This is the “accepted manuscript” version of the article:

Oser, Jennifer, and Marc Hooghe. 2018. Democratic Ideals and Levels of Political
Participation: The Role of Political and Social Conceptualisations of Democracy.
The British Journal of Politics and International Relations 20 (3):711-730.

Link to final version, published by SAGE Publishing:

https://doi.org/10.1177/1369148118768140

Democratic Ideals and Levels of Political Participation:

The Role of Political and Social Conceptualizations of Democracy

Abstract
Concerns about democratic legitimacy in contemporary democracies bring new urgency to
understanding how citizens’ attitudes and ideals affect their political activity. In this article
we analyze the relationship between citizens’ democratic ideals and political behaviour in
the European Social Survey’s 2012 uniquely extensive questions on these topics in 29
countries. Using latent class analysis, we identify two groups of citizens who emphasize
different citizenship concepts as discussed by T.H. Marshall, namely a political rights and
a social rights conception. The multilevel regression analyses indicate that those who
emphasize social rights have relatively high levels of non-institutionalized political
participation, but are less involved in institutionalized participation. In contrast, those who
emphasize political rights are more active in all forms of participation. We conclude by
discussing the implications of the findings that, even in an era of economic austerity, those
who emphasize social rights are not the most politically active.

Keywords: Political behaviour; democratic ideals; political and social citizenship; T.H.
Marshall; European Social Survey; latent class analysis

Acknowledgments:
This paper was presented at the ECPR joint sessions of workshops in Nottingham in
April 2017, and the authors thank Jeremy Albright, Ondřej Císař, Swen Hutter, Sofie
Marien, and Kateřina Vráblíková for their comments.

Funding:
The author(s) disclosed receipt of the following financial support for the research,
authorship, and/or publication of this article: M. Hooghe acknowledges funding by the
European Research Council, Advanced Grant 295920, “Democratic Linkage”.
A vibrant debate is underway regarding how citizens’ assessment of their country’s

democratic functioning has an impact on their political behaviour. A variety of studies

indicate that protest and non-institutionalized political participation are on the rise,

particularly in the wake of the economic downturn that began in 2008, and subsequent

austerity measures (Grasso & Giugni, 2016). Dissatisfaction with democratic functioning

along with the rise of populist sentiment have gained attention as important factors for

understanding these recent trends. Although this literature covers diverse aspects of

democratic dissatisfaction, it tends to share the theoretical presumption that citizens in

contemporary democracies have rather similar conceptions of what an ideal democracy

would look like. Yet, whether citizens do indeed share similar democratic ideals—and how

these ideals may be related to their political behaviour—has not yet been tested empirically.

Considering the protest wave following politics of austerity, one might assume that citizens

who stress social dimensions of democracy, such as fairness and economic quality, will be

more likely to participate than citizens who define politics mostly in strictly political and

procedural terms.

The implications of this debate are of paramount importance in light of the evidence

that electoral participation has declined in recent years (Amnå & Ekman 2014;

Dassonneville & Hooghe 2018; Hooghe & Kern 2017; McDonald 2010), while some forms

of non-electoral participation are on the rise (Dalton 2015; van Deth et al. 2007; Theocharis

& van Deth 2018). Research on the determinants of these different types of participation

has reached new insights regarding associations with conceptions of desirable behaviours

of ‘good’ citizens (Bolzendahl & Coffé 2013; Dalton 2008; Marien et al. 2010). Little is

known, however, about the relationship between citizens’ conceptions of what is important

2
in a well-functioning democracy and their patterns of political behaviour, nor about the

precise mechanism connecting ideals and participation behaviour. Concerns about

democratic legitimacy in contemporary democracies have brought new urgency to

investigating the long-standing question of how citizens’ attitudes and ideals shape their

political activity (Levi et al. 2009; Magalhães 2014).

This study aims to advance research on this topic by investigating whether citizens’

conceptions of democracy affect their political behaviour. More precisely, we test whether

citizens who hold distinct conceptions of democratic ideals also have distinctive

corresponding patterns of political participation. Given the recent concern about social

policy and inequality in Europe, it might be expected that citizens who place an emphasis

on the social responsibility of democracy would be more likely to protest against the

politics of austerity that has been pursued in recent years in numerous European countries.

An emphasis on social rights should thus be associated with a higher propensity to take

part in non-institutionalized forms of political participation.

We contribute to the ongoing debate by analyzing the relationship between citizens’

democratic ideals and political behaviour in the European Social Survey’s 2012 uniquely

extensive battery of questions on citizens’ conceptions of ideal democracy in 29 countries

(n=54,673). Using latent class analysis, we identify two groups of citizens who emphasize

distinct citizenship concepts as already discussed in the seminal work of T.H. Marshall

(1950), namely a political rights conception of democracy, versus a social rights

conception. The identification of these distinctive democratic ideals challenges the

prevailing approach which often assumes that citizens on average have the same

conception of what a democracy should look like.

3
 In multilevel regression analyses, the findings indicate that the political rights group

is highly active in all types of political behaviour, while the social rights group is less active

overall. This finding means that those who place a high priority on social rights are less

active in raising their voice toward the political system. While this group emphasizes

protection from inequality and poverty as important for the functioning of democracy,

apparently this normative ideal does not lead to more political participation, even in the

context of the economic downturn that began in 2008, and the subsequent austerity

measures in European countries. We discuss the implications of these findings that even in

an era of economic austerity, those who emphasize social rights are not the most politically

active, and therefore might not get their voice heard in the political decision-making

process.

What Affects Political Behaviour?

One of the most studied topics in political science is the question of what affects political

behaviour, including studies of resources, motivation and recruitment networks (Verba et

al. 1995; Schlozman et al. 2012), personality traits (Hibbing et al. 2013; Mondak &

Halperin 2008), contextual factors such as income inequality and institutional structures

(Hooghe & Oser 2016; Lister 2007), and most recently, online communication and

engagement (Cantijoch et al. 2016; Keating & Melis 2017). Amidst this wealth of research

on factors that affect political behaviour, empirical studies are surprisingly silent on the

relationship between democratic ideals and political behaviour—a fundamental subject of

theoretical inquiry in political science, dating back all the way to Aristotle’s writings on

political community.

4
 Recent empirical research on the topic of how citizens' conceptions of democracy

might affect how they engage with the political system focuses primarily on the effect of

citizens' dissatisfaction with democracy, with 'grievance theory' predicting that this

dissatisfaction motivates political engagement (Kriesi et al. 2012; Tarrow 1994; Webb

2013), while 'critical citizens' theory' predicts political inaction or a focus on non-

institutionalized activity (Dahlberg et al. 2015; Klingemann 1999; Norris 1999). These

approaches are based on the idea that actors are likely to show behaviour that is congruent

with the values they themselves consider to be salient. Although the main claims of these

two theories stand in opposition in terms of their prediction of political behaviour, it is

noteworthy that both depart from a shared assumption that citizens on average have similar

conceptions of what an ideal democracy should look like.

The empirical evidence for the assumption that citizens have similar democratic

ideals, however, is rather thin. Most often research relies on standard questions about the

way democracy functions in the country of the respondent (Ariely 2015). Some country-

specific research has indicated, however, that different democratic concepts can be

identified among citizens (Schedler & Sarsfeld 2007). The assumption that citizens’

concepts of democracy are the same across nations clearly requires further investigation.

