Justice in Achievement Matters: The Fairness of Educational Opportunities and Active Citizenship

Abstract

Unlike existing research which has focused mainly on the effects of educational attainment and curricula on active citizenship, the present article aims to study the relationship of subjective assessment regarding the fairness of people’s educational opportunities and their active social engagement, and how this relationship is embedded in different social environments. Theoretically, the analysis is based on the view of active citizenship as a multidimensional and domain-specific phenomenon. It is also inspired by the capability approach’s understanding of the opportunity aspect of freedom and the importance of fairness of opportunities and processes. Empirically, our study uses a multilevel linear regression model to analyse data from the European Social Survey 2018 for 29 countries. We develop a scale of active citizenship with four domains: political, social, workplace, and democratic values. Our findings show that a higher perceived unfairness of educational opportunities is associated with lower levels of active citizenship. They also reveal that although there is a negative relationship between the perceived unfairness of people’s opportunities to receive a desired level of education and their active citizenship, it is mitigated when people are living in high-trust societies and in countries which are more economically and democratically developed.

1. Introduction

John Dewey, one of the greatest sociologists and educationalists (Dewey [1916] 2001, p. 91), argues that there is a close relationship between the development of democracy and education due to the specific nature of democracy: it “is more than a form of government; it is primarily a mode of associated living”. Since then, several authors—political scientists, educationalists, sociologists—have continued to highlight the important role of education for the development of democracy and its impacts on social participation and civic engagement (Lipset 1959; Putnam 2000; Hoskins et al. 2008; Kam and Palmer 2008; Golubeva 2018). The theoretical basis of this traditional view is the understanding that civic knowledge, skills, attitudes, and values are developed through education and learning experiences that enable people to become active citizens (Hoskins et al. 2008; Hoskins and Mascherini 2009). It has also been argued that, in addition to its direct impact on political participation, the level of education “affects the acquisition of nearly all the other factors that facilitate participation: the well-educated are more likely to earn high incomes on the job; to develop civic skills at work … to be in social networks through which requests for political activity are mediated; and to be politically interested and knowledgeable” (Verba et al. 2003, p. 47). However, several recent studies have not been able to conclude in favour of education’s strong positive effect on civic participation and thus have put into question the relationship between education and active citizenship (Cassel and Lo 1997; Persson 2014; Milligan et al. 2004; Dee 2004).

These mixed findings about the relationship between education and active citizenship have provoked scholars to take into account the specificities of education beyond length or degree by investigating in their analyses, for example, how differences in the type of institution and field of study in higher education relate to political engagement (Mühleck and Hadjar 2023). Despite the increased level of education among the global population, inequalities in the opportunities that individuals from different social backgrounds have to reach the required educational level remain quite significant (Mare 1981; Blossfeld and Shavit 1993; Lucas 2001; Pfeffer 2008; Ilieva-Trichkova and Boyadjieva 2014). That is why it is worth looking not only at how having a specific kind of education influences civic participation but also at its effects on the process of acquiring the desired level of education, i.e., the availability and affordability for individuals from different social backgrounds and their assessment of this process. We will focus on one aspect of this issue—the relationship between the subjective assessment of the fairness of opportunities to achieve the levels of education which people strive for and their active citizenship.

In what follows, first, we discuss previous research and outline our theoretical considerations regarding our main concepts, “active citizenship” and “fairness of educational opportunities”. We propose an understanding of active citizenship as a multidimensional and domain-specific concept and interpret the importance of educational opportunities from the capability approach perspective. Then, we describe the data and the developed scale of active citizenship with its four dimensions. Next, we present our findings. The article concludes with a discussion of the results and some final remarks.

2. Previous Research and Theoretical Considerations

2.1. Understanding Active Citizenship

It is undoubtable that the understanding of active citizenship is based on the understanding of the core concept of citizenship. In his seminal lectures given in 1949 in Cambridge, the English sociologist T.H. Marshall (1950) defines citizenship as a legal status that captures an essential aspect of the relationship between the individual and the state. He proposes three elements of citizenship—civic, political, and social—each one composed of different rights (ibid., pp. 10–11). Based on Marshall’s legacy, thinkers adhering to different philosophical and political traditions (e.g., Beiner 1995) have not only proposed specific constellations of rights and obligations but have extended the very understanding of citizenship as a legal status. Given that citizenship practices occur within different groups, citizenship is closely linked to concepts such as role and identity and can be viewed as an expression of one’s membership in some kind of community (Kymlicka and Norman 1994; Holford and van der Veen 2006). Recently, the concept of citizenship has been further broadened and deepened as “not only linked to the national state but also to a regional identity such as European citizenship, Latin American citizenship, or Global citizenship” and as “extended from the political level to the social and cultural level” (Veugelers et al. 2017, p. 31; Milana 2008). Other authors highlight that citizenship “is undergoing radical redefinition under pressure from globalizing economies on the one hand and social movements of, for example, ecologists and feminists on the other” and thus turns into “a ‘site of struggle’” (Walters and Watters 2001, p. 473).

