirical testing across all regions, it is evident the results are more nuanced
across the different geo-political regions.
First, the disproportionality of the electoral system has a significant and
negative effect only in Western European democracies while for new dem-
ocracies the relationship tends to be non-significant except for sub-Saharan
African democracies where it is positive and significant at the conventional
level. This result simply replicates those of the comparative literature and
accentuates the debate why disproportionality of the electoral system has a
negative effect on electoral participation in older advanced democracies but
not in new democracies (see Gallego, Roco, and Anduiza, 2012). Overall, the
result reveals individuals living in countries where the disproportionality of
the electoral system is close to zero (i.e. more proportional) are more likely to
newgenrtpdf
Table 8.4 Multivariate models of voter turnout in newly consolidated democracies: individual and political-institutional variables
Regions/variables
Months since last election 0.20*** (0.02) 0.19*** (0.02) 0.15*** (0.02) 0.14*** (0.02) 0.001 (0.00) 0.01*** (0.00)
Age 0.03*** (0.00) 0.02*** (0.00) 0.02*** (0.00) 0.02*** (0.00) 0.02*** (0.00)
Gender (1 = men) 0.07 (0.04) -0.02 (0.04) -0.17*** (0.05) -0.16*** (0.05) -0.15*** (0.05)
Religion (2 = everyday/once a week; 0.12*** (0.01) 0.10*** (0.01) 0.08*** (0.01) 0.07*** (0.01) 0.11*** (0.01)
0 = never)
Nominal ethnic identity (Largest 0.96*** (0.09) 0.91*** (0.09) 0.89*** (0.10) 0.90*** (0.09) 0.94*** (0.10)
ethnic group = 1; Minority ethnic
group = 0)
Residence (1 = rural) 0.15*** (0.04) 0.16*** (0.04) 0.20*** (0.05) 0.23*** (0.05) 0.21** (0.05)
Education 0.56*** (0.03) 0.49*** (0.03) 0.31*** (0.03) 0.30*** (0.03) 0.30*** (0.03)
Party identification (1 = party ties) 1.13*** (0.05) 0.85*** (0.05) 0.86*** (0.05) 0.85*** (0.05)
158
Regions/variables
160
Regions/variables
162
Regions/variables
Source: European Social Survey (Western and Eastern European countries); Afrobarometer (Sub-Saharan Africa); AmericasBarometer (Latin America);
AsianBarometer (East Asia). The data are weighted using the weights provided for each survey data. Significance: ***p < 0.001; **p < 0.01; *p < 0.05. Entries
are standardised logistic regression coefficients, standard errors in parentheses.
The electoral context and the vote 163
participate than individuals residing in countries where the electoral system is
highly disproportional.
Second, looking at the closeness of the election,6 note this was only
tested across three newer democracies, notably sub-Saharan Africa, Latin
America, and post-communist European states. Also, recall we expected to
find a significant and positive influence on voting. Our analyses across the
different regions show that closeness of the elections meets the significance
level and in the expected direction only in sub-Saharan Africa. This result
indicates that compared to African voters who live in countries where the
elections are not considered to be close or competitive, those who live in
countries where the elections are close are more likely to vote. In contrast,
for post-communist European countries and Latin America as in Western
European democracies, the impact of the closeness of the election on vote
choice is non-significant. While these results might seem counter-intuitive,
empirically it is, in fact, not such a rare finding (Blais 2006, p. 116; Bratton
et al., 2010). Also, in their meta-analysis of individual-level research on
voter turnout, Smets and Ham (2013) find that only 35 per cent of studies
relying on national-level measures of closeness find a relationship between
closeness and vote choice while none of the studies using district-level
measures of closeness found any link.
Third, for concurrent elections, issues relating to data availability in post-
communist European states and problems of multicollinearity in East Asia
prevented the testing of this variable in both regions. Thus, we were only able
to empirically investigate the impact of concurrent elections on vote choice in
sub-Saharan Africa and Latin American democracies (as in Western European
democracies). Still, we expected that holding of simultaneous elections would
show a positive and significant effect on voting. However, I can only con-
firm that holding concurrent elections does not have a significant effect on
electoral participation in new democracies of sub-Saharan Africa7 and Latin
America (see also Bratton et al., 2010). Lastly, compulsory voting reveals a
significant and positive relationship with electoral participation in new dem-
ocracies of Latin Americans. To be precise, the coefficient estimate suggests
Latin America are twice as likely to vote when elections are compulsory and
enforced with sanctions than when they are not.
Conclusion
It was the aim of this chapter to empirically test the individual and con-
textual variables that I have previously spelt out in our theoretical section
as crucial in influencing people’s decisions of whether or not to vote. At the
individual level, most of the explanatory variables proved significant in the
expected direction in terms of our hypotheses. The most important findings
are the importance of age, education, religion, political interest, party mem-
bership and associational membership. Overall, these largely confirm earlier
studies on electoral participation in Western Europe. Turning our attention to
164 The electoral context and the vote
the contextual variables, I have mostly been unable to find that institutional
factors make much difference to people’s decision to vote in elections across all
geo-political regions. To be precise, with regards to the four main contextual
variables that I have proposed as important in determining people propensity
to vote, I find little consistency; indeed, the results are quite nuanced.8 Yet, the
effect of compulsory voting on political behaviour stands out. To be precise,
voters in new democracies (Latin America) are more likely to turn out where
these elections are mandatory and enforced with sanction than where they are
not. With regards to the other variables, I find that the closeness of elections
tends to be influential in the new democracies of sub-Saharan Africa but not
for the other regions. In sum, this chapter concludes that while the political-
institutional context matters, it is far from decisive as most of the difference
in whether or not a voter turns out on election day in the different regions still
very much depends on individual-level factors.
Notes
1 Also, “institution” means everything from formal structures like a parliament to
very amorphous entities like social class, with other components of the socio-
political universe such as law and markets, also being described as institutions
Peters (2005, p. 29).
2 Note, these turnout figures are drawn only on elections where the most important
figure of the executive is being elected (i.e. national elections).
3 www.idea.int/data-tools/data/voter-turnout/compulsory-voting.
4 However, this result is not too surprising given the dominance of mixed, majority,
and or plurality systems operating in most of these countries.
5 This is not too surprising given that I have been unable to put all the institutional
variables to test in these two regions.
6 A number of interaction effects between closeness of election and political interest,
internal and external efficacy were test across new democracies of sub-Saharan
Africa and Latin America (as in Western European democracies) but proved to
non-significant.
7 We should note that this result is somewhat inconclusive for sub-Saharan Africa
given how few cases there have been of elections held concurrently (i.e. only Ghana
and Namibia having some kind of concurrent election in this dataset).
8 However, I can confirm to observe some of the most substantial relationships on
the impact of the political-institutional variables in Western European democracies.
9
A comparative overview
The core purpose of this study has been to explain what influences people
to vote at the individual level and to establish how well the models of voter
turnout developed in established democracies can travel to newly consolidated
democracies. In order to do this, I have relied on models and or theories that
were initially formulated based on the experience of voters in older established
democracies; this, of course, flags the importance of understanding how far
these Western- generated theories could travel to other contexts (i.e. new
democracies).
Having conducted the relevant empirical analysis, I am now in a position
to address the key question. Does the theoretical explanation of voting derived
from the experience of voters from older established democracies fit the emer-
ging or new democracies of sub-Saharan Africa, Latin America, East Asia,
and post-Communist Europe well? Drawing on the empirical results across the
different geo-political regions, this chapter looks systematically at whether the
findings reported in the previous five chapters (i.e. Chapters 4, 5, 6, 7, and 8)
are generally consistent with each other. This will enable us to gauge how far
Western-derived models suit developing or emerging democracies. All said,
this chapter summarises what we have learned concerning the predictors of
voter turnout in newly consolidated democracies and the validity of our alter-
native explanation to vote choice.
Socio-demographic variables
Beginning with the first block of variables in our recursive causal system (see
Chapter 4), I examine the impact of six socio-demographic variables, not-
ably age, gender, religion, ethnicity/race, place of residence, and education
on vote choice. First, as illustrated in Table 9.1, age proves to be consistent
across all new democracies; as is generally true of established Western democ-
racies, older citizens are far more likely to vote than younger ones. Moreover,
the individual country results across the different regions showed that the
impact of age was positive and significant in 25 of the 26 countries of sub-
Saharan Africa, Latin America, East Asia, and post-communist European
newgenrtpdf
Table 9.1 Summary of empirical findings
Significance: *** p<.001, ** p<.01, * p<.05. Direction: + = positive relationship, given the coding used. –= negative relationship, given coding used. Bold
variables and results across the different regions indicates instances for which it has not been possible to use identical question wording across all four regions
(so there is a chance that any differences might simply be artefactual rather than substantive).
168 A comparative overview
democracies. These results, therefore, strongly reconfirm the transcendent
importance of age observed in the pooled data across four newer democ-
racies. One implication of this result is that it undermines the demand by
some advocates to reduce the voting age to 16, particularly with respect to
sub-Saharan Africa (Juma, 2011). As shown by our results across the four
geo-political regions (i.e. sub-Saharan Africa, Latin America, East Asia, and
Eastern Europe), such a decision to lower the voting age will not necessarily
result in a higher participation rate of young people in electoral politics as
these young individuals are revealed to be least likely to vote when I consider
the socio-demographic variables separately, but more importantly, their reluc-
tance to turnout is evident even when individual-level variables are combined
with country-level variables.
Second, turning to gender, the most striking factor is that it is consist-
ently negatively signed across all newer democracies as in older democracies
of Western Europe. To be precise, a closer look at the significance suggests
the relationship between gender and voting is non-significant across three
newer democracies, sub-Saharan Africa, East Asia, and post-Communist
Europe. However, when gender is found to be significant as in Latin America
or Western European democracies, it is usually women who turn out to
vote at higher rates than men. This result, therefore, confirms research that
continues to show the gender gap in electoral participation has gradually
declined or disappeared. Research on the emergence of a “gender gener-
ation gap” in voting across established democracies shows how the political
behaviour of women has been changing across the generations; this study
suggests that it agrees across countries and geo-political regions as well.
Could this be in part because the changes that have driven the emerging
“gender-generation gap” in Western countries have occurred at the same
rate elsewhere? Moreover, these gender-related differences might not just
affect electoral participation but equally impact non-electoral participation
(i.e. signing petition, boycotting, and demonstration). Therefore, revisiting
the global gender differences in political participation will enable us to fully
establish and be clearer if men or women tend to engage differently in polit-
ical activity other than voting.
Third, looking at religion, it is clear this variable seems to best capture
the effect of socio-demographic on vote choice in new democracies (as in
established democracies of Western Europe). To be precise, the data suggest
church attendance and religious affiliation (for Latin America) shows clear
consistency across all regions judging from the statistically significant rela-
tionship and has the expected sign (positive) in all regions. However, the
country by country results suggest the impact of religion was only significant
in 46 per cent of the cases (i.e. 12 out of 26 countries). A closer look at this
individual country result shows the impact of religion tends to vary signifi-
cantly from region to region, albeit a strong and more consistent effect of
religion in post-communist European states than it is compared to the other
global regions.
A comparative overview 169
Fourth, the impact of ethnicity/race variable tends to vary significantly from
one region to another. Although this nuanced result is somehow unavoidable
since it has not been possible to use identical question wording to tap ethni-
city in one way or another, it is evident that the effect of ethnicity is positive
and significant in post-communist European states as in Western European
democracies. This implies that those who identify as belonging to an ethnic
majority in European democracies (i.e. Eastern and Western Europe) are more
likely to vote. Concerning Latin America and East Asian democracies, the
effect of ethnicity on vote choice is non-significant but points to the expected
direction (i.e. positive sign). In summary, with sub-Saharan Africa democra-
cies, ethnicity falls short of the theoretical expectations. The implication of
the non-significant relationship between ethnic identity and voting, particu-
larly in sub-Saharan Africa, suggests that elections there are much more than
mere ethnic censuses.
Fifth, an individual place of residence and how it affects vote choice
reveals a more systematic or consistent effect across the new democracies. The
data shows that the impact of one’s place of residence (i.e. rural residence) is
statistically significant and has the expected positive sign in three newer dem-
ocracies (e.g. sub-Saharan Africa, East Asia, and Latin America) as in older
democracies of Western Europe. However, in post-communist countries, the
effect of place of residence on vote choice is non-significant.
Finally, the last socio-demographic factor, education, a widely supported
notion in the existing literature suggests that well-educated individuals are much
more likely to turn out in elections than poorly educated citizens. However,
the effect of education on voting shows some consistency only across three
regions, particularly Latin America, East Asia, and post-Communist Europe
(i.e. as in Western European democracies). In Latin America, Eastern Europe,
and East Asia, our results show that turnout is highest among those with
higher levels of education. Still, in sub-Saharan Africa, it is non-significant
and with a negative sign1. All said, in Latin America, Eastern Europe, and
East Asia, all of these effects are statistically significant, so education seems
to matter everywhere.
Still, sub-Saharan Africa stands out from the other regions in this respect.
Notwithstanding the non-significant relationship between voting and educa-
tion in Africa, this result implies two things: (1) within sub-Saharan Africa,
low material status or SES does not pose a barrier to political engagement,
which is an unexpectedly positive finding given that inequality in political par-
ticipation is often documented to affect people’s socioeconomic standings;
(2) at the macro level, what this result shows is that democratic participation
is just as likely to be established and consolidated in poor and underdeveloped
countries (i.e. sub-Saharan Africa) as in the rich and developed countries (see
Przeworski et al. 2000).
Overall, the non-significant relationship between voting and education in
sub-Saharan Africa echoes Gallego (2015), who rebuffs the general assertion
that highly educated citizens turn out to vote at higher rates everywhere by
170 A comparative overview
proposing this differs across countries and is dependent on a set of contextual
features.2 Thus, it will be therefore important for future studies to test whether
the relationship between education and voting depends on regime type or on
other country-level features such as voter registration. Overall, it is fair to say
the socio-demographic resource factors are broadly important for electoral
participation across the global regions, notwithstanding a few differences. In
particular, age, religion, education, and, to some extent, place residence all
serve to increase it.
Mobilisation variables
If we turn to the mobilisation variables that were included in the second block
(see Chapter 5), the effect of three variables has been tested: party identifi-
cation, associational membership, and clientelism. The results obtained are
better illustrated in Table 9.1 and somewhat suggest the mobilisation variables
seem to be important and perform consistently well across most of the newer
democracies. Of the three mobilisation factors, two variables that exert the
most systematic and consistent effect on vote choice are party identification
and associational membership.
First, the party identification variable displays the largest effect; it has
the expected sign (i.e. positive) in all newly consolidated democracies as in
Western European democracies and is statistically significant. Relatedly, the
country by country results reconfirm the pooled results as the impact of
party identification on vote choice is significant in 85 per cent of the cases
(i.e. 22 out of 26 countries in the four regions). Second, an examination of
associational membership shows in three regions sub-Saharan Africa, Latin
America, and post-Communist Europe except for East Asia, the impact of
associational membership on vote choice is statistically significant and posi-
tively signed. Similarly, as earlier indicated in the country by country results,
the impact of associational membership on electoral participation is signifi-
cant in 62 per cent of the cases (i.e. 16 of the 26 countries) across the four geo-
political regions. In these countries and or regions, what is clear is that those
who identified as members of trade unions or informal networks are more
likely to turn out on election day than non-members. Overall, the fact that
these mobilisation variables, particularly party identification, largely mirror
those of advanced Western democracies suggests that even in more patri-
monial and clientelist environments like those of sub-Saharan Africa and or
Latin America, political parties do have a major role in ensuring the engage-
ment of the citizenry in electoral politics. Thus, as cited by Webb and White
(2007), this seems to confirm that one of the key functions that any stable
and effective democracy might expect of parties –fostering political partici-
pation –is largely fulfilled in emerging democracies. But the non-significant
relationship between clientelism and voting in sub-Saharan Africa and Latin
America suggests that elections there are much more than mere clientelist
appeals or censuses.
A comparative overview 171
Political-psychological variables
The political-psychological variables that constitute the third block of our
recursive causal system (i.e. see Chapter 6) examine the impact of four core
variables, notably political interest, political trust, internal and external effi-
cacy on vote choice. First, here, what is evident is that political interest is
the most important political-psychological variable guiding voting in new
democracies as in Western European democracies. To be precise, as Table 9.1
reveals, political interest shows that its effect with electoral participation is
statistically significant and positively signed across all the global regions.
Moreover, though the country by country results did reveal the impact of
political interest on electoral participation tends to vary from one country
to another in the different regions, the clear message from the country by
country result is that the impact of political interest was statistically sig-
nificant across 62 per cent of the cases in all four regions (i.e. 16 of the 26
countries).
Second, turning to the other political-psychological variables (e.g. polit-
ical trust, internal, and external efficacy), the surprising finding regarding
these variables that are tested is that all these variables prove to be non-
significant for voting in three regions, sub-Saharan Africa, Latin America,
and East Asia. However, they do produce some level of consistency with
the other regions, notably Eastern Europe and Western Europe, by produ-
cing a number of parameters that run in the expected direction. Besides,
looking beyond the direction of the relationship, the political-psychological
model is revealed to be more consistent in post-Communist states as in
Western European democracies. In contrast, sub-Saharan Africa, Latin
America, and East Asia democracies prove to be unique. Thus, the most
appropriate conclusion would seem to be that political trust, internal and
external efficacy factors do not work well in new democracies, especially
those of sub-Saharan Africa, Latin America, and East Asia. How far this
might be down to the lack of identical question wording across the regions
in assessing the effects of these political-psychological variables, notably
internal and external efficacy on electoral participation, is impossible to
determine. Still, we certainly cannot regard them as important (for similar
results see, Carlin et al., (2015, p.47) for the Latin America region, Kuenzi
and Lambright (2011, p. 787) for sub- Saharan Africa, and Tambe (2016)
for East Asia.).