This question, furthermore, has become more salient given the increasing challenges to the

legitimacy of European political systems since the economic crisis that began in 2008

(Cordero & Simón 2016; Streeck & Schäfer 2013). In various European countries, citizens

have taken to the streets to protest against austerity politics (Morlino & Quaranta 2016;

Quaranta 2016), and populist challenger parties have systematically increased their vote

share in recent elections (Bermeo & Bartels 2014; Hooghe & Oser 2015; Kern, Marien &

5
Hooghe 2015; Peterson, Wahlström & Magnus 2015; Rüdig & Karyotis; Sloam 2014). One

plausible conclusion from these trends is that policies to alleviate social and economic

hardship are considered important elements of citizens’ conception of ideal democracy, at

least for a specific portion of the population. This conclusion would imply that differences

in democratic ideals can help us to explain patterns of political participation.

How Can Democratic Ideals Can Affect Political Behaviour? Theory and Hypotheses

In democratic theory, the seminal essay on citizenship concepts by T.H. Marshall (1950)

has played an important role in the debate about the rights that are considered central for

full democratic citizenship. In this essay, he reviewed the historical evolution of three kinds

of rights: civil rights that are the basic building blocks of individual freedom; political

rights, which include the rights of citizens to participate in order to exercise political power;

and finally social rights, which include the right to basic economic welfare and security. In

this three-step evolution, the final step of social rights in the 20th century is understood as

a main contribution to theories of citizenship and democracy (Lister 2005; Møller &

Skaaning 2010).

While Marshall considered these three types of citizenship rights as stages in modern

welfare states’ social and political development, an additional important perspective is the

degree to which the mass public in contemporary democracies embrace any or all of these

citizenship rights as democratic ideals. Although the basic civil rights Marshall discussed

(such as property rights) are generally taken for granted in contemporary democracies it

remains an open question to what degree the other two types of citizenship rights (political

rights and social rights) are fully implemented in democratic countries. This lack of a full

6
implementation means that empirical research is particularly important for investigating

the degree to which the mass public embraces the social and/or political rights concept as

central to their conception of an ideal democracy and how these democratic ideals have an

effect on citizens’ political engagement. Yet, empirical studies of these citizenship

concepts have been scarce (Bulmer & Reese 1996; Hooghe & Oser, forthcoming). As

Ariely (2011) notes, extant research tends to focus on just one set of citizenship rights, even

though Marshall’s theory emphasizes that all three of these rights should be ensured in a

holistic manner (Lister 2005; Revi 2014). Because of his particular focus on the social

rights concept, an important innovation in Marshall’s framework was that ensuring social

rights is not merely a matter of ideological preference, but must be considered as central in

achieving basic rights for citizens in modern democracies (Connell 2012).

Building on Marshall’s theoretical framework, we focus on the two types of

citizenship rights that are most salient for current political debates, namely the ‘political

rights’ and ‘social rights’ conceptions as democratic ideals. We are particularly interested

in how citizens who embrace these difference democratic ideals will be politically engaged

in the period following the economic downturn of 2008 and the subsequent implementation

of austerity measures. Building on recent literature that shows that democratic attitudes and

norms such as civic duty can affect traditional political behaviour such as voting (Galais &

Blais 2016), we aim to test whether distinctive democratic ideals may motivate specific

types of political participation. By building on the assumption that normative ideals indeed

influence behaviour, we can connect the more empirically oriented literature on political

participation with the broader theoretical and normative literature on ideal concepts of

democracy.

7
 A first straightforward expectation is that citizens who hold both of these types of

democratic ideals regarding what they believe to be important for democracy will be

politically engaged – particularly in comparison to citizens who do not have any strongly

expressed democratic ideal. The underlying mechanism is that we assume that actors act

upon the interiorized norms that they consider as most salient. This mechanism motivates

our first hypothesis:

H1: Citizens who express clearly defined ‘political rights’ or ‘social rights’ ideals

will be more politically engaged in all activities (institutionalized or non-

institutionalized political activities) in comparison to those who lack a strongly felt

democratic ideal.

In addition, an important theoretical issue is the comparison of the ‘political rights’

and ‘social rights’ ideals in terms of their relative engagement in different types of political

behaviour. Maintaining our focus on citizen responses to the recent economic downturn

and its aftermath, the literature would suggest that these groups will behave differently in

terms of the political activity they direct toward established institutions in comparison to

their non-institutionalized political behaviour. More specifically, we expect that citizens

who place particular emphasis on classic political rights, such as free and fair elections,

will be particularly active in institutionalized political behaviour due to their inherent

attention to political institutions and procedures. In contrast, in line with the recent

literature on protest events following the economic downturn, we would expect that those

who place particular emphasis on the importance of social rights, such as protection from

poverty, will direct their attention to a more diverse set of actors who respond to social

8
rights concerns and have an effect on economic and social policy. Therefore, the

mechanism behind this expectation is that those motivated by social rights concerns will

direct their political voice to actors such as trade unions, social movements, and civic

associations, and will therefore be particularly active in non-institutionalized political

behaviour. These expectations lead us to formulate the following two hypotheses:

H2: Citizens who hold a ‘political rights’ democratic ideal will have higher levels of

institutionalized political behaviour than citizens holding a ‘social rights’ democratic

ideal.

H3: Citizens who hold a ‘social rights’ democratic ideal will have higher levels of

non-institutionalized political behaviour than citizens holding a ‘political rights’

democratic ideal.

Data and Methods: Democratic Ideals as Determinants of Political Behaviour

Empirical measures of the quality of democracy routinely tend to focus on expert

evaluations and institutional measures (Coppedge et al. 2016). When individuals are called

upon to express their conceptions of democracy, some studies suggest that similar concepts

are evident across diverse countries (Dalton et al. 2007), but other studies are more

skeptical of this claim (Canache et al. 2001). Despite the challenge of comprehensively

measuring democratic ideals, a growing body of studies have investigated different aspects

of democratic concepts in individual countries (Bengtsson & Christensen 2016; Schedler

& Sarsfield 2007) or specific regions (Canache 2012; Doorenspleet 2012).

9
 In this study we heed Doorenspleet’s (2015) rallying cry to ask ‘where are the

people?’ when assessing democratic concepts cross-nationally by using an actor-centered

approach. This approach is made possible by the European Social Survey’s 2012 wave

(ESS 2012) which included a uniquely extensive module to investigate how Europeans

view contemporary democracy (Ferrín & Kriesi 2016). Standardized in-person interviews

were conducted between 2012 and 2013 among representative survey samples in 29

European countries. The special one-time module on democratic ideals asked respondents

to rate the importance of various aspects of democracy.

The democratic ideals battery allows us to test Marshall’s ideas as it includes both

political rights-related items such as conducting free and fair elections, as well as social

rights-related items such as protecting citizens against poverty. Although public opinion

research has extensively studied general preferences on social rights related topics

(Reeskens & van Oorschot 2013), the ESS 2012 module on democratic views is unique in

investigating these items as part of a respondent’s comprehensive conception of ideal

democracy. The question was phrased as ‘democracy in general’, in order to avoid that

respondents would interpret this as a question about the specific functioning of their own

political system.