Like citizenship, the concept of active citizenship “has no single agreed definition” and “draws from an interdisciplinary set of literature, crossing the boundaries of social science, including education, sociology, psychology, political science, civil society and community development research” (Hoskins and Mascherini 2009, p. 461). It is a complex and multidimensional concept (Golubeva 2018). The notion of active citizenship focuses on individual actions and highlights the involvement of citizens in political and social spheres. The use of the term “active citizenship” derives mainly from the sphere of education as both a research and policy field (European Commission 1998; Holford and van der Veen 2006; Hoskins and Mascherini 2009). In the European policy document Education and Active Citizenship in the European Union (European Commission 1998), three dimensions of active citizenship are defined: an affective one, which refers to individuals’ and groups’ “sense of attachment to the societies and communities to which they theoretically belong and, therefore, is closely related to the promotion of social inclusion and cohesion as well as to matters of identity and values”; a cognitive one that captures the available information and knowledge as a basis for action; and a pragmatic dimension referring to different actions (European Commission 1998, p. 10). Holford and van der Veen (2006) differentiate between several domains of civic activity: civil society, the state, workplaces, and the private sphere. They also outline the value dimension of active citizenship by pointing out that “[a]ctive citizens have a strong sense of their place and responsibility in the world, and are driven by a sense of commitment to other people, rooted in notions of justice and care” (Holford and van der Veen 2006, p. 5). The value dimension of active citizenship has also been clearly highlighted by Hoskins and Mascherini (2009, p. 462), who argue that “[a]ctivities in which persons participate should be based on values and should not contravene principles of human rights and the rule of law”. In his study, Campbell (2006) distinguishes the following dimensions of engagement: political engagement, civic engagement, voting, trust, tolerance, and political knowledge. Depending on the aims of the study and available data, an author may use different dimensions of active citizenship. Thus, Hoskins and Mascherini (2009) refer to four dimensions: protest and social change, community life, representative democracy, and democratic values. Yang and Hoskins (2020) identify three dimensions of active citizenship: voting, political protest, and volunteering.

Trying to synthesise the main characteristics of active citizenship, Hoskins et al. (2008, p. 389) suggest the following definition: “[p]articipation in civil society, community and/or political life, characterised by mutual respect and non-violence and in accordance with human rights and democracy”. The authors underline that “this definition includes a broad range of activities which are considered necessary for a stable democracy and social inclusion”, and that “[a]lthough [a]ctive [c]itizenship is specified on the individual level in terms of actions and values, the emphasis in this concept is not on the benefit to the individual but on what these individual actions and values contribute to the wider society in terms of ensuring the continuation of democracy, good governance and social cohesion” (ibid.). Kersh et al. (2021) argue that active citizenship refers to political, social, and economic participation. Veugelers et al. (2017, p. 31) argue that the concept of citizenship has recently been broadened and deepened in the sense that it “has crossed national borders and can refer to a more regional or global identity … from the political level to the social and cultural level”. Based on an empirical quantitative study in several countries, Endrizzi and Schmidt-Behlau (2021, p. 38) conclude that the active connotation clarifies that active citizenship “is not participation as objective presence in or exposure to society, rather it implies awareness acquisition and change in attitude” and conclude that its conceptualisation has been influenced by the recent migration phenomenon in different countries.

Taking into account the above discussion, we define active citizenship as individuals’ engagement with and participation in the political sphere, civil society, workplaces, and community life in accordance with human rights and democratic values and for the benefit of broader society. This definition outlines that active citizenship is domain-specific and means not only involvement and participation in different social spheres but also personal engagement with their development, inspired by democratic values as well as care for others and society as a whole. It also suggests that active citizens do not limit their actions only to the political and civic spheres but are actively concerned with the development of their community and workplaces, as well.

2.2. Education and Active Citizenship

There is a vast body of research on the effects of education on active citizenship, especially in relation to the political and civic spheres—activities such as voting, protesting, and volunteering (Putnam 2000; Dee 2004; Milligan et al. 2004; Hillygus 2005; Campbell 2006). However, although most studies find a positive association between educational attainment and active citizenship, they disagree on whether this relationship reveals causal effects. Thus, some authors (Shields and Goidel 1997; Dee 2004; Milligan et al. 2004) argue that educational attainment has a significant causal effect on voting behaviour. Other authors (Luskin 1990; Cassel and Lo 1997; Persson 2014) suggest that there is a spurious relationship between education and political engagement, and that education’s influence may be a proxy for the effects of other individual characteristics such as intelligence, occupation, or parents’ socioeconomic status.

In order to increase the reliability of their findings, some scholars have undertaken more sophisticated—both theoretically and empirically/statistically—analyses. Thus, the relationship between education and active citizenship has been specified for different types of education: formal (Campbell 2006; Hoskins et al. 2008), higher education (Kam and Palmer 2011; Veugelers et al. 2014, 2017; Mühleck and Hadjar 2023), and lifelong and adult education (Holford and van der Veen 2006; Zepke 2013; Lodigiani 2010; Kersh et al. 2021). These authors also try not to limit themselves to one model, but are rather testing different models dealing with education’s impact on various dimensions of active citizenship. Campbell (2006), for example, studies three of them: the absolute education model, according to which the individual’s specific level of education boosts engagement; the sorting model, which assumes that individual education levels relative to the social environment facilitate active citizenship; and the cumulative model, which states that engagement increases with the average education level of one’s compatriots. He finds that different dimensions of active citizenship are better explained by different models, e.g., voting and membership in voluntary associations with the absolute education model versus interpersonal trust with the cumulative model.