Alternative variables
Chapter 7 focused on an alternative explanation of electoral participation,
which constituted the fourth block to our casual recursive system. Among
them, I did consider three variables, notably: corruption, authoritarianism,
and the economy. As shown in Table 9.1, except for the effect of corruption
on vote choice, there appears to be little evidence of consistency as the effects
172 A comparative overview
tend to vary from one region to another. Corruption has a similar impact
on vote choice across three new democracies except for the Latin America
region. In these three regions, namely sub-Saharan Africa, East Asia, and
post-Communist European democracies, the effect of corruption on voting is
significant and negatively signed. This result, therefore, suggests that in new
democracies, voters’ perception of corruption tends to significantly decrease
turnout.
Next, regarding the effect of the economy on vote choice across the
different geo-political regions, Table 9.1 suggests our hypothesis is confirmed
only in post-communist European countries. Here, what is clear is that voters
who have a positive evaluation of their country’s economy are more likely
to participate. Apart from this, across the remaining three newer democra-
cies, sub-Saharan Africa, East Asia, and Latin America, I can only observe
some consistency in the direction of the effect of people’s perception of their
country’s economy. Still, even then, it is not always significant.3
Finally, I examine the effect of authoritarianism on electoral participation.
In both the theoretical and empirical chapters, I have shown a majority of
new democracies are now experiencing democratic erosion characterised by
the rise in media repression, curtailment of civil liberties, and the suppression
of oppositions. Relying on the Democratic Erosion Event dataset (DEED,
version 3), I have sought to examine how these authoritarian practices or
tendencies impact on vote choice. As shown in Table 9.1, the overall picture
depicts authoritarianism has a distinct effect on electoral participation across
new democracies. First, in post-communist Europe, democratic erosion, or
authoritarian tendencies is positive and significantly related to voter turnout.
Thus, the simple implication is that people will be more willing to engage in
voting and cast out such leaders as a way to protect democracies. Second, in
two newer democracies (i.e. Saharan Africa and Latin American countries)
democratic erosion or the subvention of democratic principles is negatively
signed, although statistically significant in sub-Saharan Africa. Given these
results, we can partially conclude any further increase with authoritarian
practices might pose a severe threat to electoral politics in these regions (e.g.
as voters might be less likely to participate).
Political-institutional variables
The last block of our causal recursive model (see Chapter 8) centres on the
political-institutional variables, particularly disproportionality, the closeness
of the election, concurrent elections, and compulsory voting. A picture of
the summary findings (see Table 9.1) suggests the impact of the political-
institutional model is rather mixed across the different regions. In some
respects, some institutional variables are more crucial in sub-Saharan Africa
and Latin America, and this is true of compulsory voting and the closeness
A comparative overview 173
of elections. For instance, in Latin America, voters are more likely to engage
when elections are compulsory and enforced with sanctions. In contrast, in
sub-Saharan Africa, individuals are more likely to participate when elections
are considered closed or competitive. However, our inability to empirically
test all the institutional variables hampers the analysis of institutional effects,
and, overall, it is hard to identify any clear patterns here. What I can say is
that this analysis does suggest that institutions are no guarantee of higher
turnout, but there is a clear need for more context-specific analysis across
and within countries. Detailed studies of particular cases should help in this
regard.
Benin – – × × × ×
Botswana × × × × × ×
Cape Verde × × × × ×
Ghana × × × × × ×
Sao Tome and × × × × × ×
Principe
Mauritius – – – – × ×
Namibia × × × × × ×
South Africa × × × × × ×
Argentina × × × × × × ×
Brazil × × × × × × ×
Chile × × × × × × ×
Uruguay × × × × × × ×
Costa Rica × × × × × × ×
El Salvador × × × × × × ×
Guyana × × × × × × ×
Panama × × × × × × ×
Taiwan × × × ×
Mongolia × × × ×
South Korea × × × ×
Czech Rep. × × – × × × × ×
Estonia – × × × × × ×
Hungary × × × × × × × ×
Lithuania – × × × × × ×
Poland × × × × × × ×
Slovenia × × × × × × ×
Source: European Social Survey (www.europeansocialsurvey.org/data/). Round 9 has already been initiated but data not yet released.
newgenrtpdf
Appendix 2: Questions wording from the four survey data
Table B1 Questions wording for the dependent and independent variables
Dependent variable
Voting Understanding that some people Did you vote in the last In talking to people about Some people do not
were unable to vote in the most presidential elections elections, we often find vote nowadays for one
recent national election in of (year of last that a lot of people reason or another.
[20xx], which of the following presidential elections)? were not able to vote Did you vote in the
statements is true for you? 1 = voted, 2 = did not because they were away last national election?
0 = You were not registered vote, 888888 = Don’t from home, they were Here the respondents
to vote, 1 = You voted in the know, 988888 = No sick or they just did not were invited to answer
elections, 2 = You decided not answer have time. How about with a simple Yes or
to vote, 3 = You could not find you? Did you vote in No response.
the polling station, 4 = You the most recent national
were prevented from voting, election, parliamentary
5 = You did not have time to or presidential? 1 = Yes,
vote, 6 = You did not vote 2 = No, 0 = not yet
because you could not find your eligible to vote, 7 = do
name in the voters’ register, not understand the
7 = Did not vote for some other question, 8 = can’t choose,
Appendices 183
reason, 8 = You were too young 9 = decline to answer.
to vote, 9 = Don’t know/can’t
remember.
(continued)
Table B1 Cont.
184
Africa Latin America East Asia Central and Eastern
Appendices
Afrobaometer, AmericasBarometer AsianBarometer European Countries
Round 6, 2016. Round 6, 2014 Wave 4, 2016 European Social Survey,
Round 8, 2016
Socio-demographics variables
Age How old are you? 18–105, 998– How old are you? 18–105 Actual age ranging from How old are you? 18–87,
999, -1 17–108
Gender Respondent’s gender: 1 = male, Respondent’s Respondent’s Respondent’s
2 = female gender: 1 = male, gender: 1 = male, sex: 1 = male,
2 = female 2 = female 2 = female
Religion People practise their religion What is your religion? About how often do you How often do you attend
in different ways. Aside from (1) Catholic, practise religious services religious services
weddings and funerals, how (2) Protestant or or rituals these days? apart from special
often do you personally engage Protestant non- Several times a day [01], occasions?
in religious practices like prayer, Evangelical, (3) Non- Once a day [02], Several Every day [01], More
reading a religious book, or Christian Eastern times a week [03], Once a than once a week [02],
attending a religious service or religion (Islam, week [04] Once a month Once a week [03], At
a meeting of a religious group? Buddhism, Hinduism, [05], Only during festivals least once a month
Religious practice: 0 = Never, Taoism, Confucianism, (or several times a year) [04], Only on special
1 = A few times a year, etc.). [06], Less often [07] and holy days [05], Less
2 = About once a month, Practically never [08] often [06], Never [07].
3 = About once a week, 4 = A
few times a week, 5 = About
once a day, 6 = More than
once a day, 7 = Respondent has
no religion, 9 = Don’t know
[DNR], -1 = Missing.
newgenrtpdf
Ethnicity/race Let us suppose that you had Do you consider yourselfWhat is your racial or ethnic Do you belong to a
to choose between being white, mestizo, background? 301 = Korea, minority ethnic group
a (Ghanaian/Kenyan/etc.) indigenous, black, 302 = Korea Chinese, in [country]? Yes
and being a (respondent’s mulatto, or of another 303 = Vietnamese, 1, No 2
identity group), which of race? [If respondent 501 = Khalkh, (Refusal) 7 (Don’t know)
these two groups do you feel says Afro-country, 502 = Kazak, 8.
most strongly attached to? mark (4) Black] 503 = others,
The response options range (1) White, (2) Mestizo, 701 = Hakka, 702 = Min-
across: 1 = Ethnic identity only, (3) Indigenous, nan, 703 = Mainlander,
2 = ethnic identity more than (4) Black 704 = Aboriginal,
national, 3 = national/ethnic 705 = Others,
identity equal, 4 = national 706 = Vietnamese,
identity more than ethnic and 707 = Indonesian,
5 = national identity only. 708 = Overseas Chinese,
709 = Filipino.
Place of Urban or rural primary sampling Which of the following Size of place: (1) National Which place best
residence unit: 1 = Urban, 2 = Rural. levels within the capital (metropolitan area) describes the area
country the respondent (2) Large city, (3) Medium where you live?
lives? 1 = Rural, city, (4) Small city, A big city [1], The
2 = Urban. (5) Rural area. suburbs or outskirts
of a big city [2],
A town or a small city
[3], A country village
[4], A farm or home in
the countryside [5].
Education What is your highest level of How many years of What is the highest level of What is the highest
Appendices 185
education? 0 = No formal schooling have you education that individuals level of education
schooling, 1 = Informal completed? The have completed or you have successfully
schooling only (including response ranges achieved: 1 = No formal completed? Ranges
Koranic schooling), 2 = Some from: primary, education, 2 = Incomplete from: Not completed
primary schooling, 3 = Primary secondary, university, primary/elementary, ISCED level 1 = 000, I
school completed, post-secondary not 3 = Complete
university).
(continued)
Table B1 Cont.
186
Africa Latin America East Asia Central and Eastern
Appendices
Afrobaometer, AmericasBarometer AsianBarometer European Countries
Round 6, 2016. Round 6, 2014 Wave 4, 2016 European Social Survey,
Round 8, 2016
4 = Intermediate school or Some primary/elementary, ISCED, completed
secondary school /high school, 4 = Incomplete secondary/ primary education
5 = Secondary school /high high school: technical = 113 to ISCED
school completed, 6 = Post- and vocational type, 6, Doctoral
secondary qualifications, 5 = Complete secondary/ degree = 800.
other than university e.g. a high school: technical and
diploma or degree from a vocational, 6 = incomplete
polytechnic or college, 7 = Some secondary/high school,
university, 8 = University 7 = Complete secondary/
completed, 9 = Post-graduate, high school, 8 = Some
99 = Don’t know [Do not university, 9 = University
read], 98 = Refused to answer, education completed,
-1 = Missing 10 = Post-graduate degree.
Mobilisation variables
Party Do you feel close to any Do you currently identify Among the political parties Is there a particular
identification particular political party? with a political party? listed here, which party political party you feel
0 = No (not close to any (1) Yes, (2) No. if any do you feel closest closer to than all the
party), 1 = Yes, (feels close to a (888888) Don’t know, to? Party A = 01, Party other parties? 1 = Yes
party), 8 = Refused to answer, (988888) No answer. B = 02, Party C = 03 etc. 1, 2 = No 2, (Refusal)
9 = Don’t know, -1 = Missing. Don’t feel close to any 7, (Don’t know) 8.
party = 99.
newgenrtpdf
Associational Let’s turn to your role in the I am going to read you Question asks whether Are you or have you
membership community. Now I am going a list of groups and respondents are ever been a member
to read out a list of groups organizations. Please members of any formal of a trade union or
that people join or attend. tell me if you attend organisations: Political similar organisation?
For each one, could you tell meetings of these parties, Residential and If Yes, is that currently
me whether you are an official organisations at least community associations, or previously? Yes,
leader, an active member, an once a week, once or Religious groups, Sports/ currently [1], Yes,
inactive member, or not a twice a month, once recreational clubs, previously [2], No [3].
member: Member of religious or twice a year, or Culture organisations,
group, Member of voluntary never. Meetings of any Charities, Public interest
association or community religious organization? groups, Labour unions,
group: 0 = Not a member, Meetings of a Farmer unions or
1 = Inactive member, 2 = Active parents’ association agricultural associations,
member, 3 = Official leader, at school? Meetings Professional organisations,
9 = Don’t know, 98 = Refused of a community, Business associations,
to answer, -1 = Missing. improvement Parent–Teacher
committee or Associations or PTA
association? Meetings producer cooperatives,
of a political Consumer cooperatives,
party or political organisations and Student
organisation? Meetings associations etc. 1 = Yes
of associations or and 2 = No, 8 = Can’t
groups of women or choose, 9 = Decline to
homemakers? answer.
Clientelism In your opinion, how often do And thinking about No question in survey. No question in survey.
the following things occur in the last presidential
Appendices 187
this country’s elections: Voters elections, did someone
are bribed: 0 = Never, offer you something,
1 = Sometimes, 2 = Often, like a favour, gift, or
3 = Always 9 = Don’t know, any other benefit in
98 = Refused to answer, return for your vote or
-1 = Missing. support? (1) Yes, (2) No,
(88) DK, (98) DA.
(continued)
newgenrtpdf
Table B1 Cont.
188
Africa Latin America East Asia Central and Eastern
Appendices
Afrobaometer, AmericasBarometer AsianBarometer European Countries
Round 6, 2016. Round 6, 2014 Wave 4, 2016 European Social Survey,
Round 8, 2016
Political-psychological variables
Political interest How interested would you say you How much interest do you How interested would you How interested would
are in public affairs? 0 = Not have in politics: a lot, say you are in politics? you say you are
at all interested, 1 = Not very some, little or none? Respondents had to in politics: very
interested, 2 = Somewhat (1) A lot, (2) Some, decide between four interested (1), quite
interested, 3 = Very interested, (3) Little, (4) None, categories: 1 = Not at all interested (2), hardly
9 = Don’t know, 98 = Refused (888888) Don’t know interested, 2 = Not very interested, (3) or, not
to answer, -1 = Missing. [DON’T READ], interested, 3 = Somewhat at all interested (4),
(988888) No answer interested and 4 = Very (Refusal) 7, (Don’t
[DON’T READ]. interested, 8 = Can’t choose, know) 8.
9 = Decline to answer.
Political trust How much do you trust each of the To what extent do you I am going to name a number Please tell me on a
following, or haven’t you heard trust the following: the of institutions. For each, score of 0–10 how
enough about them to say: Trust National Congress? please tell me how much much you personally
president, Trust parliament/ National Police? trust do you have in them? trust each of the
national assembly, Trust national Political parties? The president, courts, institutions I read out.
electoral commission, Trust the President/Prime national government, 0 means you do not
ruling party, Trust opposition Minister? 1 to 7 scale, political parties, parliament, trust an institution
political parties, Trust police, where 1 is “Not at all” civil service, military and at all, and 10 means
Trust army, Trust courts of and 7 is “A lot” police, local government, you have complete
law: 0 = Not at all, 1 = Just a and the election trust: Parliament,
little, 2 = Somewhat, 3 = A lot, commission. 1 = A great legal system, police,
9 = Don’t know/Haven’t heard deal of trust, 2 = Quite a politician, political
enough, 98 = Refused to answer, lot of trust, 3 = Not very parties.
-1 = Missing. much trust, 4 = None at all,
7 = Do not understand the
question, 8 = Can’t choose.
External efficacy Think about how elections work Those who govern this How well do you think the And how much would
in practice in this country. country are interested government responds to you say that the
How well do elections: Enable in what people like you what people want? Very political system in
voters to remove from office think. How much do responsive 1, Largely [country] allows
leaders who do not do what the you agree or disagree responsive 2, Not very people like you to
people want. 0 = Not at all well, with this statement? responsive 3, Not have an influence on
1 = Not very well, 2 = Well, You feel that you responsive at all 4. politics? Not at all 1,
3 = Very well, 9 = Don’t understand the most Very little 2, Some 3,
know, 98 = Refused to answer, important political A lot 4, A great deal 5.
-1 = Missing. issues of this country.
How much do you
agree or disagree with
this statement? Where
1 means “strongly
disagree” and 7 means
“strongly agree”.
A number in between
1 and 7 represents an
intermediate score.
Internal efficacy Think about how elections work Sometimes politics and And how confident
in practice in this country. How government seem so are you in your own
well do elections ensure voters’ complicated that a person ability to participate
views are reflected: 0 = Not like me can’t really in politics? 1 = Not
at all well, 1 = Not very understand what is going at all confident,
well, 2 = Well, 3 = Very well, on: 1 = Strongly agree, 2 = A little confident,
9 = Don’t know, 98 = Refused 2 = Agree, 3 = Disagree, 3 = Quite confident,
Appendices 189
to answer. 4 = Strongly disagree, 4 = Very confident,
7 = Don’t understand 5 = Completely
the question, 8 = Can’t confident,
choose, 9 = Decline to 7 = (Refusal),
answer. 8 = (Don’t know).
(continued)
newgenrtpdf
Table B1 Cont.
190
Africa Latin America East Asia Central and Eastern
Appendices
Afrobaometer, AmericasBarometer AsianBarometer European Countries
Round 6, 2016. Round 6, 2014 Wave 4, 2016 European Social Survey,
Round 8, 2016
Alternative variables
Country Looking back, how do you rate Do you think that the How would you describe the How satisfied with
economy economic conditions in this country’s current change in the economic present state of
country compared to 12 months economic situation is condition of your country economy in country?
ago? 1 = Much worse, better than, the same over the last year? Responses ranges from
2 = Worse, 3 = Same, 4 = Better, as or worse than it 1 = Much better, 2 = A 0–10. 0 = Extremely
5 = Much better, 9 = Don’t was 12 months ago? little better, 3 = About dissatisfied and
know, 98 = Refused to answer, (1) Better, (2) Same, the same, 4 = A little 10 = Extremely
-1 = Missing. (3) Worse, (888888) worse, 5 = Much worse, satisfied.
Don’t know, (988888) 7 = Do not understand the
No answer. question, 8 = Can’t choose,
9 = Decline to answer.
Corruption How many of the following Is corruption among Have you or anyone you know Based on the
people do you think are public officials personally witnessed an Transparency
involved in corruption, or very common? act of corruption or bribe- International
haven’t you heard enough about 1 = Very common, taking by a politician or Corruption
them to say: (a) Office of the 2 = Common, government official in the Perceptions Index.
presidency, (b) Members of 3 = Uncommon, past year? 1 = Personally The index measured
parliament, (c) Government 4 = Very common. witnessed, 2 = Told about the perceived level
officials, (d) Local government it by a family member of public/political
councillors, (e) Police, who personally witnessed, corruption on a scale
(f) Judges and magistrates, 3 = Told about it by a of 0 (highly corrupt)
and (g) Business executives. friend who personally to 100 (very clean).
The response options witnessed, 4 = Personally
were: 0 = None, 1 = Some of never witnessed, 5 = No
them, 2 = Most of them, 3 = All one I know has personally
of them. witnessed.