In the current study, we build upon prior work that investigates various aspects of

these ideals (Hooghe et al. 2017; Oser & Hooghe 2018) to make a subsequent connection

between democratic ideals and political behaviour. This module of the ESS also includes

standard and well-established questions on a variety of political acts that have been

included in the ESS and similar surveys, and we analyze these political behaviors using a

10
standard conceptual and empirical distinction in the literature between institutionalized and

non-institutionalized political behaviour (Barnes & Kaase 1979; Marien et al. 2010).

A clear observation from the descriptive statistics of these democratic ideals is that

all of the indicators are rated as highly important by respondents (Table 1). The highest

rated item is clearly the rule of law (‘The courts treat everyone the same’) with a mean

score of 9.23. Of particular interest to the present study is that classic items of social rights

are clearly considered important aspects of democracy, with protection against poverty

ranking fourth most important (mean=8.74) and reducing inequality ranked eighth

(mean=8.26). These mean scores indicate that European citizens view social rights as

important aspects of an ideal democracy, along with other elements that have received

more extensive attention as fundamental aspects of democracy, such as free media (ranked

10th). On average, social rights clearly are important to Europeans’ conception of ideal

democracy, and are by no means considered to be outside the realm of democratic politics.

[Table 1 about here]

While these mean scores show little variance in the average importance of these items

for survey respondents as a whole, in order to assess citizens’ comprehensive democratic

ideals it is necessary to employ an actor-centered analytical technique. To assess whether

Marshall’s citizenship concepts are salient in public opinion, it is particularly important to

identify discrete groups of citizens who give a high score to some indicators while

simultaneously giving a low score on other items in the same battery. This kind of

typological construct can be identified through the use of latent class analysis (LCA),

11
which identifies separate groups of respondents who share similar response patterns on a

battery of indicators (Magidson & Vermunt 2004; McCutcheon 1987). Finite mixture

models such as LCA are widely used in the social and health sciences in what has been

described as a ‘person-oriented approach’ (Collins & Lanza 2010, 8). LCA identifies

distinct subgroups of respondents that share similar response patterns on a series of items,

just like standard variants of cluster analysis (Hagenaars & McCutcheon 2002). A main

advantage of LCA in comparison to traditional cluster analysis, however, is that the

probabilistic estimation method used in LCA yields objective goodness of fit statistics that

provide reliable indicators for assessing the optimal number of latent classes to fit the data

(Raftery 1995; Vermunt & Magidson 2002). We therefore rely on this technique to identify

distinctive conceptions of ideal democracy.

Results

1. Democratic ideals (independent variable)

The goodness-of-fit statistics generated by the LCA of the democratic ideals battery

indicate that optimal fit to the data is provided by a five-class model. As documented in

the appendix, this five-class solution is cross-culturally equivalent so it can be used for

comparative analyses.

Among the five latent classes depicted in Figure 1, it is clear that three of these groups

attribute similar levels of importance to all surveyed aspects of democratic ideals, namely

the ‘high ideals’ (24% of the research population), ‘medium ideals’ (30%) and the ‘low

ideals’ (10%) groups. The interpretation of the uniform scores across all items for these

three groups is not clear cut, since this uniformity may suggest a lack of clear preferences,

12
or the existence of generalized priorities. For the purpose of next-step analyses, these

groups pose interpretive challenges, as the similar scores on a large battery of items may

be the result of an attempt to measure a complex concept among respondents who do not

have a coherent attitude, as well as social desirability biases. These findings clearly

indicate, however, that almost two-thirds (64%) of those surveyed in the ESS 2012 consider

all democratic ideals indicators to be of a similar level of importance (high, medium or

low), and therefore they do not place special emphasis on any particular set of elements of

democracy. As such, these groups do not provide evidence regarding the distinction

Marshall introduced between political and social rights.

In contrast, two groups depicted in Figure 1 emphasize the importance of certain

aspects of democracy over others, which is evident in the crossing connective lines that

mark the ‘social’ and ‘political’ ideals groups. The group labeled ‘social ideals’ (20%)

emphasizes the importance of social rights elements, such as reducing income level

differences and protecting citizens against poverty. Alternatively, the ‘political ideals’

group (16%) emphasizes Marshall’s classic political rights such as media’s freedom to

criticize the government and the free and fair conduct of elections.

[Figure 1 about here]

As noted, the measurement equivalence tests indicate that these groups can be validly

compared across societies, and Table 2 depicts the prevalence of these groups across the

countries included in the ESS 2012. When we compare the distribution of citizens who

adhere to these distinctive ideals we find marked differences between countries, indicating

sufficient variance to test whether these distinctive democratic ideals may have an impact

on citizens’ political behaviour.

13
 [Table 2 about here]

2. Political participation (dependent variable)

Building on the political participation literature, we draw on data that cover

institutionalized and non-institutionalized acts of political participation (Copeland 2014;

Oser 2017; Quaranta 2013; Sloam 2016; Stolle & Hooghe 2011; Vráblíková 2014). In the

European Social Survey in 2012, the questionnaire asks whether respondents participated

in various forms of participation in the last 12 months. The political acts surveyed, listed

in Table 3, include electoral-oriented activities such as working in a political party, as well

as non-institutionalized activities such as boycotting products for political reasons.

The mean values in Table 3 show that the prevalence of participation in these political

activities ranges from the most prevalent act of signing a petition (.19) to the least prevalent

act of working in a political party or action group (.04). As is customary for these kinds of

participation indicators the correlation matrix indicates that the items are not particularly

highly correlated, ranging from a high correlation of .33 (between signed petition and

boycott) to a low of 0.10 (party work and boycott).

[Table 3 about here]

The rotated factor analysis results documented in Table 3 suggest a separate factor

for institutionalized participation (party work, contact a politician, and organizational

work) versus a non-institutionalized participation index (signed a petition, boycotted a

14
product, and demonstrated). The only item that does not clearly load on a single factor,

wearing a badge or sticker, is excluded from the participation indices.

Analysis of the bivariate relationship between the democratic ideals and the three

participation measures suggests significant differences between groups (overall

participation, F=331.57, p<0.000; institutionalized participation, F=171.95, p<0.000; non-

institutionalized participation, F=314.93, p<0.000). Table 4 displays mean values for

participation measures by democratic ideals groups, and the results show that the political

rights group consistently scores highest on participation levels of all kinds. Even in these

bivariate relationships, however, the results suggest that the participation levels of the

social rights group vary depending on what kind of participation is measured. For non-

institutionalized participation the social rights group has fairly high participation levels,

but for institutionalized participation, the social rights group has a low score that is lower

than the overall mean for all respondents. The following section investigates these

relationships further in a series of multilevel logistic regressions with the inclusion of

appropriate individual-level and country-level controls.