Some studies also try to explain—both theoretically and empirically—the significant role of education in relation to active citizenship. Thus, Hillygus (2005) discusses three different hypotheses: (1) the civic education hypothesis; (2) the social network hypothesis; and (3) the political meritocracy hypothesis. His empirical analyses provide the greatest support for the civic education hypothesis, according to which education offers both the “skills necessary to become politically engaged and the knowledge to understand and accept democratic principles” (Hillygus 2005, p. 27). More concretely, the content of higher education, especially social science curricula developing language and civic skills, has been important in influencing political participation in America.

Most of the studies on higher education and citizenship focus on policies and practices of citizenship education in higher education (Brennan and Naidoo 2008; Watson 2008; Zgaga 2009; Veugelers et al. 2014, 2017). Some authors argue that there is a need for setting an Agenda for Critical-Democratic Citizenship in Universities (Veugelers et al. 2014). There are also increasing theoretical and empirical contributions to the literature on how university practices affect student participation in academic governance and how this promotes their political behaviour outside of the university (Yang and Hoskins 2020). Mühleck and Hadjar (2023) focus on how heterogeneities within higher education in terms of differences in the type of institution, kind of degree, and field of study are related to variations in political participation. Their results indicate that mechanisms that are behind the association between higher education and political participation differ by country, e.g., the association of political participation and type of institution is weaker in Norway, a less segregated country than Austria.

The existence of positive associations between having a higher education and active citizenship is widely accepted (e.g., Brennan et al. 2015). According to data from Eurostat (2017), people with a higher education are more active in all countries: 6.1% of those with low education (ISCED 0–2), 12.1% of those with medium (ISCED 3–4), and 20.6 of people with high education (ISCED 5–8) were active citizens in 2015. However, there are controversial statements on whether this association represents a causal relationship. Thus, Hoskins et al. (2008, p. 397), after controlling for a large set of variables, conclude in their study that the correlation between education and civic behaviour is a causal effect and that “the return to education is non-linear, with tertiary education being strongly associated with the three indicators of citizen behavior”.

An article by Kam and Palmer (2008) provoked discussion regarding the character of the effects of higher education attainment on active citizenship. The authors define their aim as to “question the extent to which higher education is a cause of political participation as opposed to a proxy for other, often unobserved, preadult experiences and predispositions” (Kam and Palmer 2008, p. 612). They start their analyses with the rejection of the assumption that higher education is a stochastic process, stressing instead that the likelihood of having a higher education degree is determined by a number of factors, such as parental characteristics and individual abilities. Based on this, they argue that “the durable empirical relationship between higher education and participation might not reflect higher education conveying participation-enhancing benefits, but rather, higher education serving as a proxy for preadult characteristics”, both parental and individual (ibid., p. 614). The authors highlight the main mechanisms by which parental characteristics influence higher educational attainment and subsequent adult active political engagement such as parental education, occupation, income, or values. They also point to different individual characteristics developed in preadult years like cognitive skills and personality traits. The empirical analyses suggest that “[r]ather than conferring benefits that in turn cause participation, higher education appears to proxy other factors that catalyze political participation” (ibid., p. 626, italics in the original).

The theses of Kam and Palmer have provoked a serious debate. Henderson and Chatfield (2011) and Mayer (2011) critique their findings, arguing that the conventional view that higher education per se does have an effect on political participation should not be rejected. In their response to this criticism, Kam and Palmer (2011) again repeat their argument that the existence of numerous selection issues related to the choice and completion of higher education problematises an unbiased causal estimation of its relationship with political participation. They claim (ibid., p. 662) that “what seems to be a ‘simple’ question of ‘Does higher education cause political participation?’ is nuanced, complex, and deserving of continued study” and have used alternative approaches to these methodological issues. Using longitudinal data for England, Persson (2014) also finds that there are no significant effects of higher education on political participation, concluding that educational attainment should be regarded as a proxy for rather than a cause of political participation.

A recent study by Yang and Hoskins (2020) continues the debate on whether education has an independent effect on political participation or is only a proxy for individuals’ social background, cognitive abilities, and values. They have used longitudinal data and controlled for people’s socioeconomic status, parental education, innate abilities, and some other variables. Their findings are nuanced—for young people, having a university degree has a positive independent effect on intentions to vote but a negative effect on intentions to volunteer and no significant relationship with intentions to protest in the future.

These nuanced and mixed results concerning the relationship between education and active citizenship point to the need to look for and take into account other characteristics of education, not just the fact of the completion of a given level of education but, for example, inequalities in access to education, opportunities that people have to obtain a given degree, and how they assess their fairness. In this article, we will focus on subjective assessments of the fairness of educational opportunities and their association with active citizenship.