Authoritarianism What the overall erosion of a What the overall What the overall erosion of What the overall
country’s democracy on the erosion of a country’s a country’s democracy erosion of a country’s
same five-point scale identified democracy on the on the same five-point democracy on the
above: –0: No backsliding, same five-point scale scale identified above: – same five-point scale
and weak threat of future identified above: – 0: No backsliding, and identified above: –
backsliding. –1: There are 0: No backsliding, weak threat of future 0: No backsliding, and
precursors to backsliding, e.g. and weak threat of backsliding. –1: There are weak threat of future
the rise of extremist parties, future backsliding. – precursors to backsliding, backsliding. – 1: There
but erosion of democratic 1: There are precursors e.g. the rise of extremist are precursors to
institutions has not yet taken to backsliding, e.g. parties, but erosion of backsliding, e.g. the
place. –2: There is weak erosion the rise of extremist democratic institutions rise of extremist
of democratic institutions, parties, but erosion has not yet taken place. – parties, but erosion
perhaps the institutions being of democratic 2: There is weak erosion of democratic
eroded are not critical for the institutions has not of democratic institutions, institutions has not
functioning of democracy. – yet taken place. – perhaps the institutions yet taken place. –
3: There is moderate erosion 2: There is weak being eroded are not 2: There is weak
of democratic institutions. – erosion of democratic critical for the functioning erosion of democratic
4: There is severe erosion of institutions, perhaps of democracy. –3: There institutions, perhaps
democratic institutions; it is the institutions being is moderate erosion of the institutions being
unclear whether democracy will eroded are not critical democratic institutions. – eroded are not critical
recover. for the functioning 4: There is severe erosion for the functioning
of democracy. – of democratic institutions; of democracy. –
3: There is moderate it is unclear whether 3: There is moderate
erosion of democratic democracy will recover. erosion of democratic
institutions. – 4: There institutions. – 4: There
is severe erosion of is severe erosion
Appendices 191
democratic institutions; of democratic
it is unclear whether institutions; it is
democracy will recover. unclear whether
democracy will
recover.
(continued)
newgenrtpdf
Table B1 Cont.
192
Africa Latin America East Asia Central and Eastern
Appendices
Afrobaometer, AmericasBarometer AsianBarometer European Countries
Round 6, 2016. Round 6, 2014 Wave 4, 2016 European Social Survey,
Round 8, 2016
Political and institutional variables
Electoral system Disproportionality (least square Disproportionality (least Disproportionality (least Disproportionality (least
index) the calculation is square index) the square index) the square index) the
either based on Gallagher calculation is either calculation is either calculation is either
(2019) Election indices or based on Gallagher based on Gallagher based on Gallagher
self-calculated. (2019) Election indices (2019) Election indices or (2019) Election indices
or self-calculated. self-calculated. or self-calculated.
Concurrent Dummy coding: 1 = Concurrent Dummy coding: 1 = Dummy coding: 1 = Dummy coding: 1 =
elections elections, 0 = Otherwise. Concurrent elections, Concurrent elections, 0 = Concurrent elections,
0 = Otherwise. Otherwise. 0 = Otherwise.
Closeness of I measure closeness of election I measure closeness of I measure closeness of I measure closeness of
elections as the margin of victory election as the margin election as the margin election as the margin
for the winning candidate of victory for the of victory for the of victory for the
or over the runner-up in winning candidate winning candidate or winning candidate
presidential elections. While for or over the runner- over the runner-up in or over the runner-
parliamentary democracies, we up in presidential presidential elections. up in presidential
measure closeness of election elections. While While for parliamentary elections. While
as the difference in seat shares for parliamentary democracies, we measure for parliamentary
between the top two parties. democracies, we closeness of election as the democracies, we
That said, I coded the variable measure closeness difference in seat shares measure closeness
in such a way that a winning of election as the between the top two of election as the
margin of less 6% = 1, and a difference in seat shares parties. That said, I coded difference in seat
margin greater than 6% = 0. between the top two the variable in such a way shares between the top
that a winning margin of two parties. That said,
less 6% = 1, and a margin I coded the variable
greater than 6% = 0.
parties. That said, in such a way that a
I coded the variable winning margin of less
in such a way that a 6% = 1, and a margin
winning margin of less greater than 6% = 0.
6% = 1, and a margin
greater than 6% = 0.
Compulsory 0 = No compulsory election/ 0 = No compulsory 0 = No compulsory election/ 0 = No compulsory
voting compulsory election without election/compulsory compulsory election election/compulsory
enforcement, 1 = Compulsory election without without enforcement, election without
elections with enforcement. enforcement, 1 = Compulsory elections enforcement,
1 = Compulsory with enforcement. 1 = Compulsory
elections with elections with
enforcement. enforcement.
Appendices 193
newgenrtpdf
Table B2 Variable descriptions and coding schemes across the four regions and surveys
194
Variables/models Africa Latin America East Asia Post-communist European
Appendices
Afrobarometer, AmericansBarometer AsianBarometer countries
Round 6, 2016 Round 6, 2014 Wave 4, 2016 European Social Survey,
Round 8, 2016
Dependent variable
Electoral Dummy 1 = Yes, voted, 0 = No 1 = Yes, voted, 0 = No 1 = Yes, voted, 0 = No 1 = Yes, voted, 0 = No did
participation/ did not vote. did not vote. did not vote. not vote.
voting Those who were too Those who were too Those who were too Those who were too young
young to vote, young to vote, blank young to vote, blank to vote, blank votes, and
blank votes, and votes, and abstention votes, and abstention abstention responses are
abstention responses responses are coded as responses are coded as coded as missing values.
are coded as missing missing values. missing values.
values.
Socio-demographic variables
Months since the Scale The number of The number of months The number of months The number of months
last election months from the from the last national from the last national from the last national
last national election election date to the election date to the election date to the
date to the month month of the survey month of the survey month of the survey
of the survey date. date. date. date.
Age Scale Age of respondents in Age of respondents in Age of respondents in Age of respondents in
years. years. years. years.
Gender Dummy 1 = Male and 1 = Male and 0 = Female. 1 = Male and 0 = Female. 1 = Male and 0 = Female.
0 = Female.
Religion Scale/ Respondents indicate What is your Respondents indicate Respondents indicate
dummy how often they religion: 1 = Catholic, how often they attend how often they
attend religious 0 = Other. religious services or attend religious
practices: 0 = Never/ rituals: 0 = Never/less services: 0 = Never,
few times a year, often, 2 = Once a year/ 2 = less often, 3 = Only
2 = About once a several times a year, on special holidays,
month, 3 = Once 3 = Once a month/ 4 = Once a month,
a week/few times a several times a month, 5 = Once a week/
week, 4 = Once a 4 = Once a week/ everyday
day, 5 = More than several times a week,
once a day. 5 = Once a day/several
times a day.
Ethnicity/race Scale/ Salience of Others ethnic groups = 0, Nominal ethnic Nominal ethnic
dummy ethnicity: 1 = Only Blacks/Indigenous identity: 0 = minority identity: 0 = Minority
ethnic ID, people = 0.5, Mestizos/ ethnic group, ethnic group,
0.5 = Equal ethnic/ Mulattos = 1, 1 = majority ethnic 1 = Majority ethnic
national ID, Whites = 2. group. group.
0 = Only national
ID.
Place of residence Dummy 0 = Respondents living 1 = Respondents living 0 = Respondents living 0 = Respondents living
in medium /large in medium /large in medium /large in medium /large
cities or urban areas, cities or urban areas, cities or urban areas, cities or urban areas,
1 = Respondents 0 = Respondents living 1 = Respondents living 1 = respondents living
living small/rural small/rural areas. small/rural areas. small/rural areas.
areas.
Level of education 3-point 0 = Primary education, 0 = Primary education, 0 = Primary education, 0 = Primary education,
Appendices 195
scale 0.5 = Secondary 0.5 = Secondary 0.5 = Secondary 0.5 = Secondary
education, education, 1 = Higher education, 1 = Higher education, 1 = Higher
1 = Higher education. education. education.
education.
(continued)
newgenrtpdf
Table B2 Cont.
196
Variables/models Africa Latin America East Asia Post-communist European
Appendices
Afrobarometer, AmericansBarometer AsianBarometer countries
Round 6, 2016 Round 6, 2014 Wave 4, 2016 European Social Survey,
Round 8, 2016
Mobilisation variables
Party Dummy 1 = Closed parties ties, 1 = Closed parties ties, 1 = Closed parties ties, 1 = Closed parties ties,
identification 0 = No party ties. 0 = No party ties 0 = No party ties. 0 = No party ties
Associational Scale/ 0 = Not a member, Never = 0, 0.5 = Once/ 0 = Not a member, 0 = Not a member,
membership dummy 0.5 = Inactive twice a year, 1 = Once/ 1 = Member. 0.5 = Member
member, 1 = Active twice a month, previously, 1 = Member
member/official 2 = Once a week. currently.
leader.
Clientelism Dummy Respondents identify Respondents identify as N/A N/A
as having receive having receive gift for
gift for their their vote: 1 = Receive
vote: 1 = Receive a gift, 0 = Did not
a gift, 0 = Did not receive any gift/present.
receive any gift/
present).
Political-psychological variables
Political interest Scale 0 = Not interested, 0 = None/no interest, 0 = Not interested, 0 = Not at all interested,
0.5 = Not very 0.5 = Little interest, 0.5 = Not very 0.5 = Hardly interested,
interested, 1 = some interest, interested, 1 = Quite interested,
1 = Somewhat 2 = A lot of interest. 1 = Somewhat 2 = Very interested.
interested, 2 = Very interested, 2 = Very
interested. interested.
Political trust Scale 0 = No trust, Respondents say to 0 = None at all, A 10-point scale that
0.5 = Little trust, what extent they 0.5 = Not very much measured respondents
1 = Some trust, trust political/other trust, 1 = Quite a lot of trust in society (0 to 10).
2 = A lot of trust. institutions (1 to 7). trust, 2 = A great deal
of trust.
External efficacy Scale Respondent indicates Respondents indicate Government leaders Respondent indicates
how well elections whether those who implement what voters how much the political
enable voters to govern their country want (0 = Not at all system in country allows
removed leaders are interested in what responsive, 0.5 = Not people to influence
from office (0 = Not people like you think very responsive, politics: 0 = Not at
at all well, 0.5 = Not (1 to 7). 1 = Largely responsive, all, 0.5 = Very little,
very well, 1 = Well, 2 = Very responsive). 1 = Some, 2 = A lot,
2 = Very well). 3 = A great deal.
Internal efficacy Scale Respondents indicate Respondents’ subjective Respondents see politics/ Respondents indicate
how well the feel assessment of their government as how confident they
elections allow understanding of complicated that a are in their ability to
voters view to be the most important person like me can’t participate in politics?
reflected. (0 = Not political issues (i.e. really understand 0 = Not at all confident,
at all well, 0.5 = Not measured on a 7-point what is going on 0.5 = A little confident,
very well, 1 = Well, scale). (0 = Strongly 1 = Quite confident,
2 = Very well.) agree, 0.5 = Agree, 2 = Very confident,
1 = disagree, 3 = Completely
2 = Strongly disagree). confident.
Alternative explanatory variables
Country economy Scale Sociotropic Sociotropic retrospective Sociotropic retrospective Sociotropic retrospective
retrospective economic perception economic perception economic perception
economic perception [0 = worse, 0.5 = Same, [0 = Worse, 0.5 = Same, [0 = Worse, 0.5 = Same,
[0 = worse, 1 = Better]. 1 = Better] 1 = Better].
0.5 = Same,
1 = Better]
Appendices 197
(continued)
newgenrtpdf
Table B2 Cont.
198
Variables/models Africa Latin America East Asia Post-communist European
Appendices
Afrobarometer, AmericansBarometer AsianBarometer countries
Round 6, 2016 Round 6, 2014 Wave 4, 2016 European Social Survey,
Round 8, 2016
Corruption Scale/ Respondent indicates Respondents indicate Respondents indicate Transparency Corruption
perception dummy how many people if corruption among if they or anyone Index. The index
they think are public officials is have personally measured the perceived
involved in very common? witnessed an act of level of public/political
corruption, or 1 = Very common, corruption or bribe- corruption on a scale of
haven’t heard 0 = Uncommon. taking by a politician 0 (highly corrupt) to 100
enough about them or government (very clean). Scores are
to say: 0 = None, official in the past rescaled from 0 –1.
1 = Some of them, year. 1 = Witness
2 = Most of them, corruption, 0 = Never
3 = All of them. witness corruption.
Authoritarianism Dummy What is the What is the overall What is the overall What is the overall
overall erosion erosion of a erosion of a erosion of a country’s
of a country’s country’s democracy? country’s democracy? democracy? 1 = Country
democracy? 1 = Country 1 = Country experiences democratic
1 = Country experiences democratic experiences democratic erosion, 0 = Country
experiences erosion, 0 = Country erosion, 0 = Country never experiences
democratic erosion, never experiences never experiences democratic erosion.
0 = Country democratic erosion. democratic erosion.
never experiences
democratic erosion.
Political and institutional variables
Electoral system Scale Least squares index of Least squares index of Least squares index of Least squares index of
disproportionality disproportionality disproportionality disproportionality
(0 = Perfect (0 = Perfect (0 = Perfect (0 = Perfect
proportionality proportionality proportionality proportionality
to 100 maximum to 100 maximum to 100 maximum to 100 maximum
disproportionality) disproportionality) disproportionality) disproportionality)
Concurrent Dummy 1 = Concurrent 1 = Concurrent elections, 1 = Concurrent elections, 1 = Concurrent elections,
elections elections, 0 = Otherwise 0 = Otherwise 0 = Otherwise
0 = Otherwise
Closeness of Dummy 1 = Closed/competitive 1 = Closed/competitive 1 = Closed/competitive 1 = Closed/competitive
elections elections, 0 = Not elections, 0 = Not closed. elections, 0 = Not elections, 0 = Not closed.
closed. closed.
Compulsory voting Dummy N/A 0 = No compulsory N/A 0 = No compulsory
election/compulsory election/compulsory
election without election without
enforcement, enforcement,
1 = Compulsory 1 = Compulsory
elections with elections with
enforcement. enforcement.
Appendices 199
200 Appendices
Appendix 3: Summary descriptive statistics of variables and
individual country –results
Table C1 Descriptive statistics for dependent and socio-demographic variables
Western Europe
Voting 0.82 0.37 0 1
Age 48.4 18.5 15 100
Gender 0.49 0.49 0 1
Religiosity 2.08 1.80 0 5
Place of residence 0.54 0.49 0 1
Ethnicity/race 0.94 0.22 0 1
Education 0.84 0.80 0 2
Sub-Saharan Africa
Voting 0.72 0.39 0 1
Age 40.4 54.2 18 105
gender 0.49 0.50 0 1
Religiosity 2.84 1.45 0 5
Place of residence 0.45 0.49 0 1
Ethnicity/race 0.32 0.34 0 2
Education 0.21 0.30 0 2
Latin America
Voting 0.76 0.43 0 1
Age 41.5 16.6 16 99
gender 0.48 0.49 0 1
Religiosity 0.56 0.49 0 1
Place of residence 0.27 0.44 0 1
Ethnicity/race 1.25 0.65 0 4
Education 0.55 0.38 0 2
East Asia
Voting 0.83 0.35 0 1
Age 43.5 15.4 18 93
Gender 0.48 0.49 0 1
Religiosity 2.50 1.32 0 5
Place of residence 0.81 0.38 0 1
Ethnicity/race 0.85 0.35 0 1
Education 1.07 0.79 0 2
Eastern Europe
Voting 0.67 0.47 0 1
Age 47.6 18.5 15 97
gender 0.45 0.49 0 1
Religiosity 2.29 1.78 0 5
Place of residence 0.41 0.49 0 1
Ethnicity/race 0.93 0.25 0 1
Education 0.68 0.78 0 2
Notes: Descriptives statistics across the different regions are drawn from the Afrobarometer,
AmericasBarometer, AsianBarometer, and the European Social Survey. The observations are all
weighted using the different surveys weights.
Appendices 201
Table C2 The impact of socio-
demographic variables on vote choice: individual
country results
Sub-Saharan Africa
Benin 0.35** 0.46*** 0.06 -0.67*** 0.51*** -0.88**
(0.16) (0.18) (0.06) (0.24) (0.18) (0.39)
Botswana 1.01*** -0.50** 0.15*** -0.28 0.39** -0.39
(0.19) (0.20) (0.05) (0.33) (0.22) (0.35)
Cape Verde 0.48** -0.18 0.06 -0.43* -0.11 0.17
(0.14) (0.16) (0.07) (0.36) (0.19) (0.25)
Ghana 0.39** -0.14 0.07** 0.22 0.23** 0.27
(0.10) (0.12) (0.04) (0.17) (0.12) (0.23)
Mauritius 0.54*** 0.26 0.03 -0.34 0.01 -0.15
(0.19) (0.20) (0.08) (0.33) (0.21) (0.30)
Namibia 0.95*** 0.10 0.05 -0.11 0.11 -0.10
(0.17) (0.18) (0.09) (0.28) (0.18) (0.30)
Sao Tome and 0.12 -0.50** 0.05 -0.41 0.50* -0.29
Principe (0.22) (0.02) (0.10) (0.47) (0.30) (0.43)
South Africa 0.64*** -0.09 0.05 -0.24* 0.27** 0.47**
(0.14) (0.11) (0.04) (0.14) (0.12) (0.26)
Latin America
Argentina 0.05*** 0.05 0.15 -0.05 0.35* 1.34***
(0.01) (0.13) (0.14) (0.12) (0.20) (0.21)
Brazil 0.08*** -0.23** 0.23 0.17 -0.24 0.93***
(0.01) (0.14) (0.14) (0.20) (0.20) (0.26)
Chile 0.05*** -0.06 0.35*** 0.06 0.76*** 1.03***
(0.01) (0.12) (0.13) (0.11) (0.22) (0.21)
Costa Rica 0.02*** -0.19 0.41*** 0.56*** 0.13 0.85***
(0.00) ((0.12) (0.12) (0.11) (0.12) (0.17)
El-Salvador 0.03*** 0.08 0.39*** -0.24** 0.01 0.74***
(0.01) (0.13) (0.13) (0.12) (0.14) (0.18)
Guyana 0.05*** -0.24** 0.16 -0.54** -0.03 0.35
(0.01) (0.12) (0.25) (0.15) (0.13) (0.22)
Panama 0.10*** -0.16 0.06 -0.03 0.39*** 0.14
(0.01) (0.13) (0.14) (0.12) (0.15) (0.22)
Peru 0.08*** -0.18 0.06 -0.02 0.08 0.65**
(0.01) (0.16) (0.19) (0.15) (0.05) (0.25)
Uruguay 0.11*** -0.39** 0.29 -0.18 -0.36 0.30
(0.01) (0.16) (0.19) (0.16) (0.34) (0.24)
East Asia
Taiwan 0.05*** -0.24* 0.16*** -0.26 0.20 0.43***
(0.01) (0.14) (0.05) (0.22) (0.19) (0.09)
Mongolia 0.06*** -0.08 0.14 -0.37 0.48 0.32**
(0.01) (0.22) (0.13) (0.35) (0.32) (0.016)
South Korea 0.08*** 0.13 -0.00 0.49 0.06 0.45***
(0.01) (0.16) (0.09) (0.24) (0.27) (0.15)
Post-Communist Europe
Czech Republic 0.03*** 0.46*** 0.13*** 0.70** 0.02 0.68***
(0.00) (0.09) (0.06) (0.30) (0.10) (0.07)
(continued)
202 Appendices
Table C2 Cont.