[Table 4 about here]

3. Multilevel regression analyses

In order to assess whether the democratic ideals identified in the LCA are related to

political participation patterns, we conduct a series of multivariate analyses. As the data

are hierarchically structured with individual respondents nested in countries, multi-level

regression is the appropriate technique for conducting the multivariate analyses (Gelman

and Hill 2007). In line with our theoretical expectations, we focus mainly on the difference

15
between the ‘political rights’ and the ‘social rights’ group. Our first set of analyses includes

all five democratic ideals groups that were identified in the LCA with the 'medium ideal'

as the reference group.1 We then proceed to conduct parallel analyses that compare only

the 'political rights' and 'social rights' groups to test whether a significant difference can be

discerned in the political behaviour of those who hold these two different democratic

ideals.

In the regression analyses we analyze binary dependent variables, coded 1 for

respondents who participated in any of the political acts included in the scale, and 0 for

respondents who did not. This coding allows the consistent, valid use of these items across

all models as dependent variables in multilevel logit regression analyses (Rabe-Hesketh &

Skrondal 2012). As the dependent variable of political participation additive indices is a

count outcome, a negative binomial regression model is an ideal approach when the data

allow.2 However, since the sum scales of the political participation items are highly skewed

with a large standard deviation, the negative binomial regression models successfully

converge only when all five democratic ideals groups are included, but do not converge

when we proceed to test the comparison of only the social rights and political rights groups.

We therefore include parallel tables in the appendix for all findings for which the negative

binomial regression results can be obtained, and these findings show that the analyses using

a negative binomial specification are substantively indistinguishable from the multilevel

logistic regression models.

Individual-level controls include those commonly used in the study of participation,

including socio-demographic variables of age, sex, education, and income, and attitudinal

measures of left-right ideological placement and satisfaction with democracy (Marien,

16
Hooghe & Quintelier 2010; Verba, Schlozman & Brady 1995). Country-level controls

include democratic stability, as measured by the Polity IV dataset (Marshall, Gurr &

Jaggers 2013). The Gini index is used as a measure of inequality based on income survey

data from the Luxembourg Income Study (LIS 2015) that is available for 21 countries in

the present study.3 The measure of Good Governance is based on World Bank data. GDP

per capita is also included as a control variable, since we expect participation levels to be

higher in more affluent societies. Finally, a control for (post-)authoritarian country context

is included.4 In the first regression analysis (Table 5) we introduce each of these country-

level variables one by one due to data limits of multicollinearity and the limited number of

country-level observations. We then present the findings of a model that includes a linear

combination of the country-level variables to create an overall country-level composite

variable (Model VII). As the results of all of these models are substantively

indistinguishable in all regression analyses for this study, in subsequent regressions

analyses reported in the article we present findings based on models that include only the

more parsimonious composite variable (see appendix for descriptive statistics and

additional coding information).

The first test uses the overall political participation measure as a dependent variable

(Table 5) and the findings show that both the social rights and political rights groups are

more active than the ‘medium’ reference group. The group with low ideals seems least

inclined to take part in political participation, while the high ideals group is active but with

lower scores than the political rights group. In comparison to the bivariate relationships

presented above, the fully specified regression models indicate that a social rights

democratic ideal has a positive impact on political participation, but that the effect is

17
relatively small. The individual level and country level findings for additional control

variables generally confirm prior findings in the field. It is noteworthy that even when

country-level controls are included in the model, all of the relevant controls have

significant effects in the expected directions, but the controls do not eliminate the continued

importance of the various democratic ideals for predicting overall participation.

In relation to our hypotheses, these findings lend support to our first hypothesis (H1)

by showing that citizens who express a clearly defined ‘political rights’ or ‘social rights’

ideals are indeed more politically engaged in all activities in comparison to the groups that

lack a strongly felt democratic ideal. Those who have a clear conception of their democratic

ideal are also more engaged in various forms of political behaviour.

We proceed from these findings on overall participation (Table 5) to focus on

institutionalized and non-institutionalized participation as distinct dependent variables

(Figure 2). While the regression findings from Table 5 have shown that the social ideals

group has a relatively high level of overall political participation in comparison to the

reference group, the predicted values documented in Figure 2 shows that the social rights

group engages differently in institutionalized versus non-institutionalized activities (see

appendix for full regression tables output for logistic and negative binomial specifications).

The findings in Figure 2 show that the social rights group is not particularly active in

institutionalized acts of participation, although it is relatively active in non-institutionalized

forms of participation.

Moreover, the predicted values plots show a substantively sizeable effect of

democratic ideals on political participation patterns in comparison to standard individual

level and country controls. For the political rights ideal, the only variable that has a larger

18
effect on either type of political participation is the highest level of education, which is

well documented as having among the most consistent, sizeable effects on political

engagement in decades of political science research (Verba 2003; Persson 2015). While

the social rights ideal's effect on institutionalized participation is marginally significant and

substantively small, its effect size for non-institutionalized participation exceeds the effect

size of standard control variables of age, sex and left-right political orientation.

To summarize the findings: while the political rights group is active in both types of

participation, the social rights group specializes in non-institutionalized participation while

scoring lower on institutionalized forms of political participation.

[Figure 2 about here]

The findings thus far compare the political participation patterns of the social and the

political rights groups with those of all other democratic ideals groups. Following our

interest to determine whether there are significant differences between the social rights and

the political rights group, we proceed to narrow the analytical focus to these two groups

only. We follow the same order, first comparing scores for political participation in general,

followed by the analysis for institutionalized and non-institutionalized participation.

Controlling for the same battery of individual-level and country-level determinants of

political participation, we find that those who adhere to a political rights democratic vision

are significantly more active than the social rights group. The full regression tables for

these findings are documented in the appendix, and Figure 3 presents in a concise manner

the predicted values of the multi-level logistic regression for the social rights and political

rights ideals on the different measures of political participation.

[Figure 3 about here]

19
 Figure 3 shows that for all measures of political participation, we observe a

significant and similar gap between the political rights and the social rights group. In

relation to our hypotheses, these findings support the hypothesis that citizens who hold a

political rights democratic ideal will be more politically active in institutionalized

participation in comparison to those who hold a social rights ideal (H2). However, these

findings do not support our final hypothesis that those who hold a social rights ideal will

be more politically active in non-institutionalized participation in comparison to the

political rights group (H3). Apparently, the citizens that are inherently motivated by a

concern about poverty and inequality do not convert this concern info forms of political

action, even if we apply a very broad definition of various forms of participation. The

predicted values plotted in the figure show that the gap between the political rights and

social rights group in relation to their political behaviour is meaningful in size, even after

taking into account a large battery of control variables. The results are clear: the group that

focuses on political rights is the most active one, no matter the specific form of participation

being investigated. Even in times of economic crisis, a concern about social rights does not

seem to be the most powerful driving force for political engagement.

Discussion
In this study we have taken advantage of new data on the relationship between citizens’

conceptions of ideal democracy and their patterns of political participation. A number of

studies on the protest reaction to austerity measures in Europe conjure images of masses of

concerned citizens taking to the street to express their democratic ideals through political

action (Rüdig & Karyotis 2014). The findings reported in the present study, however,

indicate that the political participation of citizens who care deeply about social citizenship

20
rights is muted in comparison to those who give priority to political rights. Contrary to our

expectations, therefore, an adherence to a social concept of democracy does not seem to be

a strong mobilizing tool. This finding might help us to explain why, in contrast to some

expectations, concern about inequality and austerity politics has not led to enduring forms

of protest or other forms of participation in Europe. Part of the answer to this question

might lie in the fact that the social rights group simply is not highly active in all forms of

political participation. Indeed, our findings show that concern about social rights is not the

main driving force for political engagement.