2.3. Fairness of Educational Opportunities

The capability approach views living as a combination of various “doings and beings” (Sen 1993). At the heart of this approach are the freedoms and opportunities that people have in choosing a life that they have reason to value. It is also important to highlight that outcomes (“functionings”) are less significant than “opportunities”. Because of this, the very way human that capabilities are realised is crucial for a better conceptualisation and measurement of human and social development. The capability approach perspective implies that the attained educational level and years of schooling are not a sufficient measure when it comes to education, but that educational inequalities have to be also considered. According to Sen (Drèze and Sen 2002, p. 6):

[t]his crucial role of social opportunities is to expand the realm of human agency and freedom, both as an end in itself and as a means of further expansion of freedom … We shall be particularly concerned with those opportunities that are strongly influenced by social circumstances and public policy […]

Sen (2009, p. 296) also argues that any theory of justice “has to be alive to both fairness of the processes involved and to the equity and efficiency of the substantive opportunities that people can enjoy”.

The vital importance of the fairness of educational opportunities reflects the fact that educational inequalities are among the most important determinants of economic disparities and differences in individual civic participation. There are two important characteristics of educational inequalities: they are strongly influenced by people’s social background, and they are cumulative (e.g., Rubenson 1998; Di Prete and Eirich 2006). That is why the issue of the legitimacy of educational inequalities becomes indispensable for any study of education. However, the legitimacy of inequalities in education is not self-evident. According to the capability approach, unjust inequality relates more to freedom to achieve rather than actual achievements. As Sen (1992, p. 148) puts it:

[i]f the social arrangements are such that a responsible adult is given no less freedom (in terms of set comparisons) than others, but he still wastes the opportunities and ends up worse off than others, it is possible to argue that no unjust inequality may be involved.

Based on the above understanding of the importance of social opportunities in individuals’ personal and social lives, we will shift the focus from the relationship between educational attainment and active citizenship to that of subjective assessments regarding the fairness of opportunities that people have to receive the education that they strive for and their active social engagement (on the association of fairness assessments of educational opportunities with some individual characteristics, see (Stoilova and Ilieva-Trichkova 2023)).

Following Hoskins et al. (2008), it is important to emphasise that the relationship between education and active citizenship does not take place in a vacuum, and that several macro characteristics at the national level affect civic participation. Empirical investigations confirm this. Thus, although many studies reveal that higher education has the biggest effect on active citizenship (Hoskins et al. 2008; Eurostat 2017), there is a big difference between countries. Data show that among individuals with higher education, the highest rate of civic engagement is in France (24.6%), and the lowest rate is in Cyprus (2.1%), and several countries (Belgium, Bulgaria, Cyprus, the Czech Republic, Hungary, Romania, and Slovakia) are far below the EU-28 average. Using a comparative perspective, other authors (Mühleck and Hadjar 2023) demonstrate that there is no universal association between civic skills and civic participation, and that it is stronger or weaker depending on specific country features. Keeping in mind these results, in our study, we will analyse the mediating role of some country-specific characteristics such as GDP, democratic regime, and cultural values in the associations between subjective assessments of fairness in educational opportunities and active citizenship.

More concretely, we are going to answer two research questions (RQs):

RQ1:

How are subjective assessments of the fairness of people’s opportunities to achieve the level of education that they desire associated with their active citizenship?

RQ2:

How is the association between subjective fairness assessments of educational opportunities and active citizenship embedded in different economic, political, and cultural contexts?

3. Methodology

3.1. Data

Some international surveys have included special modules on active citizenship. The most comprehensive is the one that was carried out within the European Social Survey (ESS) 2002. There was also the Module on Social and cultural participation and Material deprivation in EU SILC 2015 (Eurostat 2017). The International Civic and Citizenship Education Study (ICCS), carried out in 2009 and 2016 aims to investigate the ways in which young people are prepared to undertake their current and future roles as citizens (Schulz et al. 2010).

We will use data from the European Social Survey 2018 (ESS Round 9: European Social Survey Round 9 Data 2018). This is the most recent wave of the ESS which includes a module on fairness and a special question about the subjective assessments of the fairness of educational opportunities based on which our main independent variable was constructed. To the best of our knowledge, two other studies have used data from the European Social Survey (ESS). Campbell (2006) uses the first round of the ESS in 2002, and Hoskins et al. (2008) base their analyses on data from the third round of the ESS in 2006/2007. As part of a research project supported by the European-wide network “Active Citizenship for Democracy”, a measurement of active citizenship in Europe has been developed: the Active Citizenship Composite Indicator (ACCI), revised in Hoskins and Mascherini (2009). This revised model is based on 61 indicators from the special module on active citizenship in the ESS 2006 data and refers to the following distinct forms of participation: representative democracy, protest and social change, community participation, and democratic values. The Hoskins et al.’s (2008) study uses a small part of these indicators.