Cell entries are multivariate logit coefficients; standard errors are in parentheses. All relationships
are Significant: *significant at 10%, **significant at 5%, ***significant at 1%. The observations
are all weighted using the different surveys weights.
Western Europe
Party identification 0.52 0.50 0.00 1.00
Associational membership 0.33 0.42 0.00 2.00
Clientelism N/A N/A N/A N/A
Sub-Saharan Africa
Party identification 0.58 0.49 0.00 1.00
Associational membership 0.22 0.38 0.00 2.00
Clientelism 0.75 0.43 0.00 1.00
Latin America
Party identification 0.36 0.48 0.00 1.00
Associational membership 0.00 1.00 -0.90 3.82
Clientelism 0.04 0.19 0.00 1.00
East Asia
Party identification 0.58 0.49 0.00 1.00
Associational membership 0.44 0.49 0.00 1.00
Clientelism N/A N/A N/A N/A
Eastern Europe
Party identification 0.36 0.47 0.00 1.00
Associational membership 0.19 0.29 0.00 2.00
Clientelism N/A N/A N/A N/A
Notes: Descriptives statistics across the different regions are drawn from the Afrobarometer,
AmericasBarometer, AsianBarometer, and the European Social Survey. The observations are all
weighted using the different surveys weights.
Appendices 203
Table C4 The Impact of mobilisation variables on vote choice: individual-country
results
Sub-Saharan Africa
Benin 0.23 1.07*** (0.26) -0.63** (0.33)
(0.20)
Botswana 0.63*** 1.42*** (0.35) -0.35 (0.24)
(0.27)
Cape Verde 0.46 0.46 (0.72) -0.94 0.41
(0.52)
Ghana 0.93*** 0.08 (0.17) -0.42* (0.22)
(0.14)
Mauritius 1.35*** 0.91** (0.47) 0.67** (0.32)
(0.37)
Namibia 0.84*** 0.35 (0.30) -0.07 (0.20)
(0.24)
Sao Tome and Principe 0.89*** -0.05 (0.40) -0.07 (0.41)
(0.29)
South Africa 0.09*** 0.28* (0.16) -0.44*** (0.13)
(0.13)
Latin America
Argentina 0.68*** 0.24*** (0.11) 0.99 (0.62)
(0.18)
Brazil 0.41*** 0.16** (0.09) 0.18 (0.22)
(0.18)
Chile 1.23*** 0.25*** (0.08) -0.51 (0.48)
(0.25)
Costa Rica 1.89*** 0.15*** (0.08) 0.28 (0.46)
(0.14)
El-Salvador 0.59*** 0.45*** (0.09) -0.08 (0.33)
(0.13)
Guyana 0.87*** 0.35*** (0.08 -0.09 (0.46)
(0.24)
Panama 1.25*** 0.13*** 0.05) 0.11 (0.34)
(0.15)
Peru 0.01 0.32*** (0.12) 0.75** (0.42)
(0.24)
Uruguay 0.11 0.23** (0.12) 0.33 (0.52)
(0.18)
East Asia
Taiwan 1.36*** 0.01 (0.15) N/A
(0.17)
Mongolia 0.71*** -0.38 (0.23) N/A
(0.26)
South Korea 0.82*** -0.25 (0.19) N/A
(0.20)
(continued)
204 Appendices
Table C4 Cont.
Post-communist Europe
Czech Republic 2.10*** 0.22 (0.20) N/A
(0.12)
Estonia 1.03*** 0.73*** (0.25) N/A
(0.13)
Hungary 2.32*** 0.39 (0.27) N/A
(0.18)
Lithuania 2.49*** 1.14*** (0.23) N/A
(0.17)
Poland 1.15*** 0.22 0.25 N/A
(0.16)
Slovenia 1.10*** 0.43** (0.21) N/A
(0.16)
Cell entries are multivariate logit coefficients; standard errors are in parentheses. All relationship
are Significant: *significant at 10%, **significant at 5%, ***significant at 1%. This model control
for the socio-demographic variables.
Appendices 205
Table C5 Descriptive statistics for political-psychological variables
Western Europe
Political interest 0.84 0.58 0 2
Political trust 0.21 0.95 -2.26 2.60
Internal efficacy 0.74 0.70 0 3
External efficacy 0.72 0.58 0 3
Sub-Saharan Africa
Political interest 0.93 0.72 0 2
Political trust 0.007 0.96 -1.82 1.76
Internal efficacy 0.84 0.55 0 2
External efficacy 2.83 0.60 0 2
Latin America
Political interest 0.59 0.60 0 2
Political trust 0.02 0.98 -1.94 2.62
Internal efficacy 3.93 1.77 1 7
External efficacy 3.37 1.86 1 7
East Asia
Political interest 0.65 0.46 0 2
Political trust -0.53 0.79 -2.53 2.33
Internal efficacy 0.72 0.52 0 2
External efficacy 0.58 0.35 0 2
Eastern Europe
Political interest 0.59 0.48 0 2
Political trust -0.21 0.98 -2.26 2.6
Internal efficacy 0.42 0.55 0 3
External efficacy 0.47 0.47 0 3
Notes: Descriptives statistics across the different regions are drawn from the Afrobarometer,
AmericasBarometer, AsianBarometer, and the European Social Survey. The observations are all
weighted using the different surveys weights.
206 Appendices
Table C6 The impact of political-psychological variables on vote choice: individual
country results
Sub-Saharan Africa
Benin 0.01 0.26*** 0.18 0.01
(0.15) (0.11) (0.25) (0.23)
Botswana -0.24 0.17 0.50* -0.23
(0.18) (0.26) (0.31) (0.27)
Cape Verde 0.75 -0.12 0.62 0.00
(0.48) (0.39) (0.60) (0.49)
Ghana 0.03 -0.05 0.08 -0.04
(0.10) (0.07) (0.15) (0.14)
Mauritius 0.52** 0.14 -0.19 0.54*
(0.22) (0.17) (0.35) (0.35)
Namibia 0.96*** -0.01 0.09 -0.05
(0.17) (0.14) (0.24) (0.21)
Sao Tome and Principe 0.25 -0.07 -0.62 0.27
(0.24) (0.22) (0.39) (0.37)
South Africa 0.26*** 0.09 0.05 0.25*
(0.10) (0.08) (0.15) (0.14)
Latin America
Argentina -0.04 -0.01 -0.05 -0.06
(0.15) (0.09) (0.05) (0.05)
Brazil 0.12 -0.04 -0.006 0.01
(0.15) (0.08) (0.05 (0.04)
Chile 1.35*** 0.25*** 0.01 -0.05
(0.17) (0.09) (0.04) (0.04)
Costa Rica 1.10*** -0.21*** 0.01 -0.02
(0.16) (0.08) (0.05) (0.04)
El-Salvador 0.17 -0.01 0.09*** -0.00
(0.12) (0.08) (0.04) (0.04)
Guyana 0.27** -0.14 -0.03 0.06
(0.18) (0.10) (0.05) (0.05)
Panama 0.72*** 0.04 0.08** -0.06
(0.17) (0.09) (0.05) (0.05)
Peru 0.07 -0.12 -0.01 0.02
(020) (0.11) (0.06) (0.05)
Uruguay 0.21 -0.08 0.01 0.04
(0.17) (0.11) (0.05) (0.05)
East Asia
Taiwan 1.45*** -0.21 0.24 0.02
(0.45) (0.26) (0.50) (0.25)
Mongolia 1.10*** -0.28 0.05 -0.04
(0.37) (0.19) (0.26) (0.04)
South Korea 0.76*** 0.09 -0.01 0.24
(0.25) (0.13) (0.17) (0.32)
Post-Communist Europe
Czech Republic 1.38*** 0.15** 0.37** 0.26*
(0.19) (0.06) (0.16) (0.16)
Appendices 207
Table C6 Cont.
Cell entries are multivariate logit coefficients; standard errors are in parentheses. All relationships
are Significant: *significant at 10%, **significant at 5%, ***significant at 1%. The model control
for both the socio-demographic and mobilisation variables.
Western Europe
Corruption 0.22 0.42 0.00 1.00
Authoritarianism N/A N/A N/A N/A
Economic perception 0.54 0.40 0.00 2.00
Sub-Saharan Africa
Corruption -0.09 0.96 -2.09 2.27
Authoritarianism 0.37 0.48 0.00 1.00
Economic perception 0.47 0.41 0.00 2.00
Latin America
Corruption 0.81 0.39 0.00 1.00
Authoritarianism 0.66 0.47 0.00 1.00
Economic perception 0.36 0.36 0.00 2.00
East Asia
Corruption 0.17 0.37 0.00 1.00
Authoritarianism 0.66 0.47 0.00 1.00
Economic perception 0.36 0.39 0.00 2.00
Eastern Europe
Corruption 0.81 0.38 0.00 1.00
Authoritarianism 0.42 0.49 0.00 1.00
Economic perception 0.41 0.39 0.00 2.00
Notes: Descriptives statistics across the different regions are drawn from the Afrobarometer,
AmericasBarometer, AsianBarometer, and the European Social Survey. The observations are all
weighted using the different surveys weights.
208 Appendices
Table C8 Descriptive statistics for political-institutional variables
Western Europe
Disproportionality 6.93 5.39 0.99 17.34
Closeness of election 0.44 0.49 0.00 1.00
Concurrent election 0.25 0.44 0.00 1.00
Compulsory voting 0.08 0.28 0.00 1.00
Sub-Saharan Africa
Disproportionality 8.47 7.25 0.37 20.44
Closeness of election 0.25 0.43 0.00 1.00
Concurrent election 0.25 0.43 0.00 1.00
Compulsory voting N/A N/A N/A N/A
Latin America
Disproportionality 5.28 3.24 0.81 10.23
Closeness of election 0.33 0.47 0.00 1.00
Concurrent election 0.77 0.42 0.00 1.00
Compulsory voting 0.44 0.49 0.00 1.00
East Asia
Disproportionality 9.33 1.89 7.15 11.77
Closeness of election N/A N/A N/A N/A
Concurrent election 0.33 0.47 0.00 1.00
Compulsory voting N/A N/A N/A N/A
Eastern Europe
Disproportionality 9.42 5.11 2.34 17.80
Closeness of election 0.58 0.49 0.00 1.00
Concurrent election N/A N/A N/A N/A
Compulsory voting N/A N/A N/A N/A
Notes: Descriptives statistics across the different regions are drawn from the Afrobarometer,
AmericasBarometer, AsianBarometer, and the European Social Survey. The observations are all
weighted using the different surveys weights.
References
Aarts, Kees, and Bernhard Wessels (2005). Electoral turnout. In Jacques Thomassen
(ed.), The European Voter. New York: Oxford University Press.
Abramson, P. R., and J. H. Aldrich (1982). The decline of electoral participation in
America. American Political Science Review 76, 502–521.
Achen, Christopher H. and T. Y. Wang (2017). The Taiwan Voter. Ann Arbor: The
University of Michigan Press.
Acton, Ciaran, Robert L. Miller, D. Fullerton, John Maltby, and Ciaran Acton (2009).
SPSS Statistics for Social Scientists (2nd ed.). Basingstoke: Palgrave Macmillan.
Ades, A., and R. Di Tella (1996). The causes and consequences of corruption: A review
of recent empirical contributions. Institute of Development Studies 27(2), 6–11.
Afrobarometer Data (2016). Round 6. Retrieved from www.afrobarometer.org.
Ahn, Chung-si and Jaung Hoon (2000). South Korea. In I. Marsh, J. Blondel and
T. Inoguchi, Democracy, Governance and Economic Performance. Tokyo, Japan,
New York, NY: United Nations University Press.
Ahuja, Amit, and Pradeep K. Chhibber (2012). Why the poor vote in India: “If I don’t
vote, i am dead to the state”. Studies in Comparative International Development
47(4), 389–410.
Albacete G. G. (2014). Young People’s Political Participation in Western Europe:
Continuity or Generational Change?. Basingstoke and New York: Palgrave
Macmillan.
Albritton, Robert B. (2004). Cambodia in 2003: On the road to democratic consolida-
tion. Asian Survey 44(1), 102–109.
Aldrich, J. (1993). Rational choice and turnout. American Journal of Political Science
37, 246–278.
Alexander, A. C., and H. Coffé (2018). Women’s political empowerment through
public opinion research: The citizen perspective. In A. C. Alexander, C.
Bolzendahl, and F. Jalalzai (eds), Measuring Women’s Political Empowerment
across the Globe: Strategies, Challenges and Future Research. Basingstoke: Palgrave
Macmillan.
Almond, Gabriel A., and Sidney Verba (1963). The Civic Culture: Political Attitudes
and Democracy in Five Nations. Newbury Park: Sage Publications.
Ames, Barry (2001). The Deadlock of Democracy in Brazil. Ann Arbor: University of
Michigan Press.
Ames, Barry, Andy Baker, and Lucio R. Rennó (2008). The quality of elections in
Brazil: Policy, performance, pageantry, or pork? In Peter Kingstone and Timothy
Power (eds), Democratic Brazil Revisited. Pittsburgh: University of Pittsburgh Press.
210 References
Anderson, C. J., and M. M. Singer 2008. The sensitive left and the impervious
right: multilevel models and the politics of inequality, ideology, and legitimacy in
Europe. Comparative Political Studies 41, 564–569.
Andreas Hadjar, and Michael Beck (2010). Who does not participate in elections in
Europe and why is this?. European Societies 12(4), 521–542.
Arzheimer, K., J. Evans, and M. S. Lewis-Beck (2017). The SAGE Handbook of elect-
oral behaviour (Vols. 1–2). London: Sage.
Baker, Andy, Barry Ames, and Lucio R. Rennó (2006). Social context and cam-
paign volatility in new democracies: Networks and neighborhoods in Brazil’s 2002
elections. American Journal of Political Science 50, 382–399.
Balch, George I. (1974). Multiple indicators in survey research: The concept “sense of
political efficacy”. Political Methodology 1, 1–43.
Barnes, S. H. (2006). The changing political participation of postcommunist citizens.
International Journal of Sociology 36, 2.
Barnes, T. D., and S. M. Burchard (2013). Engendering politics: The impact of
descriptive representation on women’s political engagement in sub-Saharan Africa.
Comparative Political Studies 46, 767–790.
Beauregard, Katrine (2014). Gender, political participation and electoral systems: A
cross-national analysis. European Journal of Political Research 53, 617–634.
Becker, R. (2004). Voter turnouts in East and West Germany. German Politics 13, 1–19.
Beer, S. H. (1974). Modern Political Development. New York: Random House.
Bélanger, E., and Nadeau, R. (2005). Political trust and the vote in multiparty
elections: The Canadian case. European Journal of Political Research 44, 121–146.
Bennett, Stephen E. (1986). Apathy in America, 1960–1984: Causes and Consequences
of Citizens’ Political Indifference. New York: Transnational Publisher, Inc.
Benton, Allyson (2005). Dissatisfied Democrats or retrospective voters?. Comparative
Political Studies 38, 417–442.
Berelson, Bernard R., Paul F. Lazarsfeld, and William N. McPhee (1954). Voting: A
Study of Opinion Formation in a Presidential Campaign. Chicago: University of
Chicago Press.
Berman, B., D. Eyoh, and W. Kymlicka (eds) (2004). Ethnicity and Democracy in
Africa. Oxford: James Currey.
Bermeo, Nancy (2016). On democratic backsliding. Journal of Democracy 27(1), 5–19.
Bernhagen, P. and Michael Marsh (2007). Voting and protesting: Explaining citizen
participation in old and new European democracies. Democratization 14(1), 44–72.
Bhatti, Y., and K. M. Hansen (2012). The effect of generation and age on turnout
to the European parliament –how turnout will continue to decline in the future.
Electoral Studies 31(2), 269–272.
Bhavnani, R., and D. Miodownik (2009). Ethnic polarization, ethnic salience, and civil
war. Journal of Conflict Resolution 53(1), 30–49.
Birch, S. (2010). Perceptions of electoral fairness and voter turnout. Comparative
Political Studies 43, 1601–1622.
Birch, Sarah (2009). Full Participation: A Comparative Study of Compulsory Voting.
Manchester: Manchester University Press.
Birch, Sarah (2018). Elections and Electoral Participation. In Adam Fagan, and
Petr Kopecký (eds), The Routledge Handbook of East European Politics. London:
Routledge.
Black, J. (1991). Reforming the context of the voting process in Canada: Lessons
from other democracies. In H. Bakvis (ed.), Voter Turnout in Canada. Toronto:
Dundurn Press.
References 211
Blais, André (2000). To Vote or Not to Vote –The Merits and Limits of rational
Choice Theory. Pittsburgh: University of Pittsburgh Press. Choice Theory.
Pittsburgh: University of Pittsburgh Press.