Three main findings from the research advance the current understanding of the

relationship between political attitudes and political behaviour. First, even though

Europeans express very high average levels of support for a wide variety of elements of

democracy, a person-centered approach shows that distinct concepts of democracy can be

identified among contemporary Europeans. We find a plurality that considers all elements

of democracy to be of similar levels of importance, but we also find those who place

particular emphasis on a social or political rights conception of democracy as introduced

in the work of T.H. Marshall. Second, it is clear that these concepts of democracy are

predictive of distinctly different patterns of political behaviour, even when controlling for

a variety of individual and country-level determinants that have been widely studied in the

literature. The findings show that our interest in citizens' democratic ideals has implications

beyond the realm of normative theory, as these democratic ideals clearly have real life

consequences for the way citizens interact with their political system. Finally, the findings

show that the two most distinctive democratic ideals of political rights and social rights

have different effects on political participation. While those who emphasize political rights

21
are highly engaged in all kinds of political activities, the findings show that the social rights

supporters tend to be less active. Although the social rights group adheres to a democratic

ideal that holds political institutions responsible for ensuring lower levels of inequality,

apparently they are less inclined to participate to put pressure on the political decision-

makers in order to advance this ideal.

The current analysis does not allow us to explain why we find this relation between

an emphasis on social rights and lower levels of political participation. As it has been noted

that governments’ responsiveness to citizens’ electoral participation often declines in a

period of austerity (Streeck and Schafer 2013), the results raise the question of the direction

of causality in the relationship between democratic ideals and political behaviour. Since

the current study is based on a cross-sectional survey fielded in 2012 that included a

uniquely comprehensive battery of questions on democratic ideals, the requisite

longitudinal data do not exist to determine the direction of the causal mechanism. We also

have to take into account the possibility that the causal effect might flow in the direction

from participation to attitudes, as experiences with political participation can also have an

effect on citizens’ attitudes and expectations.

This unresolved question of causal mechanisms poses something of a challenge for

the broader debate on democratic legitimacy. On the one hand it is clear that rising levels

of inequality might have an effect on the way citizens view their political system. The

social rights concept essentially means that a substantial part of a country’s population

places emphasis on the importance of protecting its citizens against poverty and social

exclusion. If the democratic system fails to provide this social and economic protection, it

does not live up to normative standards, which should have a negative effect on democratic

22
legitimacy. Simultaneously, however, the findings of the current study show that the social

rights democratic ideal is less easily translated into protest and political participation of all

types.

In contrast, if the political rights group is not satisfied with the political system’s

ability to live up to their democratic ideal, (for example, the democratic character of the

elections in their country), a likely outcome is that they will protest or participate in other

forms of political behaviour to get their voice heard. For those who feel left out because

their government does not protect them against the risk of poverty or inequality, this

manner to express their concern is less likely. The findings of the present study raise a

puzzle, since non-institutionalized participation is often considered to be a more

challenging and resource-intensive form of behaviour. An important topic for future

investigation is therefore whether citizens who highly value social citizenship rights and

simultaneously refrain from institutionalized participation are doing so due to a lack of

belief in the efficacy or legitimacy of the political system to realize their democratic ideals.

What does remain clear, however, is that the mismatch between citizens’ expectations on

social rights, and the input the political system receives on this issue, might pose a

structural problem for the legitimacy of democratic governance.

23
References

Agresti, A. (2007). An Introduction to Categorical Data Analysis (2nd ed.). Hoboken,

N.J.: Wiley.

Amnå, E., & Ekman, J. (2014). Standby citizens: diverse faces of political passivity.
European Political Science Review, 6(2), 261-281.

Ariely, G. (2011). Exploring citizenship spheres of inclusion/exclusion: rights as

'potential for power'. Patterns of Prejudice, 45(3), 241-258.

Ariely, G. (2015). Democracy-Assessment in Cross-National Surveys: A Critical

Examination of How People Evaluate Their Regime. Social Indicators Research,
121(3), 621-635.

Barnes, S. H., & Kaase, M. (Eds.). (1979). Political action: Mass participation in five
Western democracies. Beverly Hills CA: Sage.

Bermeo, N., & Bartels, L. (Eds.). (2014). Mass politics in tough times: Opinions, Votes,
and Protest in the Great Recession. New York: Oxford University Press.

Bengtsson, Å., & Christensen, H. (2016). Ideals and actions: Do citizens' patterns of
political participation correspond to their conceptions of democracy? Government
and Opposition, 51(2), 234-260.

Bolzendahl, C., & Coffé, H. (2013). Are ‘good’ citizens ‘good’ participants? Testing
citizenship norms and political participation across 25 nations. Political Studies,
61(S1), 45-65.

Bulmer, M., & Rees, A. M. (1996). Citizenship today: The contemporary relevance of
T.H. Marshall. London: UCL Press.

Canache, D. (2012). Citizens' conceptualizations of democracy: Structural complexity,

substantive content, and political significance. Comparative Political Studies,
45(9), 1132-1158.

Canache, D., Mondak, J. J., & Seligson, M. A. (2001). Meaning and measurement in
cross-national research on satisfaction with democracy. Public Opinion Quarterly,
65(4), 506-528.

Cantijoch, M., Cutts, D., & Gibson, R. (2016). Moving slowly up the ladder of political
engagement: A ‘spill-over’ model of internet participation. British Journal of
Politics & International Relations, 18(1), 26-48.

Collins, L. M., & Lanza, S. T. (2010). Latent class and latent transition analysis.
Hoboken: Wiley.

24
Connell, A. (2012). Civil rights and social welfare: Some thoughts on the contemporary
relevance of T. H. Marshall. The Political Quarterly, 83(3), 551-559.

Copeland, L. (2014). Conceptualizing political consumerism: How citizenship norms

differentiate boycotting from buycotting. Political Studies, 62(S1), 172-186.

Coppedge, M., Lindberg, S., Skaaning, S.-E., & Teorell, J. (2016). Measuring high level
democratic principles using the V-Dem data. International Political Science
Review, 37(5), 580-593.

Cordero, G., & Simón, P. (2016). Economic crisis and support for democracy in Europe.
West European Politics, 39(2), 305-325.

Dahlberg, S., Linde, J., & Holmberg, S. (2015). Democratic discontent in old and new
democracies: Assessing the importance of democratic input and governmental
output. Political Studies, 63, 18-37.

Dalton, R. J. (2008). Citizenship norms and the expansion of political participation.

Political Studies, 56(1), 76-98.

Dalton, R. J. (2015). The good citizen: How a younger generation is reshaping American
politics (2nd ed.). Washington DC: CQ Press.

Dalton, R. J., Shin, D. C., & Jou, W. (2007). Understanding democracy: Data from
unlikely places. Journal of Democracy, 18(4), 142.

Dalton, R., & Welzel, C. (Eds., 2014). The civic culture transformed: From allegiant to
assertive citizens. Cambridge: Cambridge University Press.