More specifically, we have used data for 29 European countries and limited the age of respondents to 25–65 years and those who are citizens in the country where they were interviewed. The analytical sample in the present study consists of 27,472 individuals grouped in 29 countries. More specifically, these are Austria (n = 1432); Belgium (n = 952); Bulgaria (n = 1108); Croatia (n = 1080); Cyprus (n = 406); the Czech Republic (n = 1530); Denmark (n = 904); Estonia (n = 1034); Finland (n = 1049); France (n = 1009); Germany (n = 1283); Great Britain (n = 1106); Hungary (n = 981); Iceland (n = 536); Ireland (n = 1165); Italy (n = 1389); Latvia (n = 461); Lithuania (n = 1032); Montenegro (n = 785); the Netherlands (n = 967); Norway (n = 838); Poland (n = 883); Portugal (n = 568); Serbia (n = 1087); Slovakia (n = 626); Slovenia (n = 771); Spain (n = 904); Sweden (n = 849); and Switzerland (n = 737).

The data at the country level were extracted from a report from The Economist Intelligence Unit (2019); Eurostat, the UNESCO Institute for Statistics website, and the authors’ own calculations were obtained based on the European Social Survey 2018 (ESS Round 9: European Social Survey Round 9 Data 2018).

3.2. Variables

Our dependent variable is a scale of active citizenship. Based on our theoretical understanding of active citizenship as a multidimensional and domain-specific phenomenon and the available data, we calculated a scale of active citizenship which includes the following theoretically predefined dimensions and indicators: the political domain, the social domain, the workplace domain, and the democratic values domain. We selected items from the survey’s questionnaire and then mapped them to the four domains in order to cover their content:

Table A1. Items included in the active citizenship scale.

	Political Domain
1.	B13 Some people don’t vote nowadays for one reason or another. Did you vote in the last [country’s] national election in [month/year]?
2.	B15 During the last 12 months, have you done any of the following: …contacted a politician, government, or local government official?
3.	B16 During the last 12 months, have you done any of the following: … worked in a political party or action group?
4.	B20 During the last 12 months, have you done any of the following: … taken part in a public demonstration?
5.	B22 During the last 12 months, have you done any of the following: … posted or shared anything about politics online, for example, on blogs, via email, or on social media such as Facebook or Twitter?
	Social domain
6.	B17 During the last 12 months, have you done any of the following: … worked in another organization or association?
7.	B18. During the last 12 months, have you done any of the following: … worn or displayed a campaign badge/sticker?
8.	B19 During the last 12 months, have you done any of the following: … signed a petition?
9.	B21 During the last 12 months, have you done any of the following: … boycotted certain products?
10.	C4 Compared to other people of your age, how often would you say you take part in social activities?
11.	Section H L … how much each person is or is not like you: It’s very important to him to help the people around him. He wants to care for their well-being.
	Workplace domain
12.	F27 … how much the management at your work allows/allowed you … … to decide how your own daily work is/was organized?
13.	F28 … how much the management at your work allows/allowed you … … to influence policy decisions about the activities of the organisation?
14.	F39 Are you or have you ever been a member of a trade union or similar organisation?
	Democratic values domain
15.	B34 … to what extent you agree or disagree with each of the following statement: Gay men and lesbians should be free to live their own life as they wish.
16.	B43 Is [country] made a worse or a better place to live by people coming to live here from other countries?
17.	Section H C He thinks it is important that every person in the world should be treated equally. He believes everyone should have equal opportunities in life.
18.	Section H H It is important to him to listen to people who are different from him. Even when he disagrees with them, he still wants to understand them.
19.	Section H S … how much each person is or is not like you. He strongly believes that people should care for nature. Looking after the environment is important to him.

Source: Authors’ selection of the ESS Round 9 questionnaire.

in Appendix A shows the 19 items that are included in the four domains comprising the active citizenship scale.

Due to the fact that the 19 selected variables were constructed using different scale formats, they had to be standardised prior to their aggregation into a composite score. In our case, we standardised the initial variables by converting them into z-scores. This approach converts all indicators to a common scale with an average of zero and standard deviation of one. Before summing the variables into the composite active citizenship scale and its domain subindices, a series of reliability tests were conducted in order to find out whether the selected variables fit into coherent scales. The internal consistency coefficients’ Cronbach Alpha for the four domain subindices were acceptable: for the political domain, α = 0.49; for the social domain, α = 0.533; for the workplace domain, α = 0.51; and for the democratic values domain, α = 551. According to Nunnally (1967), values of the Cronbach’s Alpha coefficients as low as 0.50 are appropriate for exploratory research such as ours. The internal consistency of the composite scale active citizenship, which we used in the analysis, was α = 0.722. The McDonald omega of this scale was 0.724. Values of these coefficients of 0.7 indicate that the composite scale can be considered a measure with an acceptable internal consistency (George and Mallery 2003; Stensen and Lydersen 2022).

The subindices and the composite scale were formed by summing the corresponding variables. Correlations between the active citizenship scale and its components are shown in

Table 1. Bivariate Pearson correlations between the active citizenship scale and its components.