Blais, A. (2006). What affects voter turnout? Annual Review of Political Science 9,
111–125.
Blais, A., and K. Aarts (2006). Electoral Systems and Turnout. Acta Politica 41, 180–
196. doi:10.1057/palgrave.ap.5500148.
Blais A. and J.-F. Daoust (2020). The Motivation to Vote. Vancouver: UBC Press.
Blais, A. and A. Dobrzynska (1998). Turnout in electoral democracies. European
Journal of Political Research 33, 239–261.
Blais, A. and K. Carty (1990). Does proportional representation foster voter turnout?
European Journal of Political Research 18, 167–181..
Blais, A., E. Gidengil, N. Nevitte, and R. Nadeau (2004). Where does turnout decline
come from? European Journal of Political Research 43(2), 221–236.
Blais, André, Elisabeth Gidengil, Richard Nadeau, and Neil Nevitte (2002). Anatomy
of a Liberal Victory: Making Sense of the Vote in the 2000 Canadian Election.
Peterview: Broadview Press.
Blaydes, l. (2011). Elections and Distributive Politics in Mubarak’s Egypt. Cambridge:
Cambridge University Press.
Bleck, J., and N. Van de Walle (2018). Electoral Politics in Africa since 1990: Continuity
in Change. Cambridge: Cambridge University Press.
Boas, Taylor, and Amy Erica Smith (2015). Religion and the Latin American voter. In
Ryan E. Carlin, Matthew M. Singer, and Elizabeth J. Zechmeister (eds), The Latin
American Voter. Ann Arbor: University Michigan Press.
Boix, C. (2011). Democracy, development and the international system. American
Political Science Review 105(4), 809–828.
Boix, Carles, Michael K. Miller, and Sebastian Rosato (2013). A complete data set
of political regimes, 1800–2007. Comparative Political Studies 46(12), 1523–1554.
Bollen, K. A. (1993). Liberal democracy: Validity and source biases in cross-national
measures. American Journal of Political Science 37, 1207–1230.
Booth, J. A., and M. A. Seligson (2008). Inequality and democracy in Latin
America: Individual and contextual effects of wealth on political participation.
In A. Krishna (ed.), Poverty, Participation and Democracy: A Global Perspective.
New York: Cambridge University Press.
Boulding, Carew, and David S. Brown. (2013). Do political parties matter for
turnout: Number of parties, electoral rules and local elections in Brazil and Bolivia.
Party Politics 21(3), 404–416.
Brader, T., J. A. Tucker, and D. Duell (2013). Which parties can lead opinion?
Experimental evidence on partisan cue taking in multiparty democracies.
Comparative Political Studies 46(11), November, 1485–1517.
Brady, H. (1999). Political participation. In J. P. Robinson, P. R. Shaver, and L. S.
Wrightsman (eds), Measures of Political Attitudes. San Diego, CA: Academic Press.
Brady, Henry E., Sidney Verba, and Kay Lehman Schlozman (1995). Beyond SES: A
resource model of political participation. The American Political Science Review
89(2), 271–294.
Bratton, M. (1999) Political participation in a new democracy –institutional consider-
ations from Zambia, Comparative Political Studies 32, 549–588.
Bratton, M. (1998). Second elections in Africa. Journal of Democracy 9(3), 51–66.
Bratton, M. (2003). Islam, democracy and public opinion in Africa. African Affairs
102, 493–501.
212 References
Bratton, M. (2008). Vote buying and violence in Nigerian election campaigns. Electoral
Studies, 27, 621–632.
Bratton, M. (2013). Voting and Democratic Citizenship in Africa. Boulder: Lynne
Rienner Publishers.
Bratton, M., R. Bhavnani, and C. Tse-Hsin (2012) Voting intentions in Africa: Ethnic,
economic or partisan? Commonwealth and Comparative Politics 50(1), 27–52.
Bratton, M., Y. Chu, and M. Lagos (2010) Who votes? Implications for new democra-
cies. Taiwan Journal of Democracy 6(1), 1–30.
Bratton, M., R. Mattes, and E. Gyimah-Boadi (2005). Public Opinion, Democracy and
Market Recovery in Africa. Cambridge: Cambridge University Press.
Braungart, R. G., and Braungart, M. M. (1986). Life course and generational politics.
Annual Review of Sociology 12, 205–231.
Brigevich, Anna (2008). The Post- Cold War Wave of Democratization: Regime
Transitions in Sub-Saharan Africa and Postcommunist States Compared. A thesis
submitted to the faculty of the University of North Carolina at Chapel Hill in
partial fulfillment of the requirements for the degree of Master of Arts in the
Department of Political Science.
Brockington, D. (2004). The paradox of proportional representation: The effect of
party systems and coalitions on individuals’ electoral participation. Political
Studies 52, 469–490.
Broughton, David and Hans-Martien ten Napel (2000). Religion and Mass Electoral
Behaviour in Europe. London: Routledge/ European Consortium of Political
Research.
Bufacchi, V. (2001). Voting, rationality and reputation. Political Studies 49, 714–729.
Bühlmann, M. and Freitag, M. (2006). Individual and contextual determinants of
electoral participation. Swiss Political Science Review 12, 13–48.
Bunce, Valerie (1999). The political economy of postsocialism. Slavic Review 58(4),
756–793.
Burden, B. C. (2009) The dynamic effects of education on voter turnout. Electoral
Studies 28(4), December, 540–549.
Burkhart, R. E., and M. Lewis Beck (1994). Comparative democracy: The economic
development thesis. American Political Science Review 88, 903–910.
Burnet, J. (2011). Women have found respect: Gender quotas, symbolic representation
and female empowerment in Rwanda. Politics & Gender 7, 303–334.
Burns, N., K. L. Schlozman, and S. Verba (2001). The Private Roots of Public
Action: Gender, Equality and Political Participation. Cambridge, MA: Harvard
University Press.
Bustikova, L., and E. Zechmeister (2017). Voting in new(er) democracies. In K.
Arzheimer, J. Evans and M. S. Lewis-Beck (eds), The SAGE Handbook of Electoral
Behaviour (Vol. 2). London: Sage.
Campbell, A. (1960). Surge and decline: A study of electoral change. Public Opinion
Quarterly 24, 397–418.
Campbell, R. (2006). Gender and Voting Behaviour in Britain. Colchester: ECPR Press.
Campbell, A., Converse, P. E., Miller, W. E., and Stokes, D. E. (1960). The American
Voter. New York, NY: Wiley.
Campbell, A., G. Gurin, W. E. Miller, and D. E. Stokes (1954). The Voter Decides.
Westport, CT: Greenwood Press.
Cancela, J. and B. Geys (2016). Explaining voter turnout: A meta-analysis of national
and subnational elections. Electoral Studies 42(2), 264–275.
References 213
Caramani, Daniele (2008). Comparative Politics. Oxford: Oxford University Press.
Caramani, Daniele (2009). Introduction to Comparative Method with Boolean algebra.
Quantitative Application in the Social Sciences. Thousand Oaks, CA: Sage
Publication.
Carlin, R. E. (2006). The decline of citizen participation in electoral politics in
postauthoritarian Chile. Democratization 13, 632–651.
Carlin, Ryan E., Matthew M. Singer, and E. J. Zechmeister (2015). The Latin American
Voter: Pursuing Representation and Accountability in Challenging Context. Ann
Arbor: The University of Michigan Press.
Carmines, E. G. and Huckfeldt, R. (1996). Political behavior. In R. E. Goodin and
H.-D. Klingemann (eds), A New Handbook of Political Science. Oxford: Oxford
University Press.
Carothers, Thomas (2002). The end of the transition paradigm. Journal of Democracy
13(1), 5–21.
Carreras, M. (2018). Why no gender gap in electoral participation? A civic duty explan-
ation. Electoral Studies 52, 36–45.
Carreras, M. and N. Castañeda-Angarita (2014). Who votes in Latin America? A test
of three theoretical perspectives. Comparative Political Studies 47(8), 1079–1104.
Carter, J. R. (1984). Early projections and voter turnout in the 1980 presidential
election. Public Choice 43, 195–202.
Castillo, J. C., D. Miranda, M. Bonhomme, C. Cox, and M. Bascopé (2014). Social
inequality and changes in students’ expected political participation in Chile.
Education, Citizenship and Social Justice 9, 140–156.
Ceka, Besir (2012). The perils of political competition: Explaining participation and
trust in political parties in Eastern Europe. Comparative Political Studies 46(12),
1610–1635.
Chang, Eric C., and Yun-han Chu (2006). Corruption and trust: Exceptionalism in
Asian democracies? Journal of Politics 68(2), 259–271.
Cheeseman, Nic (2010). African elections as vehicles for change. Journal of Democracy
21(4), 139–153.
Cheibub, José Antonio, Jennifer Gandhi, and James Raymond Vreeland (2010).
Democracy and dictatorship revisited. Public Choice 143(2–1), 67–101.
Cheng, Tun-jen, and Yun-han Chu (2018). Routledge Handbook of Democratization in
East Asia. Abingdon: Routledge.
Chesnut, R. Andrew (2003). Competitive Spirits Latin America’s New Religious
Economy. New York: Oxford University Press.
Childs, Sarah (2004). A British gender gap? Gender and political participation.
Political Quarterly 75(4), 422–424.
Chin, James Ung-Ho (2002). Malaysia: The Barisan National supremacy. In John
Fuh-Sheng Hsieh and David Newman (eds), How Asia Votes. New York: Chatham
House.
Chung-li Wu and Tzu-Ping Liu (2017). Political participation in Taiwan. In Achen,
Christopher H. and T. Y. Wang (eds), The Taiwan Voter. Ann Arbor: University of
Michigan Press.
Clarke, H. D., and A. C. Acock (1989). National elections and political attitudes.
British Journal of Political Science 19, 551–562.
Cleary, Matthew R., and Susan Carol Stokes (2006). Democracy and the Culture
of Skepticism: Political Trust in Argentina and Mexico. New York: Russell Sage
Foundation.
214 References
Coffé, H. and C. Bolzendahl (2011). Gender gaps in political participation across sub-
Saharan African nations. Social Indicators Research 102, 245–264.
Coffé, Hilde, and Tanja van der Lippe (2010). Citizenship norms in Eastern Europe.
Social Indicators Research 96(3), 479–496.
Cohen. M. L. (2017). Support for electoral democracy in the Americas. In M. J.
Cohen, N. Lupu, and E. J. Zechmeister (eds), The Political Culture of Democracy
in the Americas: A Comparative Study of Democracy and Governance. The
AmericasBarometer 2016/17. A Comparative Study of Democracy and Government.
Nashville, TN: LAPOP, Vanderbilt University.
Coleman, J. (1990). The Foundations of Social Theory. Cambridge, MA: Harvard
University Press.
Coppedge, Michael, John Gerring, Carl Henrik Knutsen, Staffan I. Lindberg, Svend-
Erik Skaaning, Jan Teorell, David Altman, Michael Bernhard, Agnes Cornell, M.
Steven Fish, Haakon Gjerløw, Adam Glynn, Allen Hicken, Joshua Krusell, Anna
Lührmann, Kyle L. Marquardt, Kelly McMann, Valeriya Mechkova, Moa Olin,
Pamela Paxton, Daniel Pemstein, Brigitte Seim, Rachel Sigman, Jeffrey Staton,
Aksel Sundtröm, Eitan Tzelgov, Luca Uberti, Yi-ting Wang, Tore Wig, and Daniel
Ziblatt (2018). Democracy for All? V-Dem Annual Democracy Report.
Coppedge, Michael, John Gerring, Carl Henrik Knutsen, Staffan I. Lindberg, Jan
Teorell, David Altman, Michael Bernhard, M. Steven Fish, Adam Glynn, Allen
Hicken, Anna Luhrmann, Kyle L. Marquardt, Kelly McMann, Pamela Paxton,
Daniel Pemstein, Brigitte Seim, Rachel Sigman, Svend- Erik Skaaning, Jeffrey
Staton, Steven Wilson, Agnes Cornell, Nazifa Alizada, Lisa Gastaldi, Haakon
Gjerløw, Garry Hindle, Nina Ilchenko, Laura Maxwell, Valeriya Mechkova, Juraj
Medzihorsky, Johannes von Römer, Aksel Sundström, Eitan Tzelgov, Yi- ting
Wang, Tore Wig, and Daniel Ziblatt. (2020). V-Dem [Country–Year/Country–
Date] Dataset v10. Varieties of Democracy (V- Dem) Project. https:// doi.org/
10.23696/vdemds20.
Córdova, A. and G. Rangel (2017) Addressing the gender gap: The effect of compul-
sory voting on women’s electoral engagement. Comparative Political Studies 50(2),
264–290.
Cox, G. W., and M. C. Munger (1989). Closeness, expenditures and turnout in the
1982 US house elections. American Political Science Review 83(1), 217–230.
Cox, M. (2003). When trust matters: Explaining differences in voter turnout. Journal
of Common Market Studies 41(4), 757–770.
Craig, S. C., and M. A. Maggiotto (1982). Measuring political efficacy. Political
Methodology 8, 85–109.
Crewe, I. (1981). Electoral participation. In D. Butler, H. R. Penniman and A. Ranney
(eds), Democracy at the Polls: Comparative Study of Competitive National Elections.
Washington, DC: American Enterprise Institute.
Croke. K, G. Grossman, H. A. Larreguy, and J. Marshall (2016). Deliberate disengage-
ment: How education can decrease political participation in electoral authoritarian
regimes. American Political Science Review 110(3), 579–600.
Curtice, J. (2002). The state of election studies: Mid-life crisis or new youth? Electoral
Studies 21(2002), 161–168.
Curtice J. (2007). Comparative opinion surveys. In Russell J. Dalton and Hans-Dieter
Klingemann (eds), The Oxford Handbook of Political Behavior. Oxford: Oxford
University Press.
References 215
Curtis G. (1971). Election Campaigning Japanese Style. New York: Columbia
Univ. Press.
Cutler, Neal E., and Vern l. Bengtson (1974) Age and political alienation: Maturation,
generation and period effects. In R. J. Dalton and H. Klingemann (eds), Oxford
Handbook of Political Behavior. Oxford: Oxford University Press.
Dahl, Robert A. (1971). Polyarchy: Participation and Opposition. New Haven: New
Haven Press.
Dahlberg, S., and Maria Solevid (2016). Does corruption suppress voter turnout?,
Journal of Elections, Public Opinion and Parties 26(4), 489–510.
Dalton R. J. (1984). Cognitive mobilization and partisan dealignment in advanced
democracies. The Journal of Politics 46(1), 264–284.
Dalton, R. J. (2000). The decline of party identification. In R. J. Dalton and M. P.
Wattenberg (eds), Parties without Partisans: Political Change in Advanced Industrial
Democracies. Oxford and New York: Oxford University Press.
Dalton, Russell J., and Christopher J. Anderson (eds) (2011). Citizens, Context,
and Choice: How Context Shapes Citizens’ Electoral Choices. New York: Oxford
University Press.
Dalton, Russell J., David M. Farrell, and Ian McAllister (2011). Political Parties and
Democratic Linkage: How Parties Organize Democracy. Oxford: Oxford Scholarship
Online (published online Feb. 2015).
Dalton, R. J., and Hans-Dieter Klingemann (2009). The Oxford Handbook of Political
Behaviour. Oxford: Oxford University Press.
Dalton, R. J., D. C. Shin, and Y.- H. Chu (eds) (2008). Party Politics in East
Asia: Citizens, Elections, and Democratic Development. Colorado: Lynne Rienner
Publishers.
Dassonneville, R. (2017). Age and voting. In K. Arzheimer, J. Evans and M. S. Lewis-
Beck (eds), The SAGE Handbook of electoral behaviour (Vol. 2). London: Sage.
Dawson, H. (2014). Youth and political unrest: Waiting and envy in a South African
informal settlement. Journal of Southern African Studies 40, 861–882.
de Vries, C. (2017). Cross-national data sources: Opportunities and challenges. In K.
Arzheimer , J. Evans and M. S. Lewis-Beck (eds), The SAGE Handbook of Electoral
Behaviour (Vol. 2). London: Sage.
Denton, Ginger L. (2016). Political Participation in Asia: Typologies of Political
Behavior across Democratizing States. Singapore: World Scientific.
Denver, D., Christopher Carman, and Robert Johns (2012). Elections and Voters in
Britain. Basingstoke: Red Global Press.
Desposato, S., and B. Norrander (2009). The gender gap in Latin America: Contextual
and individual influences on gender and political participation. British Journal of
Political Science 39, 141–162.
Diamond, L. (1999). Developing Democracy: Toward Consolidation. Baltimore:
JHU Press.
Diamond, L. (2002) Consolidating democracies. In Lawrence LeDuc, Richard G.
Niemi, and Pippa Norris (eds), Comparing Democracies 2: New Challenges in the
Study of Elections and Voting. London: Sage.
Diwakar, Rekha (2008). Voter turnout in the India states: An empirical analysis.
Journal of Elections, Public Opinion and Parties 18(1), 75–100.
Domínguez, Jorge I., and Chappell Lawson (eds) (2003). Mexico’s Pivotal Democratic
Election: Candidates, Voters, and the Presidential Campaign of 2000. Stanford,
216 References
CA: Stanford University Press and the Center for U.S.-Mexican Studies, University
of California, San Diego.
Domínguez, Jorge I., and James A. McCann (1996). Democratizing Mexico: Public
Opinion and Electoral Choice. Baltimore: Johns Hopkins University Press.
Domínguez, Jorge I., and James A. McCann (1998). Mexicans react to electoral fraud
and political corruption: An assessment of public opinion and voting behavior.
Electoral Studies 17, 483–503.
Dow, Jay (1998). A spatial analysis of the 1989 Chilean presidential election. Electoral
Studies 17, 61–76.
Downs, A. (1957). An Economic Theory of Democracy. New York: Harper and Row.
Duch, Raymond M. (2007). Comparative studies of the economy and the vote. In
Carles Boix and Susan C. Stokes (eds), The Oxford Handbook of Comparative
Politics. Oxford: Oxford University Press.
Duverger, Maurice (1955). The Political Role of Women. New York: UNESCO.