Dassonneville, R., & Hooghe, M. (2018). Indifference and alienation: Diverging

dimensions of electoral dealignment in Europe. Acta Politica, 53(1), 1-23.

Doorenspleet, R. (2012). Critical citizens, democratic support and satisfaction in African

democracies. International Political Science Review, 33(3), 279-300.

Doorenspleet, R. (2015). Where are the people? A call for people-centred concepts and
measurements of democracy. Government and Opposition, 50(3), 469-494.

ESS Round 6 (2012). European Social Survey Round 6 (ESS-6) 2012, data file edition
2.3. Oslo: NSD - Norwegian Centre for Research Data.

Ferrín, M., & Kriesi, H. (Eds., 2016). How Europeans View and Evaluate Democracy.
Oxford: Oxford University Press.

25
Galais, C., & Blais, A. (2016). Beyond rationalization: Voting out of duty or expressing
duty after voting? International Political Science Review, 37(2), 213-229.

Gelman, A., & Hill, J. (2007). Data analysis using regression and multilevel/hierarchical
models. Cambridge: Cambridge University Press.

Grasso, M. & Giugni, M. (2016). Protest participation and economic crisis: The
conditioning role of political opportunities. European Journal of Political
Research, 55(4), 663-680.

Hagenaars, J. A., & McCutcheon, A. L. (Eds., 2002). Applied latent class analysis. New
York: Cambridge University Press.

Hibbing, M. V., Ritchie, M., & Anderson, M. R. (2013). Personality and political
discussion. Political Behavior, 33(4), 601-624.

Hooghe, M., & Kern, A. (2017). The tipping point between stability and decline: Trends
in voter turnout, 1950–1980–2012. European Political Science, 16(4), 535-552.

Hooghe, M., Marien, S., & Oser, J. (2017). Great expectations: the effect of democratic
ideals on political trust in European democracies. Contemporary Politics, 23(2),
214-230.

Hooghe, M., & Oser, J. (2015). The electoral success of the Movimento 5 Stelle. An
example of a left populist vote? Austrian Journal of Political Science, 44(4), 25-
36.

Hooghe, M., & Oser, J. (2016). Trade union density and social expenditure: a
longitudinal analysis of policy feedback effects in OECD countries, 1980-2010.
Journal of European Public Policy, 23(10), 1520-1542.

Hooghe, M., & Oser, J. (Forthcoming). Social and political citizenship in European
public opinion: An empirical analysis of T.H. Marshall’s concept of social rights.
Government and Opposition, 1-26. Published online before print May 4, 2017,
doi:10.1017/gov.2017.11.

Kankaraš, M., Moors, G., & Vermunt, J. K. (2011). Testing for measurement invariance
with latent class analysis. In E. Davidov, P. Schmidt, & J. Billiet (Eds.), Cross-
cultural analysis: Methods and applications (pp. 359-384): Routledge.

Kankaraš, M., & Vermunt, J. K. (2014). Simultaneous latent class analysis across groups.
In A. C. Michalos (Ed.), Encyclopedia of Quality of Life and Well-Being Research
(pp. 5969-5974). Heidelberg: Springer.

26
Keating, A., & Melis, G. (2017). Social media and youth political engagement: Preaching
to the converted or providing a new voice for youth? The British Journal of
Politics and International Relations, 19(4), 877-894.

Kern, A., Marien, S., & Hooghe, M. (2015). Economic crisis and levels of political
participation in Europe (2002-2010): The role of resources and grievances. West
European Politics, 38(3), 465-490.

Klingemann, H.-D. (1999). Mapping political support in the 1990s: A global analysis. In
P. Norris (Ed.), Critical citizens: Global support for democratic governance (pp.
31-56). Oxford: Oxford University Press.

Levi, M., Sacks, A., & Tyler, T. (2009). Conceptualizing Legitimacy, Measuring
Legitimating Beliefs. American Behavioral Scientist, 53(3), 354-375.

LIS, Luxembourg Income Study (2015). Inequality and poverty key figures.
http://www.lisdatacenter.org/lis-ikf-webapp/app/search-ikf-figures. Accessed
June 19, 2015.

Lister, M. (2005). ‘Marshall-ing’ Social and Political Citizenship: Towards a Unified

Conception of Citizenship. Government and Opposition, 40(4), 471-491.

Lister, M. (2007). Institutions, inequality and social norms: Explaining variations in

participation. British Journal of Politics & International Relations, 9(1), 20-35.

Magalhães, P. C. (2014). Government effectiveness and support for democracy.

European Journal of Political Research, 53(1), 77-97.

Magidson, J., & Vermunt, J. K. (2004). Latent class models. In D. Kaplan (Ed.), The
Sage handbook of quantitative methodology for the social sciences (pp. 175-198).
London: Sage.

Marien, S., Hooghe, M., & Quintelier, E. (2010). Inequalities in non-institutionalised

forms of political participation: A multi-level analysis of 25 countries. Political
Studies, 58(1), 187-213.

Marshall, M. G., Gurr, T. R., & Jaggers, K. (2013). POLITY IV Project: Political regime
characteristics and transitions, 1800-2013 (Dataset Users’ Manual). Vienna:
Center for Systemic Peace. Retrieved from
http://www.systemicpeace.org/inscrdata.html.

Marshall, T. H. (1950). Citizenship and social class and other essays. Cambridge:
Cambridge University Press.

McCutcheon, A. L. (1987). Latent class analysis. London: Sage.

27
McDonald, M. P. (2010). American voter turnout in historical perspective. In J. Leighley
(Ed.), Oxford handbook of American elections and political behavior. Oxford:
Oxford University Press.

Møller, J., & Skaaning, S.-E. (2010). Marshall revisited: The sequence of citizenship
rights in the twenty-first century. Government and Opposition, 45(4), 457-483.

Mondak, J. J., & Halperin, K. D. (2008). A framework for the study of personality and
political behaviour. British Journal of Political Science, 38(2), 335-362.

Morlino, L., & Quaranta, M. (2016). What is the impact of the economic crisis on
democracy? Evidence from Europe. International Political Science Review, 37(5),
618-633.

Norris, P. (Ed., 1999). Critical Citizens: Global support for democratic governance.
Oxford: Oxford University Press.

Oser J. (2017) Assessing how participators combine acts in their "political tool kits": A
person-centered measurement approach for analyzing citizen participation. Social
Indicators Research 133(1): 235-258.

Oser, J., & Hooghe, M. (2018). Give me your tired, your poor? Support for social
citizenship rights in the United States and Europe. Sociological Perspectives,
61(1), 14-38.

Persson, M. (2015). Education and Political Participation. British Journal of Political

Science, 45(3), 689-703.

Peterson, A., Wahlström, M., & Wennerhag, M. (2015). European anti-austerity protests
– Beyond "old" and "new" social movements? Acta Sociologica, 58(4), 293-310.

Quaranta, M. (2013). Measuring political protest in Western Europe: assessing cross-

national equivalence. European Political Science Review, 5(3), 457-482.

Quaranta, M. (2016). Protesting in "hard times": Evidence from a comparative analysis of

Europe, 2000-2014. Current Sociology, 64(5), 736-756.