	Active Citizenship	Political Domain	Social Domain	Workplace Domain	Democratic Values Domain
Active citizenship	1	0.710	0.826	0.566	0.662
Political domain		1	0.512	0.239	0.191
Social domain			1	0.306	0.387
Workplace domain				1	0.241
Democratic values domain					1

. The correlations have been calculated on an individual level across the whole sample. The composite scale is most strongly correlated with its social domain and its political domain. These two dimensions contribute the most to its overall content. The workplace domain makes the smallest contribution.

Our main independent variable reflects subjective assessments of the fairness of educational opportunities. It is linked to the following question: “To what extent do you think this statement applies to you? ‘Compared to other people in [country], I have had a fair chance of achieving the level of education I was seeking’”. Following Liebig (2016), we have chosen the individual-level item (the reflexive one) from the ESS questionnaire which measures the equality of opportunity in relation to education. Respondents rated this statement on an 11-point Likert scale, with 0 indicating “does not apply at all” and 10 meaning “applies completely”. We have reversed the coding of the answer.

We also included five individual-level control variables: the highest level of education measured with the International Standard Classification of Education (ISCED) classification 2011 scheme (five categories, ref. category: ISCED 0–1), the highest parents’ level of education as an indicator of social background (five categories, ref. category: ISCED 0–1), age (continuous), gender (0 = female; 1 = male), and working in the last 7 days (0 = no, 1 = yes).

In order to study the social embeddedness of the relationship between subjective assessments of the fairness of educational opportunities and active citizenship, we selected some important indicators of political, economic, and cultural contexts. More concretely, we included the following independent variables at the country level:

• Democracy index. The Economist Intelligence Unit (2019) report presents the democracy index results for 2018. This index ranges from 0 to 10 and is composed of 60 indicators grouped into five categories: electoral process and pluralism; civil liberties; functioning of government; political participation; and political culture. Among the 29 countries in our analysis, the highest score on the democracy index is Norway’s at 9.87, and the lowest is in Montenegro at 5.74. We have used the democracy index as an indicator of the political context.
• GDP per capita. We have used this as indicator of a country’s economic development. More specifically, we have used the GDP per capita, PPP (current international USD) as of 2018 (extracted from (The UNESCO Institute for Statistics (UIS) n.d.). Although we are aware that this has some limitations (e.g., Brinkman and Brinkman 2011; Dědeček and Dudzich 2022), keeping in mind that there is no commonly accepted index of economic development, for the purposes of our analysis, we use it as a proxy for economic development and control its influence with income inequalities, measured using the Gini index. The highest GDP is in Ireland at 84,556, and the lowest is in Serbia at 17,715.
• Generalised trust. We have used this crucial cultural value as an indicator of the cultural context. Generalised trust was measured at the country level as the average value of those who answered the question “[G]enerally speaking, would you say that most people can be trusted, or that you can’t be too careful in dealing with people? Please tell me on a score of 0 to 10, where 0 means you can’t be too careful and 10 means that most people can be trusted” in the European Social Survey 2018 (ESS Round 9: European Social Survey Round 9 Data 2018). The lowest mean value for this indicator is in Bulgaria at 3.507, and the highest is in Denmark at 7.063. When doing the calculations, we weighted the data by design weight (dweight).

We also included the Gini coefficient of equivalised disposable income as of 2018 as a country-level control variable. We used it as a measure of income inequalities. It ranges from 0 to 100, and the higher the index is, the higher income inequalities in a given country are (extracted from Eurostat on 23 November 2023, data code: ilc_di12). The highest Gini coefficient is in Bulgaria (39.6) and the lowest in Slovakia (20.9).

To facilitate the interpretation of interaction effects and provide comparability of results, all country-level variables were standardised and entered into our analysis, being mean-centred and having a standard deviation of one.

Table 2. Descriptive statistics.

Variables	Mean	SD	Min	Max
Dependent variable
Scale of active citizenship	0.012	0.426	−1.766	2.097
Independent variable at individual level
Unfairness of educational opportunities	3.430	2.350	1	10
Control variables at individual level
Highest level of education
ISCED 0–1	0.025	0.157	0	1
ISCED 2	0.112	0.316	0	1
ISCED 3	0.413	0.492	0	1
ISCED 4	0.067	0.251	0	1
ISCED 5–8	0.382	0.486	0	1
Parents’ highest level of education
ISCED 0–1	0.178	0.383	0	1
ISCED 2	0.189	0.392	0	1
ISCED 3	0.355	0.478	0	1
ISCED 4	0.054	0.226	0	1
ISCED 5–8	0.224	0.417	0	1
Age	46.643	11.536	25	65
Gender (Male)	0.472	0.499	0	1
Working in the last 7 days (Yes)	0.749	0.434	0	1
Independent variables at country level
GDP per capita PPP	0	1	−1.708	2.559
Democracy index	0	1	−2.104	1.815
Generalised trust	0	1	−1.478	1.910
Gini coefficient	0	1	−1.906	2.237

Abbreviation: GDP per capita in PPP = gross domestic product (GDP) at purchasing power parity (PPP).

provides descriptive statistics on all the variables used in this study after a listwise deletion of individual-level independent and control variables.