Edgell, A. B., V. Mechkova, D. Altman, M. Bernhard, and S. I. Lindberg (2017). When
and where do elections matter? A global test of the democratization by elections
hypothesis, 1900–2010. Democratization 1–23.
Eiffert, B., E. Miguel, and D. Posner (2010). Political conflict and ethnic identification
in Africa. American Journal of Political Science 54(2), 494–510.
Ekström, M., and J. Östman (2013). Family talk, peer talk, and young people’s
civic orientation. European Journal of Communication 28, 294–308. doi: 10.1177/
0267323113475410.
Ellis, Andrew, Maria Gratschew, Joh H. Pammett, and Erin Thiessen et al. (2006).
Engaging the electorate: Initiative to promote voter turnout from around the
world. Stockholm: International IDEA.
Endersby, J. W., and J. T. Krieckhaus (2008). Turnout around the globe: The influ-
ence of electoral institutions on national voter participation, 1972–2000. Electoral
Studies 27, 601–610.
Evans, Geoffrey (2006). The social bases of political divisions in post-communist
Eastern Europe. Annual Review of Sociology 32, 245–270.
Evans, Geoffrey and Nan Dirk De Graaf (eds) (2013). Political Choice Matters.
Explaining the Strength of Class and Religious Cleavages in Cross- National
Perspective. Oxford: Oxford University Press.
Evans, Geoffrey, and Stephen Whitefield (1993). Identifying the bases of party compe-
tition in Eastern Europe. British Journal of Political Science 23(4), 521–548.
Evans, J. (2004). Voters and Voting –An Introduction. London: Sage Publications.
Fagan, A., and P. Kopecký (2017). The Routledge Handbook of East European Politics
(1st ed.). Milton: Taylor & Francis.
Farrell, David (2001). Electoral Systems: A Comparative Introduction. Houndmills,
Basingstoke: Palgrave Macmillan.
Ferree, K. (2006). Explaining South Africa’s racial census. Journal of Politics 68(4),
803–815.
Filer, J. E., and L. W. Kenny (1980). Voter turnout and the benefits of voting. Public
Choice 35, 575–585.
Fisher, J., D. Denver, and G. Hands (2006) Party membership and campaign activity in
Britain: the impact of electoral performance. Party Politics 12(4), 505–519.
Fisher, J., E. Fieldhouse, M. N. Franklin, R. Gibson, M. Cantijoch, and C. Wlezien
(2018). The Routledge Handbook of Elections, Voting Behavior and Public Opinion.
London: Routledge.
References 217
Fishman, R., and J. Svensson (2000). Are Corruption and Taxation Really Harmful
to Growth? Firm Level Evidence. Policy Research Working Paper Series No. 2485.
Washington, DC: World Bank.
Flanagan, Scott, Shinsaku Kohei, Ichiro Miyake, and Bradley M. Richardson (1991).
The Japanese Voter. New Haven, CT: Yale University Press.
Flanagan, Scott, and Bradley M. Richardson. (1977). Japanese Electoral
Behavior: Social Cleavages, Social Networks, and Partisanship. London: Sage.
Flores, T. E. and I. Nooruddin (2016). Elections in Hard Times: Building Stronger
Democracies in the 21st Century. Cambridge: Cambridge University Press.
Fornos, C. A., Power, T. J., and Garand, J. C. (2004). Explaining voter turnout in Latin
America, 1980 to 2000. Comparative Political Studies 37(8), 909–940.
Fox J. (1994). The difficult transition from clientelism to citizenship: Lessons from
Mexico. World Polit. 46(Jan.), 151–184.
Franklin, M. (1996). Electoral participation. In Lawrence LeDuc, Richard G. Niemi
and Pippa Norris (eds), Comparing Democracies: Elections and Voting in Global
Perspective. Thousand Oaks, California: Sage.
Franklin, M. (2002). The dynamics of electoral participation. In Lawrence LeDuc,
Richard G. Niemi and Pippa Norris (eds), Comparing Democracies 2: New
Challenges in the Study of Elections and Voting (2nd ed.). London; Thousand
Oaks: Sage.
Franklin, Mark N. (2004). Voter Turnout and the Dynamics of Electoral Competition
in Established Democracies since 1945. Cambridge, UK; New York: Cambridge
University Press.
Franklin, M., and W. P. Hirczy de Miño (1998). Separated powers, divided govern-
ment, and turnout in U.S. presidential elections. American Journal of Political
Science 42(1), 316–326.
Fukuyama, Francis (1989). The end of history? The National Interest 16, 3–18.
Fukuyama, Francis (2001). Asian values in the wake of the Asian crisis. In Farrukh
Iqbal and John-Il You (eds), Democracy, Market Economics and Development.
Washington, DC: World Bank.
Furnham, A., and B. Gunter (1989). The Anatomy of Adolescence. Young People’s
Social Attitudes in Britain. London: Routledge.
Gabriel, O. W. and J. W. van Deth (1995) Political interest. In J. W. van Deth and E.
Scarbrough (eds), The Impact of Values. Oxford: Oxford University Press.
Gallagher, M. (1991). Proportionality, disproportionality and electoral systems.
Electoral Studies 10, 33–51.
Gallagher, M. and Paul Mitchell (2005). The Politics of Electoral Systems.
Oxford: Oxford University Press.
Gallego, A. (2009). Where else does turnout decline come from? Education, age, gener-
ation and period effects in three European countries. Scandinavian Political Studies
32(1), 23–44.
Gallego, A. (2010). Understanding unequal turnout: Education and voting in com-
parative perspective. Electoral Studies 29, 239–248.
Gallego, A. (2015). Unequal Political Participation Worldwide. Cambridge: Cambridge
University Press.
Gallego, A., Rico, G., and Anduiza, E. (2012). Disproportionality and voter turnout
in new and old democracies. Electoral Studies 31(1), 159–169.
Gastil, R. D. (1979). Freedom in the World: Political Rights and Civil Liberties.
New York: Freedom House.
218 References
Geddes, Barbara (1990). How the cases you choose affect the answers you get: Selection
bias in comparative politics. Political Analysis 2(1), 131–150.
Geertz, Clifford (1963). Old Societies and New States. New York: Free Press.
Gélineau, François (2013). Electoral accountability in the developing world. Electoral
Studies 32(3), 418–424.
Gélineau, François, and Matthew Singer (2015). The economy and incum-
bent support in Latin America. In Ryan E. Carlin, Matthew M. Singer, and
Elizabeth J. Zechmeister (eds), The Latin American Voter. Ann Arbor: University
Michigan Press.
Gerber, A. S., and D. P. Green (2000a). The effects of canvassing, telephone calls, and
direct mail on voter turnout: A field experiment. American Political Science Review
94(3), 653–663.
Gerber, A. S., and D. P. Green (2000b). The effect of a nonpartisan get-out-the-vote
drive: An experimental study of leafleting. Journal of Politics 62(3), 846–857.
Gerring, John (2007). Case Study Research: Principles and Practices. Cambridge:
Cambridge University Press.
Geys, B. (2006). Voter turnout: A review of aggregate-level research. Electoral Studies
25, 637–663.
Gordon, S. B. and Segura, G. M. (1997). Cross-National Variations in the Political
Sophistication of Individuals: Capability or Choice? Journal of Politics 59,
126–147.
Gottlieb, Jessica, Robert Blair, Hannah Baron, Aries Arugay, Cameron Ballard-Rosa,
Don Davidson, Laura Gamboa, Guy Grossman, Shelby Grossman, Christina
Kulich- Vamvakas, Nancy Lapp, Jennifer McCoy, Amanda Robinson, Steven
Rosenzweig, Eric Royer, Cathy Lisa Schneider, Susan Stokes, and Megan Turnbull.
2019. Democratic Erosion Event Dataset v3. Democratic Erosion: A Cross-
University Collaboration.
GRAY, M., & CAUL, M. (2000). Declining Voter Turnout in Advanced Industrial
Democracies, 1950 to 1997: The Effects of Declining Group Mobilization.
Comparative Political Studies, 33(9), 1091–1122.
Graziano L. (1973). Patron-client relationships in Southern Italy. Eur. J. Polit. Res.
1(1), 3–34.
Guardado, Jenny R., and Leonard Wantchekon (2013). Do electoral handouts affect
voting behavior? Paper presented at the annual meeting of the Midwest Political
Science Association, Chicago, IL, April.
Gupta, Sanjeev, Hamid Davoodi, and Rosa Alonso-Teme (2002). Does corruption
affect income inequality and poverty? Economics of Governance 3(1), 23–45.
Guy, Peters B. (2005). Institutional Theory in Political Science: The “New
Institutionalism”. Gosport: Ashford Colour Press Ltd.
Hall, P. A. and R. C. R. Taylor (1996). Political science and the three new
institutionalisms. Political Studies 44(5), 936–957.
Hamilton, L. C. (2002). Statistics with Stata. College Station, TX: Stata Corporation.
Harder J. and J. A. Krosnick (2008). Why do people vote? A psychological analysis of
the causes of voter turnout. Journal of Social Issues 64(3), 525–549.
Harkness, J. A., F. J. R. Van de Vijver, and P. P. Mohler (2003). Cross-Cultural Survey
Methodology. New York: Wiley.
Harris, F. C. (1994). Something within: Religion as a mobilizer of African-American
political activism. Journal of Politics 56(1), February, 42–68.
Healey, Joseph F. (2012). Statistics: A Tool for Social Research (9th ed.), International
edition. Belmont, CA: Wadsworth Cengage Learning.
References 219
Heath, A., S. Fisher and S. Smith (2005). The globalization of public opinion research.
Annual Review of Political Science 8, 297–333.
Heath, A. F., R. Jowell, J. Curtice, G. A. Evans, J. Field, and S. Witherspoon
(1991). Understanding Political Change: The British Voter 1964– 1987. Oxford:
Pergamon Press.
Helliwell, J. F. (1994). Empirical linkage between democracy and economic growth.
British Journal of Political Science 24, 225–248.
Hellwig, Timothy (2010). Elections and the economy. In Lawrence LeDuc, Richard G.
Niemi, and Pippa Norris (eds), Comparing Democracies 3: Elections and Voting in
the 21st Century (3rd ed.). Thousand Oaks, CA: Sage.
Herron, Erik S. (2009). Elections and Democracy After Communism? Basingstoke:
Palgrave Macmillan.
Hicken, Allen (2011). Clientelism. Annual Review of Political Science 14(1), 289–310.
Hillygus, D. Sunshine (2005). The missing link: Exploring the relationship between
higher education and political engagement. Political Behavior 27(1), 25–47.
Hoffmann-Martinot, V. (1994). Voter turnout in French municipal elections. In L.
Lopez-Nieto (ed.), Local elections in Europe (13–42). Barcelona, Spain: Institut de
ciènces politiques I socials.
Holbein, John B. and D. Sunshine Hillygus (2020). Making Young Voters: Converting
Civic Attitudes into Civic Action. New York: Cambridge University Press.
Honwana, Alcinda (2012). The Time of Youth: Work, Social Change, and Politics in
Africa. Sterling, VA: Kumarian Press.
Hooghe, M., and B. Wilkenfeld (2008). The stability of political attitudes and
behaviors across adolescence and early adulthood. A comparison of survey data on
adolescents and young adults in eight countries. Journal of Youth and Adolescence
37, 155–167.
Hooghe, M., and E. Quintelier (2014). Political participation in European coun-
tries: The effect of authoritarian rule, corruption, lack of good governance and
economic downturn. Comparative European Politics 12, 209–232.
Horn, J. L., and J. J. McArdle (1992). A practical and theoretical guide to meas-
urement invariance in aging research. Experimental Aging Research 18(3),
117–144.
Horowitz, D. (1985). Ethnic Groups in Conflict. Berkeley: University of California
Press.
Horowitz, D. (1991). A Democratic South Africa? Constitutional Engineering in a
Divided Society. Berkeley: University of California Press.
Hsieh, John Fuh-Sheng, and David Newman (2002). How Asia Votes. New York:
Chatham House.
Hsieh, John Fuh- sheng, and David Newman (2002). Continuity and change in
Taiwan’s electoral politics. In J. Hsieh and D. Newman (eds), How Asia Votes.
New York: Seven Bridges.
Huntington, S. (1991). The Third Wave: Democratization in the Late Twentieth Century.
Norman, OK: University of Oklahoma Press.
Hyug Baeg, Im. (2004). Faltering democratic consolidation in South Korea: Democracy
at the end of the “Three Kims” era. Democratization 11(5), 187.
Inglehart, R. and P. Norris (2000). The developmental theory of the gender
gap: Women’s and men’s voting behavior in global perspective. International
Political Science Review 21(4), 441–463.
Inglehart, R., and P. Norris (2003). Rising Tide –Gender Equality and Cultural Change
around the World. New York: Cambridge University Press.
220 References
Inman, Kris, and Josephine T. Andrews (2009). Corruption and political participation
in Africa. Evidence from survey and experimental research. Paper presented at the
Midwest Political Science Association, April 3–6, Chicago.
International Institute for Democracy and Electoral Assistance (2002). Voter Turnout
since 1945: A Global Report (2nd edition). Stockholm: International IDEA.
Isaksson, Ann-Sofie (2010). Political participation in Africa: Participatory inequal-
ities and the role of resources. Working papers in Economics, No. (462),
Department of Economics School of Business, Economics and Law at University
of Gothenburg.
Isaksson, Ann-Sofie (2014). Political participation in Africa: The role of individual
resources. Electoral Studies 32, 244–260.
Ishiyama, J. (1996). Electoral systems experimentation in the new Eastern Europe: The
single transferable vote and the additional member system in Estonia and Hungary.
East European Quarterly 29(4), 487–507.
Ishiyama, J. and Fox, C. (2006). What affects the strength of partisan identity in sub-
Saharan Africa. Politics and Policy 34(4), 748–773.
Jackman, R. W. (1987). Political Institutions and Voters Turnout in the Industrial
Democracies, American Political Science Review, 81(2), 405–424.
Jackman, R. W. and Miller, Ross M. A. (1995). Voter turnout in the industrial democ-
racies during the 1980s. Comparative Political Studies 27, 467–492.
Jankowski, Thomas B. and John M. Strate (1995). Modes of participation over the
adult life span. Political Behavior 17(1), 89–106.
Jasiewicz, K. (2007). Citizens and politics. In S. White, J. Batt, and P. G. Lewis (eds),
Developments in Central and East European Politics 4. Basingstoke: Palgrave
Macmillan.
Jennings, M. Kent (1979). Another look at the life cycle and political participation.
American Journal of Political Science 23(4), 755–771.
Johnson, Gregg, and Sooh-Rhee Ryu (2010). Repudiating or rewarding neoliberalism?
How broken campaign promises condition economic voting in Latin America.
Latin American Politics and Society 52, 1–24.
Johnston, R., D. Cutts, C. Pattie, and J. Fisher (2012) We’ve got them on the
list: Contacting, canvassing and voting in a British general election campaign.
Electoral Studies, 31(2), 317–329.
Jones, M. P. (2009). Gender quotas, electoral laws, and the election of women: Evidence
from the Latin American vanguard. Comparative Political Studies 42, 56–81.
Jones-Correa, M. A., and D. L. Leal (2001). Political participation: Does religion
matter? Political Research Quarterly 54(4), December, 751–770.
Juma, Calestous (2011). Why Africa needs to lower its voting age to 16. The Guardian,
9 February 2011.
Karahan, G. R., Coats, M., and Shugart, W. F. II (2006). Corrupt political jurisdictions
and voter participation. Public Choice 126(1), 87–196.
Karp, J. A., and S. A. Banducci (2008). Political efficacy and participation in twenty-
seven democracies. British Journal of Political Science 38, 311–334.
Karp, J. A., S. A. Banducci, and S. Bowler (2007). Getting out the vote: Party
mobilization in a comparative perspective. British Journal of Political Science 38,
91–112.
Karp, J. A., and D. Brockington, (2005). Social desirability and response validity: A
comparative analysis of overreporting voter turnout in five countries. Journal of
Politics 67, 825–840.
References 221
Karp, Jeffrey A., and Caitlin Milazzo (2015). Democratic scepticism and political
participation in Europe. Journal of Elections, Public Opinion and Parties 25(1),
97–110.
Kasara, Kimuli, and Pavithra Suryanarayan (2015). When do the rich vote less than
the poor and why? Explaining turnout inequality across the world. American
Journal of Political Science 59(3), 613–627.
Katz, R. (1997). Democracy and Elections. Oxford and New York: Oxford
University Press.
Keefer, P. 2007. Clientelism, credibility, and the policy choices of young democracies.
American Journal of Political Science 51(4), 804–821.
Kim, Seok-Eun (2005) The role of trust in the modern administrative state: An inte-
grative model. Administration and Society 37(5), 611–635.
Kinder, D. R. (2006). Politics and the life cycle. Science 312, 1905–1908.
King, A. (1969). Political parties in Western democracies. Polity 2, 112–141.
Kitschelt, H., K. A. Hawkins, J. P. Luna, G. Rosas, E. J. Zechmeister (2010). Latin
American Party Systems. Cambridge, UK: Cambridge University Press.
Kittilson, M. C. (2007). Research resources in comparative political behavior. In J.
R. Dalton and H. Klingemann (eds), Oxford Handbook of Political Behavior.
Oxford: Oxford University Press.
Kittilson, M. C., and C. J. Anderson (2011). Electoral supply and voter turnout.
In Russell J. Dalton and Christopher J. Anderson (eds), Citizens, Context,
and Choice: How Context Shapes Citizens’ Electoral Choices. Oxford: Oxford
University Press.
Kittilson, M. C., L. Schwindt-Bayer (2012). The gendered effects of electoral institutions.
Oxford: Oxford University Press.
Klesner, J. L. (2007). Social capital and political participation in Latin America:
Evidence from Argentina, Chile, Mexico, and Peru. Latin American Research
Review 42, 1–32.
Klingemann, H.- D. (2007). Turnout in European and national parliamentary
elections. In C. Lequesne, and MacDonagh-Pajerova, M. (eds), La Citoyennete ´
De ´mocratique Dans l’Europe Des Vingt-Sept. Paris: L’Harmattan.