Rabe-Hesketh, S., & Skrondal, A. (2012). Multilevel and Longitudinal Modeling Using
Stata: Stata Press.

Raftery, A. E. (1995). Bayesian model selection in social research. Sociological

Methodology, 25, 111-163.

Reeskens, T., & van Oorschot, W. (2013). Equity, equality, or need? A study of popular
preferences for welfare redistribution principles across 24 European countries.
Journal of European Public Policy, 20(8), 1174-1195.

28
Revi, B. (2014). T.H. Marshall and his critics: reappraising 'social citizenship' in the
twenty-first century. Citizenship Studies, 18(3-4), 452-464.

Rüdig, W., & Karyotis, G. (2014). Who protests in Greece? Mass opposition to austerity.
British Journal of Political Science, 44(3), 487-513.

Schedler, A., & Sarsfield, R. (2007). Democrats with adjectives: Linking direct and
indirect measures of democratic support. European Journal of Political Research,
46(5), 637-659.

Schlozman, K. L., Verba, S., & Brady, H. E. (2012). The unheavenly chorus: Unequal
political voice and the broken promise of American democracy. Princeton:
Princeton University Press.

Sloam, J. (2014). "The outraged young": Young Europeans, civic engagement and the
new media in a time of crisis. Information, Communication & Society, 17(2), 217-
231.

Sloam J. (2016) Diversity and voice: The political participation of young people in the
European Union. British Journal of Politics and International Relations 18(3):
521-537.

Streeck, W., & Schäfer, A. (2013). Politics in the Age of Austerity. Cambridge: Polity
Press.

Stolle, D., & Hooghe, M. (2011). Shifting inequalities: Patterns of exclusion and
inclusion in emerging forms of political participation. European Societies, 13(1),
119-142.

Tarrow, S. (1994). Power in movement: Social movements, collective action, and politics.
Cambridge: Cambridge University Press.

Theocharis, Y., & van Deth, J. W. (2018). The continuous expansion of citizen
participation: a new taxonomy. European Political Science Review, 10(1), 139-
163.

Van Deth, J. W., Montero, J. R., & Westholm, A. (Eds., 2007). Citizenship and
involvement in European democracies: A comparative analysis. London:
Routledge.

Verba, S. (2003). Would the dream of political equality turn out to be a nightmare?
Perspectives on Politics, 1(4), 663-679.

Verba, S., Schlozman, K. L., & Brady, H. E. (1995). Voice and equality: Civic
voluntarism in American politics. Cambridge, MA: Harvard University Press.

29
Vermunt, J. K., & Magidson, J. (2002). Latent class cluster analysis. In J. A. Hagenaars,
& A. L. McCutcheon (Eds.), Applied latent class analysis (pp. 89-106). New
York: Cambridge University Press.

Vráblíková, K. (2014). How context matters? Mobilization, political opportunity

structures, and nonelectoral political participation in old and new democracies.
Comparative Political Studies, 47(2), 203-229.

Webb, P. (2013). Who is willing to participate? Dissatisfied democrats, stealth democrats

and populists in the United Kingdom. European Journal of Political Research,
52(6), 747-772.

30
Table 1. Mean scores on ‘democratic ideals’ battery in 29 European countries

Description Abbreviation Mean

1. The courts treat everyone the same courts fair 9.23
2. The government explains its decisions to voters govt expl. 8.89
3. National elections are free and fair fair elec. 8.85
4. The government protects all citizens against poverty poverty 8.74
5. The media provide citizens with reliable information to judge the govt. media info. 8.73
6. Governing parties are punished in elections when they have done a bad job party acc. 8.46
7. The rights of minority groups are protected minority 8.27
8. The government takes measures to reduce differences in income levels income eq. 8.26
9. Opposition parties are free to criticise the government opposition 8.18
10. The media are free to criticise the government free media 8.06
11. Different political parties offer clear alternatives to one another party alter. 7.95

Notes: European Social Survey, 29 countries (n=48,805); ESS 2012 v2.3. Appropriate weights applied:
population size weight ('pweight') and post-stratification weights that account for design weight as well as
sampling error and non-response error ('pspwght', newly available in v2.3 of the ESS 2012 data). Prefatory
survey question: ‘Using this card, please tell me how important you think it is for democracy in general
that…’. Responses coded on a 0 to 10 scale where 0 indicates ‘not at all important’ and 10 indicates
‘extremely important’.

31
Table 2. Democratic ideals, distribution of citizens across countries into five ideals

Social Political High Medium Low Total n

Albania 31.59 7.31 46.48 13.49 1.13 1,149
Belgium 18.47 15.69 12.75 39.95 13.13 1,835
Bulgaria 20.42 18.53 39.29 17.76 4.01 2,072
Switzerland 19.66 21.85 12.66 38.97 6.86 1,414
Cyprus 18.05 13.71 40.93 25.58 1.74 1,036
Czech Republic 18.93 17.68 18.11 28.47 16.81 1,844
Germany 20.63 28.84 16.76 29.35 4.42 2,899
Denmark 16.43 31.66 16.82 32.29 2.80 1,570
Estonia 16.96 14.73 28.16 29.36 10.80 2,241
Spain 29.71 6.80 32.05 25.42 6.02 1,794
Finland 19.41 16.86 11.74 42.61 9.38 2,112
France 19.57 14.15 18.38 38.36 9.55 1,937
United Kingdom 20.84 12.43 20.21 34.52 12.00 2,083
Hungary 15.53 11.64 37.91 21.70 13.22 1,899
Ireland 15.36 11.65 22.03 34.64 16.32 2,506
Israel 19.96 16.34 23.23 34.64 5.83 2,350
Iceland 14.86 29.16 24.48 27.92 3.58 727
Italy 29.62 11.37 25.90 29.07 4.04 915
Lithuania 17.31 10.98 21.92 31.61 18.19 1,930
Netherlands 12.85 18.34 10.36 44.43 14.02 1,805
Norway 17.62 27.19 17.81 33.79 3.59 1,589
Poland 26.82 17.02 26.82 25.48 3.86 1,786
Portugal 17.92 4.01 30.82 25.13 22.12 2,093
Russian Federation 21.32 12.75 27.21 23.65 15.08 2,275
Sweden 13.53 31.17 25.14 26.72 3.44 1,774
Slovenia 33.96 11.18 21.84 28.12 4.90 1,163
Slovakia 15.49 12.23 16.16 35.52 20.59 1,782
Ukraine 22.26 10.88 32.13 27.12 7.62 1,995
Kosovo 21.93 6.35 37.95 23.24 10.53 1,149
TOTAL 19.82 15.98 24.04 30.36 9.80 51,724

Source: ESS 2012. Notes: Entries are latent class analysis findings for percentage distribution of population
in each country among the five latent classes. Row = 100 % in every country.

32
Table 3. Political participation acts, dimensionality

Factor 1 Factor 2
Political act Mean Institutionalized Non-institutionalized
Worked in a political party or action group 0.036 0.786 0.000
Contacted a politician, government or local official 0.108 0.671 0.125
Worked in another organization or association 0.124 0.538 0.335
Signed a petition 0.194 0.183 0.744
Boycotted certain products 0.151 -0.027 0.758
Taken part in a lawful public demonstration 0.079 0.225 0.534
Work or displayed a campaign badge/sticker 0.066 0.497 0.372
Source: European Social Survey 2012 (n=52,606). Factor analysis results based on exploratory principle
components factor analysis with a varimax rotation. Substantively similar results obtain with a promax
rotation. Eigenvalues for the rotated factors are 1.69 for Factor 1, and 1.68 for Factor 2.