3.3. Strategy of Analysis

The analysis uses data for 29 countries. As such, it has a nested structure with individuals (level 1) nested in countries (level 2). Nested data imply that individual observations are not independent of each other, and this makes ordinary least squares regression techniques inadvisable. To take into account this problem and to answer our research questions, this study uses multilevel linear regression models (see Rabe-Hesketh and Skrondal 2012). These models were chosen because they allow for the estimation of models with data at different levels and their cross-level interactions. The analyses were conducted in Stata 16 using the xtreg command. As a first step, we estimated a null model representing a baseline model with only the intercept. The intraclass correlation (ICC) in the empty model (Model 0) is 0.259. This shows that 25.9% of the variation in active citizenship is due to differences between the countries where people live. Also, this ICC value is higher than 0.05, which is sufficient for applying a multilevel modelling approach (see Hox 1998). Furthermore, the number of groups in our analysis is above the minimum number of 25 groups that is required for applying multilevel linear models (Bryan and Jenkins 2016).

Model 1 includes unfairness of educational opportunities. In Model 2, all the control variables are included. In Models 3, 5, and 7, we added the country-level characteristics one by one, whereas Models 4, 6, and 8 include cross-level interaction terms separately between each of the three aspects of the opportunity structures, respectively: generalised trust, GDP per capita, and democracy index and unfairness of educational opportunities.

4. Results

Model 1 in

Table 3. Results of multilevel linear models, regression coefficients, and standard errors in parentheses.

	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6	Model 7	Model 8
Unfair educational opportunities	−0.036 *** (0.001)	−0.020 *** (0.001)	−0.020 *** (0.001)	−0.019 *** (0.001)	−0.020 *** (0.001)	−0.020 *** (0.001)	−0.020 *** (0.001)	−0.019 *** (0.001)
Highest level of education Ref., ISCED 0-1
ISCED 2		0.083 *** (0.015)	0.083 *** (0.015)	0.082 *** (0.015)	0.083 *** (0.015)	0.083 *** (0.015)	0.083 *** (0.015)	0.085 *** (0.015)
ISCED 3		0.147 *** (0.015)	0.147 *** (0.015)	0.146 *** (0.015)	0.147 *** (0.015)	0.147 *** (0.015)	0.147 *** (0.015)	0.149 *** (0.015)
ISCED 4		0.213 *** (0.016)	0.213 *** (0.016)	0.212 *** (0.016)	0.213 *** (0.016)	0.213 *** (0.016)	0.213 *** (0.016)	0.215 *** (0.016)
ISCED 5–8		0.310 *** (0.015)	0.310 *** (0.015)	0.310 *** (0.015)	0.310 *** (0.015)	0.311 *** (0.015)	0.310 *** (0.015)	0.313 *** (0.015)
Parents’ highest level of education Ref., ISCED 0-1
ISCED 2		0.014 (0.008)	0.014 (0.008)	0.014 (0.008)	0.014 (0.008)	0.014 (0.008)	0.014 (0.008)	0.015 * (0.008)
ISCED 3		0.040 *** (0.007)	0.040 *** (0.007)	0.040 *** (0.007)	0.040 *** (0.007)	0.040 *** (0.007)	0.040 *** (0.007)	0.040 *** (0.007)
ISCED 4		0.064 *** (0.011)	0.064 *** (0.011)	0.064 *** (0.011)	0.064 *** (0.011)	0.064 *** (0.011)	0.064 *** (0.011)	0.064 *** (0.011)
ISCED 5–8		0.110 *** (0.008)	0.109 *** (0.008)	0.109 *** (0.008)	0.110 *** (0.008)	0.110 *** (0.008)	0.109 *** (0.008)	0.110 *** (0.008)
Age		0.003 *** (0.0002)	0.003 *** (0.0002)	0.003 *** (0.0002)	0.003 *** (0.0002)	0.003 *** (0.0002)	0.003 *** (0.0002)	0.003 *** (0.0002)
Gender, Ref. Female
Male		−0.020 *** (0.004)	−0.020 *** (0.004)	−0.020 *** (0.004)	−0.020 *** (0.004)	−0.020 *** (0.004)	−0.020 *** (0.004)	−0.020 *** (0.004)
Working last 7 days, Ref. No
Yes		0.053 *** (0.005)	0.053 *** (0.005)	0.053 *** (0.005)	0.053 *** (0.005)	0.053 *** (0.005)	0.053 *** (0.005)	0.053 *** (0.005)
Country-level features and cross-level interactions
Gini coefficient			−0.022 (0.022)	−0.022 (0.023)	−0.027 (0.021)	−0.027 (0.021)	−0.023 (0.020)	−0.023 (0.020)
Generalised trust			0.107 *** (0.021)	0.097 *** (0.022)
Generalised trust × Unfair educational opportunities				0.003 ** (0.001)
GDP					0.106 *** (0.021)	0.100 *** (0.021)
GDP × Unfair educational opportunities						0.002 * (0.001)
Democracy index							0.123 *** (0.020)	0.108 *** (0.020)
Democracy index × Unfair educational opportunities								0.004 *** (0.001)
Constant	0.145 *** (0.033)	−0.334 *** (0.029)	−0.331 *** (0.027)	−0.330 *** (0.028)	−0.331 *** (0.027)	−0.332 *** (0.027)	−0.335 *** (0.027)	−0.337 *** (0.026)
ICC	0.193	0.101	0.088	0.094	0.080	0.080	0.078	0.075
Observations	27,472	27,472	27,472	27,472	27,472	27,472	27,472	27,472