Konzelmann, L., Wagner, C., and Rattinger, H. (2012). Turnout in Germany in the
course of time: Life Cycle and cohort effects on electoral turnout from 1953 to
2049. Electoral Studies 31(2), 250–261.
Kostadinova, Tatiana. (2003). Voter turnout dynamics in postcommunist Europe.
European Journal of Political Research 42(6), 741–759.
Kostadinova, Tatiana (2009). Abstain or rebel: Corruption perceptions and voting in
East European elections. Politics & Policy 37(4), 691–714.
Kostadinova, T. and Power, T. J. (2007). Does democratization depress participation?
Voter turnout in the Latin American and Eastern European transitional democra-
cies. Political Research Quarterly 60(3), 363–377.
Kostelka, F. (2014). The state of political participation in post-communist democ-
racies: Low but surprisingly little biased citizen engagement. Europe-Asia Studies
66(6), 945–968.
Kostelka, F. (2015). To mobilise and demobilise: The puzzling decline of voter turnout
in post-communist democracies. PhD dissertation, Centre d’études européennes,
Sciences Po, Paris.
Kostelka, F. (2017). Does democratic consolidation lead to a decline in voter turnout?
Global evidence since 1939. American Political Science Review 111(4), 653–667.
222 References
Kostelka, Filip, Blais A., and E. Gidengil (2019). Has the gender gap in voter turnout
really disappeared? West European Politics 42, 437–463.
Kramon, E. (2017). Money for Votes: The Causes and Consequences of Electoral
Clientelism in Africa. Cambridge: Cambridge University Press.
Krishna, A. (2002). Enhancing political participation in democracies: What is the role
of social capital? Comparative Political Studies 35(4), 437–460.
Krishna, A. (2008). Do poor people care less for democracy? Testing individual-level
assumptions with individual-level data. In A. Krishna (ed.), Poverty, Participation
and Democracy: A Global Perspective. New York: Cambridge University Press.
Kuenzi, M., and G. M. S. Lambright (2007). Voter turnout in Africa’s multiparty
regimes. Comparative Political Studies 40(6), 665–690.
Kuenzi, M., and G. M. S Lambright, (2011). Who votes in Africa? An examination of
electoral participation in 10 African countries. Party Politics 17, 767–799.
La Due Lake, Ronald, and Robert Huckfeldt (1998). Social capital, social networks,
and political participation. Political Psychology 19(3), 567–584.
Ladner, A. and H. Milner (1999). Do voters turn out more under proportional than
majoritarian systems? The evidence from Swiss communal elections. Electoral
Studies 18, 235–250.
Lane, Roberts E. (1959). Political Life: Why People Get Involved in Politics. Glencoe,
Illinois: The Free Press.
Langston, Joy (2007). Strong parties in a struggling party system: Mexico in the
democratic era. In Paul Webb, and Stephen White (eds), Political Parties in New
Democracy. Oxford: Oxford University Press.
Lanning, K. (2008). Democracy, voting, and disenfranchisement in the United
States: A social psychological perspective. Journal of Social Issues 64, 431–446.
LaPiere, R. T. (1934). Attitudes vs. actions. Social Forces 13, 230–237.
Lavezzolo, Sebastián (2008). Adversidad económica y participación electoral en
América Latina, 1980–2000. Revista Española de Ciencia Política 18, 67–93.
Lazarsfeld, Paul F., Bernard Berelson, and Hazel Gaudet. 1948. The People’s
Choice: How the Voter Makes Up His Mind in a Presidential Campaign (2nd ed.).
New York: Columbia University Press.
LeDue, L., Niemi Richard, G., and Norris Pippa (1996). Comparing Democracies:
Elections and Voting in Global Perspective. Thousand Oaks, CA: Sage.
Lee, A.-R., and Glasure, Y. U. (2007). Social Capital and Political Participation in
South Korea. Asian Affairs: An American Review 34, 101–118.
Lehoucq, F., and D. Wall (2004). Explaining voter turnout rates in new democra-
cies: Guatemala. Electoral Studies 23, 485–500.
Leighley, J. E. and J. Nagler (1992). Individual and systemic influences on turnout: Who
votes? Journal of Politics 54(3), 718–740.
Leighley, Jan E., and J. Nagler (2007). Unions, voter turnout, and class bias in the US
electorate, 1964–2004. Journal of Politics 69(2), 430–441.
Leighley, J. E. and J. Nagler (2014). Who Votes Now? Demograhpic, Issues, Inequality
and Turnout in the United States. Princeton, NJ: Princeton University Press.
Letki, N. (2003). Explaining political participation in East Central Europe: Social cap-
ital, democracy and the communist past. Nuffield College Politics Working Paper
2003-W2, Oxford: University of Oxford.
Letsa, N. (2019). Expressive voting in autocracies: A theory of non-economic partici-
pation with evidence from Cameroon. Perspectives on Politics 18(2), 439–453.
References 223
Lever, H. (1979). Ethnicity and voting patterns in South Africa. Political Studies 27(3),
458–468.
Levitsky, Steven and Lucan Way (2002). The rise of competitive authoritarianism.
Journal of Democracy 13(2), 51–65.
Levitsky, S. and L. A. Way (2010). Competitive Authoritarianism: Hybrid Regimes after
the Cold War. Cambridge: Cambridge University Press.
Lewis-Beck, William G. Jacoby, Helmut Norpoth, and Herbert F. Weisberg (2008).
The American Voter Revisited. Ann Arbor: University of Michigan, Press.
Lewis-Beck, M., and M. Stegmaier (2000). Economic determinants of election
outcomes. Annual Review of Political Science 3, 183–219.
Lewis-Beck, Michael. S., and Mary Stegmaier (2007). Economic models of voting.
In Russell Dalton and Hans-Dieter Klingemann (eds), The Oxford Handbook of
Political Behavior. Oxford: Oxford University Press.
Lewis-Beck, Michael S., and Mary Stegmaier (2008). The economic vote in transi-
tional democracies. Journal of Elections, Public Opinion and Parties 18(3), 303–323.
Lindberg, S. (2006a). Democracy and Elections in Africa. Baltimore, MD: Johns
Hopkins Univ. Press.
Lindberg, S. (2006b). The surprising significance of African elections. Journal of
Democracy 17(1), 139–151.
Lindberg, S. I. (2009). Democratization by Election: A New Mode of Transition.
Baltimore, MD: Johns Hopkins Univ. Press.
Lindberg, Staffan I. (2010). What accountability pressures do MPs in Africa face and
how do they respond? Evidence from Ghana. Journal of Modern African Studies
48(1), 117–142.
Linz, J. J., and Stepan, A. (1996). Problems of Democratic Transition and Consolidation.
Baltimore: Johns Hopkins University Press.
Lührmann, Anna and Staffan I. Lindberg (2019). A third wave of autocratization is
here: What is new about it?, Democratization 26(7), 1095–1113.
Lust, Ellen and David Waldner (2015). Unwelcome Change: Understanding,
Evaluating, and Extending Theories of Democratic Backsliding. Washington, DC:
USAID.
Lyne, M. 2007. Rethinking economics and institutions: the voter’s dilemma and demo-
cratic accountability. See Kitschelt and Wilkinson 2007c, 206–226.
Macpherson, C. B. (1977). The Life and Time of Liberal Democracy. Oxford: Oxford.
Mainwaring, Scott (1999). Rethinking Party Systems in the Third Wave of
Democratization: The Case of Brazil. Stanford, CA: Stanford University Press.
Mainwaring, Scott, and Mariano Torcal (2006). Party system institutionalization and
party system theory after the third wave of democratization. In Richard S. Katz
and William Crotty (eds), Handbook of Political Parties. London: Sage.
Michael, Marsh, Richard Sinnott, John Garry, and Fiachra Kennedy (2008).
The Irish Voter: The Nature of Electoral Competition in the Republic of Ireland.
Manchester: Manchester University Press.
Mattes. R., and D. Mughogho (2010). The Limited Impacts of Formal Education
on Democratic Citizenship in Africa. Higher Education Research and Advocacy
Network in Africa Paper. Cape Town: Centre for Higher Education Transformation.
Mauro, P. (1997). The effects of corruption on growth, investment and government
expenditure: A crosscountry analysis. In K. A. Elliott (ed.), Corruption and the
Global Economy. Washington, DC: Institute for International Economics.
224 References
Maydom, B. (2017). Migrant Remittances and Democracy. DPhil diss., University of
Oxford.
McAllister, Daniel J. (1995). Affect and cognition-based trust as foundations for
interpersonal cooperation in organizations. Academy of Management Journal
38(1), 24–59.
McCann, J. A., and Domínguez, J. I. (1998). Mexicans react to political fraud and
corruption: An assessment of public opinion and voting behavior. Electoral Studies
17, 483–503.
McLaughlin, E. (2008). Beyond the racial census: The political salience of ethno-
linguistic cleavages in South Africa. Comparative Political Studies 40(4),
435–456.
Melo, D. F., and D. Stockemer (2014). Age and political in Germany, France and the
UK: A comparative analysis. Comparative European Politics 12(1), 33–53.
Mill, J. S. (1848/1875). A System of Logic: Ratiocinative and Inductive: Being a Connected
View of the Principles of Evidence and the Method of Scientific Investigation, Vol 1
(9th ed.). London: Longmans, Green Reader, and Dyer.
Millard, Frances (1999). Polish Politics and Society. London; New York: Routledge.
Miller, A. H., P. Gurin, G. Gurin, and O. Malanchuk (1981). Group conscious-
ness and political participation. American Journal of Political Science 25(3),
494–511.
Miller, A. H. and O. Listhaug (1990). Political parties and confidence in government.
British Journal of Political Science 20, 357–386.
Miller, W. E. (1994). An organizational history of the intellectual origins of the
American national election studies. In J. J. A. Thomassen (ed.), The Intellectual
History of Election Studies. Special Issue of the European Journal for Political
Research. Dordrecht: Kluwer.
Miller, W. E., and J. Merrill Shanks (1996). The New American Voter. Cambridge:
Harvard University Press.
Mo, Jongryn, David Brady, and Jaehun Ro. (1991). Urbanization and voter turnout in
Korea: An update. Political Behavior 13, 21–31.
Monyake Moletsane (2018). Can the poor fight back? A multilevel analysis of
corruption, poverty and collective demand for accountability in Africa, Journal of
Contemporary African Studies 36(4), 514–531.
Montinola, Gabriella R. (1999). The Philippines in 1998: Opportunities amid crisis.
Asian Survey 39(1), 67.
Monyake, Moletsane, and Dan Hough (2019). Citizens, bribery and the propensity to
protest. Commonwealth & Comparative Politics 57(3), 282–302.
Moreno, Alejandro (2003). El Votante Mexicano: Democracia, Actitudes Políticas y
Conducta Electoral. Mexico City: Fondo de Cultura Económica.
Moreno, Alejandro. (2009). La Decisión Electoral: Votantes, Partidos, y Democracia en
México. Mexico City: Miguel Ángel Porrúa.
Morgenbesser, L., and Pepinsky, T. B. (2019). Elections as causes of democratiza-
tion: Southeast Asia in comparative perspective. Comparative Political Studies
52(1), 3–35.
Nadeau, Richard, Éric Bélanger, and Thomas Didier (2013). The Chávez vote and the
national economy in Venezuela. Electoral Studies Sept. 32(3), 482–488.
Nadeau, Richard, Éric Bélanger, Michael S. Lewis- Beck, Mathieu Turgeon and
François Gélineau (2017). Latin American Elections: Choice and Change. Ann
Arbor: The University of Michigan Press.
References 225
Nathan, Noah (2019). Does participation reinforce patronage? Policy preferences,
turnout, and class in urban Ghana. British Journal of Political Science 49(1),
229–255.
Neundorf, A., J. Gerschewski, and R.-G. Olar (2019). How do inclusionary and
exclusionary autocracies affect ordinary people?. Comparative Political Studies.
doi: 10.1177/0010414019858958.
Newman, David (2002). Hong Kong: A decade of change. In John Fuh-Sheng Hsieh
and David Newman (eds), How Asia Votes. New York: Chatham House,
Nichter, S. (2008). Vote buying or turnout buying? Machine politics and the secret
ballot. American Political Science Review 102(1), 19–31.
Nie, N. H., J. Junn, and K. Stehlik-Barry (1996). Education and Democratic Citizenship
in America. Chicago: University of Chicago Press.
Niemi, Richard G., Stephen C. Craig, and Franco Mattei (1991). Measuring internal
political efficacy in the 1988 national election study. American Political Science
Review 85, 1407–1413.
Nohlen, Dieter (2005). Elections in the Americas: A Data Handbook Volume 2: South
America. Oxford: Oxford University Press.
Norris, P. (2002). Democratic Phoenix: Reinventing Political Activism. New York:
Cambridge University Press.
Norris, P. (2003). Young people & political activism: From the politics of loyalty to the
politics of choice? Retrieved from http://ksghome.harvard.edu/~pnorris/Acrobat/
COE.pdf.
Norris, P. (2004, October). Young people and political activism: From the pol-
itics of loyalties to the politics of choice? Paper presented at the Conference for
Civic Engagement in the 21st century: Toward a Scholarly and Practical Agenda,
Columbia, South Carolina.
Norris, P. (2009). The globalization of comparative public opinion research. In T.
Landman and N. Robinson (eds), The Sage Handbook of Comparative Politics,
London: Sage.
Norris, Pippa (2002). Democratic Phoenix: Reinventing Political Activism. New York:
Cambridge University Press.
Norris, Pippa (2002). Social capital and the news media. Harvard International Journal
of Press/Politics 7 (2), 3–8.
Norris, Pippa (2004). Electoral Engineering: Voting Rules and Political Behavior.
New York: Cambridge University Press.
North, D. C. (1990). Institutions, Institutional Change, and Economic Performance:
Political Economy of Institutions and Decisions. Cambridge: Cambridge University
Press.
O’Donnell, G., and P. C. Schmitter (1986). Transitions from Authoritarian Rule:
Tentative Conclusions about Uncertain Democracies. Baltimore: Johns Hopkins
University Press.
O’Neill, Brenda (2003). Examining declining electoral turnout among Canada’s youth.
Electoral Insight 5, 2.
Ochoa, Enrique C. (1987). The rapid expansion of voter participation in Latin
America: Presidential elections, 1845–1986. In James W. Wilkie and David Lorey
(eds), Statistical Abstract of Latin America. Los Angeles: UCLA Latin American
Center Publications, University of California.
Offe, Claus (1991). Capitalism by democratic design? Democratic theory facing the
triple transition in East Central Europe. Social Research 58(4), 865–892.
226 References
Olorunsola, V. (1972). The Politics of Cultural Sub-nationalism in Africa. Garden City,
NJ: Doubleday.
Orvista, Marta, Caplanova Anetta, and Hudson John (2005). Intended electoral par-
ticipation in transitional countries. Economics of Governance 6(3), 211–228.
Pacek, Alexander (1994). Macroeconomic conditions and electoral politics in East
Central Europe. American Journal of Political Science 38(3), 723–744.
Pacek, Alexander C., Grigore Pop-Eleches, and Joshua A. Tucker (2009). Disenchanted
or discerning: Voter turnout in post-communist countries. The Journal of Politics
71(2), 473–491.
Park, C.-M. (2016). Political trust in the Asia-Pacific region. In S. Zmerli and T. W. G.
van der Meer (eds), Handbook on Political Trust. Cheltenham, UK: Edward Elgar
Publishing.
Park, Chan Wook. (2002). Elections in democratizing Korea. In John Fuh-Sheng
Hsieh and David Newman (eds), How Asia Votes. New York: Chatham House.
Pattie, C., and R. Johnston (2001). Losing the voters’ trust. British Journal of Politics
and International Relations 3, 191–222.
Perea, E. A. (2002) Individual characteristics, institutional incentives and electoral
abstention in Western Europe. European Journal of Political Research 41, 643–673.
Pérez-Liñán, A. (2001). Neoinsitutional accounts of voter turnout: Moving beyond
industrial democracies. Electoral Studies 20, 281–297.
Persson, Mikael (2015). Education and political participation. British Journal of
Political Science 45, 689–703.
Persson, M., Wass, H., and Oscarsson, H. (2013). The generational effect in turnout
in the Swedish general elections, 1960–2010. Scandinavian Political Studies 36(3),
249–269.
Persson, M., and M. Solevid (2014). Measuring political participation –Testing social
desirability bias in a web-survey experiment. International Journal of Public Opinion
Research 26, 98–112.
Persson, T., G. Tabellini, F. Trebbi (2003). Electoral rules and corruption. J. Eur. Econ.
Assoc. 1(4), 958–989.
Peter, Guy (2018). Institutions and voting behavior. In J. Fisher et al. (ed.), The
Routledge Handbook of Elections, Voting Behaviour and Public Opinion. Abingdon:
Routledge.
Pierce, Roy (1995). Choosing the Chief: Presidential Elections in France and the United
States. Ann Arbor: University of Michigan Press.
Pietsch, J., Michael Miller and Jeffrey A. Karp (2015). Public support for democ-
racy in transitional regimes. Journal of Elections, Public Opinion and Parties
25(1), 1–9.
Plutzer, E. (2002). Becoming a habitual voter: Inertia, resources, and growth in young
adulthood. American Political Science Review 96(1), 41–56.
Plutzer, E. (2018). Demographics and social bases of voter turnout. In J. Fisher,
E. Fieldhouse, M. N. Franklin, R. Gibson, M. Cantijoch, and C. Wlezien (eds),
The Routledge Handbook of Elections, Voting Behaviour and Public Opinion.
Abingdon: Routledge.
Podobnik, B., J. Shao, D. Njavro, P. C. Ivanov and H. E. Stanley (2008). Influence
of corruption on economic growth rate and foreign investment. The European
Physical Journal 63(3), 547–550.
Poole. K T., and Zeigler, L. H (1985). Women, Public Opinion, and Politics: The
Changing Political Attitudes of American Women. New York: Longman.
References 227
Pop-Eleches, Grigore (2010). Throwing out the bums: Protest voting and unorthodox
parties after communism. World Politics 62(2), 221–260.
Posner, D. (2005). Institutions and Ethnic Politics in Africa. New York: Cambridge
University Press.