33
Table 4. Participation levels of democratic ideals groups
Overall Institutionalized Non-institutionalized
participation participation participation
Mean S.D Mean S.D Mean S.D
Political rights group 0.534 0.499 0.316 0.465 0.427 0.495
Social rights group 0.386 0.487 0.206 0.405 0.300 0.458
High ideals group 0.385 0.487 0.216 0.412 0.303 0.458
Medium ideals group 0.378 0.485 0.217 0.412 0.276 0.447
Low ideals group 0.227 0.419 0.133 0.339 0.151 0.358
All respondents 0.380 0.485 0.216 0.411 0.289 0.454

Source: ESS 2012 (n=54,520).

34
Table 5. Explaining overall political participation

Model Model Model Model Model Model Model

I II III IV V VI VII
Individual-level variables
Democratic ideals
(ref: Medium ideals)
Low ideals -0.573*** -0.418*** -0.467*** -0.419*** -0.419*** -0.419*** -0.418***
(0.041) (0.043) (0.046) (0.043) (0.043) (0.043) (0.043)
High ideals 0.190*** 0.327*** 0.290*** 0.330*** 0.330*** 0.330*** 0.328***
(0.028) (0.031) (0.034) (0.031) (0.031) (0.031) (0.031)
Political rights 0.445*** 0.454*** 0.504*** 0.455*** 0.455*** 0.455*** 0.454***
(0.030) (0.033) (0.035) (0.033) (0.033) (0.033) (0.033)
Social rights 0.179*** 0.202*** 0.209*** 0.204*** 0.204*** 0.204*** 0.203***
(0.029) (0.031) (0.034) (0.031) (0.031) (0.031) (0.031)
Age -0.001* -0.001* -0.001 -0.001* -0.001* -0.001* -0.001*
(0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001)
Sex (1=female) -0.101*** -0.072** 0.005 -0.075*** -0.075*** -0.075*** -0.072**
(0.020) (0.022) (0.024) (0.022) (0.022) (0.022) (0.022)
Education (ref: Low)
Medium 0.245*** 0.449*** 0.410*** 0.445*** 0.444*** 0.444*** 0.448***
(0.027) (0.030) (0.033) (0.030) (0.030) (0.030) (0.030)
High 0.686*** 1.016*** 0.970*** 1.012*** 1.012*** 1.011*** 1.016***
(0.027) (0.031) (0.033) (0.030) (0.030) (0.030) (0.031)
Left-right -0.057*** -0.030*** -0.042*** -0.030*** -0.030*** -0.030*** -0.030***
(0.004) (0.005) (0.005) (0.005) (0.005) (0.005) (0.005)
Income feeling
(ref: very difficult)
Difficult 0.233*** 0.115* 0.077 0.110* 0.110* 0.110* 0.115*
(0.043) (0.047) (0.058) (0.047) (0.047) (0.047) (0.047)
Coping 0.580*** 0.172*** 0.138** 0.167*** 0.167*** 0.168*** 0.171***
(0.040) (0.045) (0.054) (0.044) (0.044) (0.044) (0.045)
Living comfortably 1.024*** 0.307*** 0.265*** 0.299*** 0.298*** 0.300*** 0.306***
(0.043) (0.049) (0.057) (0.048) (0.048) (0.048) (0.049)
Satisfaction democracy 0.055*** -0.021*** -0.022*** -0.021*** -0.021*** -0.021*** -0.021***
(0.004) (0.005) (0.006) (0.005) (0.005) (0.005) (0.005)
Country-level variables
Established democracy 0.017***
(0.003)
Gini -10.090*
(4.036)
Good governance 0.028***
(index) (0.006)
GDP/cap 0.048***
(1000 USD) (0.009)
(Post-)authoritarian -1.255***
(0.231)
Country-level 0.334***
composite variable (0.056)
Constant -1.226*** -1.803*** 2.380* -3.057*** -2.424*** -0.295 -0.975***
(0.060) (0.204) (1.182) (0.511) (0.318) (0.173) (0.121)
Random intercept 0.342*** 0.519** 0.463*** 0.393*** 0.381*** 0.292***
(0.093) (0.162) (0.123) (0.105) (0.101) (0.079)
Observations 43739 43081 34769 43739 43739 43739 43081
Notes: ESS 2012 in 29 countries. Multi-level logistic regression. * p < 0.05, ** p < 0.01, *** p < 0.001

35
Figure 1. Democratic ideals held by five groups of citizens
Medium (30%) High (24%) Social (20%) Political (16%) Low (10%)

3.0

2.5

2.0

1.5

1.0
courts fair govt expl. fair elec. poverty media info. party acc. minority income eq. opposition free media party alter.

Source: European Social Survey, 2012 (n=51,724).

Notes: Latent class analysis conditional probabilities for optimal partial equivalence model that includes
country covariate and applies design weights. Y-axis: conditional probabilities that the indicator is important
for democracy. Findings based on 3-point coding of the original 11-category democratic ideal items: 0-7
recoded as 1; 8-9 recoded as 2; 10 recoded as 3. For documentation of alternate coding that yielded similar
results see appendix Section III for discussion of model selection, measurement equivalence tests and
robustness tests.

36
Figure 2. Institutionalized and non-institutionalized participation, predicted values

Note: see appendix for related regression tables for multilevel logistic specification, as well as negative
binomial specification and marginal effects plots.
Figure 3. Social and political groups compared for all types of participation

Participation All

Institutionalized Participation

Non-institutionalized Participation

Notes: x-axis measures the predicted values for the social and political democratic ideals groups on the
different measures of political participation, based on regressions with the full battery of controls (see
appendix for additional documentation).

38
Endnotes

1
As a robustness test to our findings, in an alternate formulation of the reference group
we combined the three democratic ideals groups that make no distinction between the
importance level of the democratic ideals indicators (the ‘low’, ‘medium’ and ‘high’
democratic ideals groups) into a single reference variable category. There are no
substantive differences between the findings using this alternate reference group in
comparison to those presented in the article.
2
Although a viable alternative count model specification to the negative binomial is the
Poisson model, a likelihood ratio test that compares the fit of these two regression models
for our data shows a significant improvement of a negative binomial model compared to
a Poisson model.
3
The eight countries in the ESS 2012 that lack Gini coefficients in the LIS study are
Albania, Bulgaria, Cyprus, Iceland, Lithuania, Portugal, Ukraine and Kosovo.
4
The countries in the data that qualify as ‘(post-)authoritarian’ include Albania, Bulgaria,
Czech Republic, Estonia, Spain, Hungary, Lithuania, Poland, Portugal, Russia, Slovenia,
Slovakia, Ukraine, and Kosovo. We also ran all models with a ‘post-communist’ control,
which excluded from the above list Spain and Portugal. As there were no significant
differences in the use of these different country-level controls, we report on the
authoritarian control only.

39