Potgieter, E. (2013). Predictors of political participation in new democracies: A com-
parative study. MA Thesis, Stellenbosch University.
Powell, G. B. (1980). Voting turnout in thirty democracies. In. R. Rose (ed.), Electoral
Participation. Beverly Hills: Sage.
Powell, G. B. (1982). Contemporary Democracies: Participation, Stability and Violence.
Cambridge: Harvard University Press.
Przeworski, A. (1991). Democracy and the Market. Cambridge: Cambridge University
Press.
Przeworski, A., M. Alvarez, J. Cheibub, and F. Limongi (2000). Democracy and
Development: Political Institutions and Well- Being in the World. New York:
Cambridge University Press.
Putnam, R. D (1993). Making Democracy Work: Civic Traditions in Modern Italy.
Princeton, NJ: Princeton University Press.
Putnam, R. D. (1995). Bowling alone: America’s declining social capital. Journal of
Democracy 6(1), 65–78.
Putnam, R. D. (2000). Bowling Alone: The Collapse and Revival of American
Community. New York: Simon and Schuster.
Pye, Lucien (1966). Aspects of Political Development. Boston: Little Brown.
Ragin, C. (1987). The Comparative Method: Moving Beyond Qualitative and
Quantitative Strategies. Berkeley, CA; London: University of California Press.
Ragsdale, L. and J. G. Rusk (1993). Who are non-voters? American Journal of Political
Science 37, 721–746.
Reilly, Benjamin (2006). Democracy and Diversity: Political Engineering in the Asia-
Pacific. Oxford: Oxford University Press.
Remmer, K. L. (2010) Political scale and electoral turnout: Evidence from the less
industrialized world. Comparative Political Studies 43(3), 275–303.
Resnick, Danielle, and Daniela Casale (2014). Young populations in young democra-
cies: Generational voting behaviour in Sub-Saharan Africa. Democratization 21(6),
1172–1194.
Riker, William H. and Ordeshook, Peter C. (1968). A theory of the calculus of voting.
American Political Science Review 62, 25–42.
Roberts, Andrew (2018). Democracy and democratization in post-communist Europe.
In A. Fagan and P. Kopecký (eds), Routledge Handbook of East European Politics,
1st ed. London: Routledge.
Roberts, Nancy C. (2008). The Age of Direct Citizen Participation. New York: M.
E. Sharpe.
Rood, Steven (2002). Election as complicated and important events in the
Philippines. In John Fuh-Sheng Hsieh and David Newman (eds), How Asia Votes.
New York: Chatham House,
Rose, R. (1995). Mobilizing demobilized voters in post-communist societies. Party
Politics 1, 4.
Rose, R., and W. Mishler (1998). Negative and positive party identification in post-
communist countries. Electoral Studies 17(2), 217–234.
Rose, R. and N. Munro (2003). Elections and Parties in New European Democracies.
Washington, DC: CQ Press.
228 References
Rose, Richard, William Mishler, and Christian Haerpfer (1998). Democracy and Its
Alternatives: Understanding Post-communist Societies. Baltimore, MD: John
Hopkins University Press.
Rose, R. and C. Peiffer (2019). Bad Governance and Corruption: Political Corruption
and Governance. London: Palgrave Macmillan.
Rosenbaum, W. A (1975). Political Culture. New York: Praeger.
Rosenstone, Steven J., and John Mark Hansen (1993). Mobilization, Participation and
Democracy in America. New York: Macmillan.
Sanches, Edalina Rodrigues, José Jaime Macuane, and Chipo Dendere (2020).
Introduction. Cadernos de Estudos Africanos 38, 9–14.
Schedler, Andreas (1998). What is democratic consolidation? Journal of Democracy
9(2), 91–107.
Schlozman. K.L, N. Burns, and Sidney Verba (1994). Gender and the pathways to par-
ticipation: The role of resources. The Journal of Politics 56(4), 963–990.
Schmitt, Hermann and Renato Mannheimer (1991). About voting and non-voting in
the European elections in 1989. European Journal of Political Research 19, 31–54.
Schraufnagel, Scot, and Barbara Sgouraki (2005). Voter turnout in Central and South
America. The Latin Americanist 49, 39–69.
Schumpeter J. A. (1942). Capitalism, Socialism and Democracy. London and
New York: Routledge.
Schwindt-Bayer, L. (2009). Making quotas work: The effect of gender quota laws on
the election of women. Legislative Studies Quarterly 34, 5–28.
Scott, Ian (1996). Party politics and elections in transitional Hong Kong. Asian Journal
of Political Science 4(1), 130–152.
Segura, G., and S. Bowler (2012). The Future Is Ours: Minority Politics, Political
Behavior, and the Multiracial Era of American Politics. Thousand Oaks: Sage.
Seligson, M. A. (2002). Trouble in paradise? The erosion of system support in Costa
Rica, 1978–1999. Latin American Research Review 37, 160–185.
Seligson, Mitchell A., Annabelle Conroy, Ricardo Cordova Macias, Orlando J. Perez,
and Andrew J. Stein (1995). Who votes in Central America? A comparative ana-
lysis. In Mitchell A. Seligson and John A. Booth (eds), Elections and Democracy
in Central America, Revisited. Chapel Hill, NC: The University of North
Carolina Press.
Shachar, R., and B. Nalebuff (1999). Follow the leader: Theory and evidence on polit-
ical participation. American Economic Review 89(3), 525–547.
Sheerin, C. (2008). Political Efficacy and Youth Non- Voting. Saarbrucken: VDM
Verlag Dr. Muller.
Shi, T. (2014). The Cultural Logic of Politics in Mainland China and Taiwan. Cambridge,
UK: Cambridge University Press.
Shleifer, A. and R. W. Vishny (1993). Corruption. Quarterly Journal of Economics 108,
599–617.
Singer, M. (2009). Buying voters with dirty money: the relationship between clien-
telism and corruption. Presented at the Annual Meeting of the American Political
Science Association 3–6 September, Toronto.
Singh, Shane P. (2010). Contextual influences on the decision calculus: A cross-
national examination of proximity voting. Electoral Studies 29, 425–134.
Smets, K. (2010). A widening generational divide? Assessing the age gap in voter
turnout between younger and older citizens, PhD dissertation, European University
Institute.
References 229
Smets, K. (2012). A widening generational divide? The age gap in voter turnout
through time and space. Journal of Elections, Public Opinion and Parties 22(4),
407–430.
Smets, K. and Van Ham, C. (2013). The embarrassment of riches? A meta-analysis of
individual level research on voter turnout. Electoral Studies 32(2), 344–359.
Smith, A. E. (2009). Legitimate grievances: Preferences for democracy, system
support, and political participation in Bolivia. Latin American Research Review
44, 102–126.
Stegmaier, Mary, and Michael S. Lewis- Beck (2013). Economic voting. In Rick
Valelly (eds), Oxford Bibliographies in Political Science Online. New York: Oxford
University Press.
Stepan, Alfred (1990). On the tasks of a democratic opposition. Journal of Democracy
1: 41–9.
Stockemer, D. (2015). Turnout in developed and developing countries: Are the two
turnout functions different or the same? Political Science 67(1), 3–20.
Stockemer, Daniel (2015). District magnitude and electoral turnout: A macro-level
global analysis. Acta Politica 50(1), 82–100.
Stockemer, D. (2016). Is the turnout function in democracies and nondemocracies
alike or different?. Politics and Policy 44, 889–915.
Stockemer, D. (2017). What affects voter turnout? A review article/meta-analysis of
aggregate research. Government and Opposition 52(4), 698–722.
Stockemer, Daniel, Bernadette LaMontagne, and Lyle Scruggs (2013). Bribes and
ballots: The impact of corruption on voter turnout in democracies. International
Political Science Review 34(1), 70–92.
Stokes S. C. (2005). Perverse accountability: A formal model of machine politics with
evidence from Argentina. American Political Science Review 99(3), 315–325.
Stolle, Dietlind and Marc Hooghe (2005). Inaccurate, exceptional, one-sided or irrele-
vant? The debate about the alleged decline of social capital and civic engagement in
Western societies. British Journal of Political Science 35, 149–167.
Strate John M., Charles J. Parrish, Charles D. Elder and Coit Ford (1989). Life span
civic development and voting participation. The American Political Science Review
83(2), 443–464.
Studlar, D. T., I. McAllister, and B. C. Hayes (1998). Explaining the gender gap in
voting: A cross-national analysis. Social Science Quarterly 79(4), 779–798.
Sundström, Aksel, and Daniel Stockemer (2015). Voter turnout in the European
regions: The impact of corruption perceptions. Electoral Studies 40, 158–169.
Surin, Maisrikrod (2002). Political reform and the new Thai electoral system. In
John Fuh-Sheng Hsieh and David Newman (eds), How Asia Votes. New York:
Chatham House.
Svolik, Milan W. (2015). Which democracies will last? Coups, Incumbent takeovers,
and the dynamic of democratic consolidation. British Journal of Political Science
45, 715–738.
Tambe, E. B. (2016). Who votes in East Asia? European Journal of East Asian Studies
15(2), 149–173.
Tambe, E. B. (2017a) Electoral participation in African democracies: The impact of
individual and contextual factors. Commonwealth & Comparative Politics 55(2),
119–140.
Tambe, E. B. (2017b). [Review] Aina Gallego (2015). Unequal Political Participation
Worldwide. Party Politics 23(5), 605–606.
230 References
Tambe, E. B. (2018a) Electoral participation in new democracies: applying existing
models of turnout to new democracies in Africa, East Asia and Post-Communist
Europe. Doctoral thesis, University of Sussex.
Tambe, E. B. (2018b). Who Does Not Vote and Why? Implication for New Democracies,
Elections –A Global Perspective. In Ryan M. Yonk (ed.), A Global Perspective.
London: IntechOpen.
Tambe, E. B. (forthcoming). Why do the poor and less educated vote more? Explaining
the reverse relationship between socio-economic status and turnout in Africa.
Paper presented at the African Politics Conference in 2020.
Tavits, Margit (2005). The development of stable party support: Electoral Dynamics
in post-communist Europe. American Journal of Political Science 49(2), 283–298.
Tavits, Margit (2008). Policy positions, issue importance, and party competition in
new democracies. Comparative Political Studies 41(1), 48–72.
Teixeira, R. A. (1984). Why Americans Don’t Vote: Turnout Decline in the United States
1960–1984. New York: Greenwood Press.
Teixeira, R. A. (1987), Why Americans Don’t Vote: Turnout Decline in the United States
1960–1984. New York: Greenwood Press.
Teixiera, R. A. (1992). The Disappearing American Voter. Washington: Brookings
Institute.
Telles, Helcimara, and Alejandro Moreno (eds) (2013). Comportamento Eleitoral e
Comunicação Política na América Latina. Belo Horizonte: Editora UFMG.
Teorell, J., and A. Hadenius (2009). Elections as levers of democracy: A global inquiry.
In S Lindberg (ed.), Democratization by Elections: A New Mode of Transition.
Johns Hopkins University Press.
Teorell, J., P. Sum, and M. Tobiasen (2007). Participation and political equality: An
assessment of large- scale democracy. In J. W. van Deth, J. R. Montero and
A. Westholm (eds), Citizenship and Involvement in European Democracies: A
Comparative Analysis. New York: Routledge.
Teorell, Jan, and Michael Wahman (2018). Institutional stepping stones for democ-
racy: How and why multipartyism enhances democratic change. Democratization
25(1), 78–97.
Terry, Lynn Karl (1986). Imposing consent: Electoralism and democratization in El-
Salvador. In Paul W. Drake and Eduardo Silva (eds), Elections and Democratization
in Latin America, 1980–1985. La Jolla, CA: University of California-San Diego,
Center for International Studies.
Thies, Michael F. (2002). Changing how the Japanese vote: The promise and pitfalls
of the 1994 electoral reform. In John Fuh-Sheng Hsieh and David Newman (eds),
How Asia Votes. New York: Chatham House.
Thomas, Craig W. (1998). Maintaining and restoring public trust in government
agencies and their employees. Administration & Society 30(2), 166–193.
Thomassen, J. (1994). Special issue, The intellectual history of election studies.
European Journal of Political Research 25 (3).
Thomassen, J. (2000). From comparable to comparative electoral research. In J. van
Deth, H. Rattinger, and E. Roller (eds), Die Republik auf dem Weg zur Normalita
¨t? Wahlverhalten und Politische Einstellungen nach Acht Jahren Einheit. Opladen:
Leske þ Budrich.
Thomassen, J. (ed.) (2005). The European Voter. Oxford: Oxford University Press.
Timpone, R. J. (1998). Structure, behavior, and voter turnout in the United States.
American Political Science Review 92(1), 145–158.
References 231
Tocqueville A. D. (1994). Democracy in America. New York: Knopf:
Topf, Richard (1995). Electoral participation. In H. Klingemann, and D. Fuchs (eds),
Citizens and the State. Oxford: Oxford University Press.
Tripp, A. M., A. Kang (2008). The global impact of quotas on the fast track to increased
female legislative representation. Comparative Political Studies 41, 338–361.
Tsai, Chia-hung (2001). Why do Taiwanese vote? Journal of Electoral Studies 8(2),
125–158.
Tucker, Joshua (2001). Economic conditions and the vote for incumbent parties in
Russia, Poland, Hungary, Slovakia and the Czech Republic from 1990–1996. Post-
Soviet Affairs 17(4), 309–331.
Tucker, Joshua (2006). Regional Economic Voting: Russia, Poland, Hungary, Slovakia
and the Czech Republic, 1990–1999. Cambridge: Cambridge University Press.
Turner, A. W. (1993). Postauthoritarian Elections. Comparative Political Studies 26, 3.
Uhlaner, C. J., B. E. Cain, and D. R. Kiewiet (1989). Political participation of ethnic
minorities in the 1980s. Political Behavior 11(3), 195–231.
Van Biezen, Ingrid, Peter Mair, and Thomas Poguntke (2012). Going, going … gone?
The decline of party membership in contemporary Europe. European Journal of
Political Research 51, 24–56.
van der Brug, Wouter, Mark Franklin, and Gabor Toka (2008). One electorate or
many? Differences in party preference formation between new and established
European democracies. Electoral Studies 27(4), 589–600.
van der Meer, Tom W. G. (2017). Political trust and the “crisis of democracy”. Oxford
Research Encyclopedia of Politics.
Van Deth, J. (1989). Interest in politics. In M. K. Jennings and J. van Deth (eds),
Continuities in Political Action. Berlin: De Gruyter.
Van Egmond, M. (2003). Rain falls on all of us (but some manage to get more wet
than others): Political context and electoral participation. Doctoral dissertation,
University of Amsterdam.
Van Egmond, De Graafi and Van Der Eijk (1998). Electoral participation in the
Netherlands: Individual and contextual influences. European Journal of Political
Research 34, 281–300.
Verba, S., and N. H. Nie (1972). Participation in America. New York: Harper &
Row.
Verba, S., N. H. Nie, and J.-O. Kim, (1978). Participation and Political Equality:
A Seven-nation Comparison. Chicago, IL: University of Chicago Press.
Verdery, Katherine (1996). What Was Socialism, and What Comes Next? Princeton,
NJ: Princeton University Press.
Warren. M. E. (2004). What does corruption mean in a democracy? American Journal
of Political Science 48(2), 328–343.
Wass, H. (2007). The effects of age, generation and period on turnout in Finland 1975–
2003. Electoral Studies 26(3), 648–659.
Wattenberg, Martin P. (1991). The Rise of Candidate-centered Politics: Presidential
Elections of the 1980s. Cambridge: Harvard University Press.
Webb, P. D. (1995). Are British Political Parties in Decline?. Party Politics 1,
299–322.
Webb, Paul, David Farrell, and Ian Holliday (eds) (2002). Political Parties in Advanced
Industrial Democracies. Oxford: Oxford University Press.
Webb, Paul, and Stephen White (2007). Political Parties in New Democracy.
Oxford: Oxford University Press.
232 References
Wedeman, Andrew (2002). Development and corruption: The East Asian paradox.
In Edmund Terence Gomez (ed.), Political Business in East Asia. London, UK:
Routledge.
Welzel, C., and Dalton, R. J. (2015). From allegiant to assertive citizens. In R. J. Dalton
and C. Welzel (eds), The Civic Culture Transformed: From Allegiant to Assertive
Citizens. Cambridge, UK: Cambridge University Press.
Weyland, K. (2003). Economic voting reconsidered: Crisis and charisma in the election
of Hugo Chávez. Comparative Political Studies 36(7), 822–848.
Whitefield, Stephen, and Robert Rohrschneider (2009). Understanding cleavages in
party systems issue position and issue salience in 13 post-communist democracies.
Comparative Political Studies 42(2), 280–313.
Wirls, D. (1986). Reinterpreting the gender gap. Public Opinion Quarterly 50, 316–330.
Wofinger, R. E (1965). The development and persistence of ethnic voting. The
American Political Science Review 59(4), 896–908.
Wolfinger, R. E. and Rosenstone, S. J. (1980). Who Votes? New Haven: Yale University
Press.
Yadav, Yogendra (2000). Understanding the second democratic upsurge: Trends of
Bahujan participation in electoral politics in the 1990s. In Francine Frankel, Zoya
Hasan, Rajeev Bhargava and Balveer Arora (eds), Transforming India: Social and
Political Dynamics of Democracy. Oxford: Oxford University Press.
Young, D. J. (2009). Is clientelism at work in African elections? A study of voting
behavior in Kenya and Zambia. Afrobarometer Working Paper 106.
Zakaria, Fareed (1994). A conversation with Lee Kwan Yew. Foreign Affairs 73,
109–127.
Zhengxu, Wang, and Leihua Ye (2017). Political participation in East Asia. In T.
Cheng, and Y. Chu (eds), Routledge Handbook of Democratization in East Asia.
London: Routledge.
Zmerli, S., and Newton, K. (2016). Objects of political and social trust: Scales and
hierarchies. In S. Zmerli and T. W. G. van der Meer (eds), Handbook on Political
Trust. Cheltenham, UK: Edward Elgar Publishing.
Index
Note: Page numbers in italics indicate figures and in bold indicate tables on the
corresponding pages.