Source: European Social Survey 2018, own calculations. Significance: * p < 0.05, ** p < 0.01, *** p < 0.001. N(countries) = 29.

indicates that a higher perceived unfairness of educational opportunities is associated with lower levels of active citizenship. This negative relationship persists, although its strength decreases (as indicated by the change in the regression coefficient from −0.036 to −0.020), even when control variables are added in the analysis (see Model 2). The analysis of the control variables shows that one’s highest level of education, parents’ highest level of education, age, and work status are positively associated with active citizenship, whereas male gender is negatively associated with active citizenship. This means that the higher one’s level of education, one’s parents’ level of education, and one’s age are, the more pronounced one’s active citizenship is. Those citizens who had worked in the last seven days were also more active as citizens than those who had not. At the same time, men are less active as citizens than women.

The estimates also show that all three country-level variables included in our analyses (generalised trust, GDP per capita, and democracy index) positively relate to active citizenship (Models 3, 5, and 7). This means that the higher the level of trust in a given country is and the more economically and democratically developed it is, the higher the level of active citizenship among the citizens who live in it is. The cross-level interactions which we have added in Models 4, 6, and 8 allowed us to estimate whether the association between perceived unfairness of educational opportunities and active citizenship varies with the extent to which people trust other random people in a country that is economically or democratically developed. More specifically, the analyses of the first interaction term show that although unfairness of educational opportunities is associated with lower levels of active citizenship, as the generalised trust in a given country becomes higher, the perceived unfairness of educational opportunities increases individual active citizenship (Model 4). The same trend can also be observed for GDP and the democratic index (Models 6 and 8). This illustrates that these three aspects of the country context in which people live (political, economic, and cultural) have a moderating role in the relationship between active citizenship and the perceived unfairness of educational opportunities. In other words, despite the relationship between the perceived unfairness of educational opportunities and active citizenship being negative, it is mitigated when people are living in high-trust societies and in countries which are more economically and democratically developed.

5. Discussion and Concluding Remarks

Zgaga (2009, p. 185) argues that “[t]he ‘citizenship function’ is not limited to the curriculum and cannot be exhausted in citizenship education alone”. The present article follows this line of reasoning. It looks not inside educational institutions but at their entry and focuses on individuals’ subjective assessments of the fairness of their opportunities to achieve a desired level of education and their effects on active citizenship.

Although we use ‘effects’, ‘influences’, and ‘contributions’ as overlapping terms for the analysis of education’s role in active citizenship, we agree that these emphasise different aspects and accept ‘contributions’ as the most relevant term that “suggests more a process than an end” (Marginson et al. 2023, p. 4). Marginson (ibid., italics in the original) refers to the Cambridge Dictionary, which “defines a contribution as ‘something that you contribute or do to help produce or achieve something together with other people’”. This definition is very important, because it outlines that the activities of education “are often relational (‘together with other people’)”. Previous research has provided the greatest support for the hypothesis that education affects active citizenship through developing relevant knowledge and skills (e.g., Hillygus 2005) through individualised contributions, for example. Our article demonstrates that active citizenship could also be regarded as a collectivised contribution of education “emerging only through social relations” (Marginson et al. 2023, p. 4, italics in the original). Assessments of the fairness of educational opportunities are always relational, as people compare their chances to achieve their desired level of education with the chances of other people in given social circumstances.

The present article makes several contributions to the existing research. Theoretically, it offers an innovative approach for conceptualising the relationship between education and active citizenship. By investigating how the subjective assessment of people’s opportunities to obtain the level of education that they desire is associated with their active citizenship, our study broadens the very understanding of the ways and mechanisms through which education contributes to active citizenship. The article also reveals the heuristic potential of the capability approach for understanding the importance of educational opportunities as a separate factor influencing people’s activities. The article argues that active citizenship is a multidimensional and domain-specific phenomenon and suggests an understanding of its dimensions and indicators. Lastly, the analyses demonstrate the social embeddedness of the association between subjective assessments of the fairness of people’s opportunities to achieve a desired level of education and their active citizenship.

Our findings are in line with previous research which reveals a positive association between individual educational attainment and active citizenship—more concretely, that the higher people’s level of education is, the more pronounced their active citizenship is (Campbell 2006; Hoskins et al. 2008; Kam and Palmer 2011; Eurostat 2017). However, our analysis goes further and broadens the understanding of how education affects active citizenship. According to Hillygus (2005), education positively influences active citizenship mainly through its role in developing individuals’ skills and knowledge about the democratic society. We have shown that the subjective assessment of people’s opportunities to achieve the level of education that they strive for is a separate mechanism through which an education system may contribute to active citizenship. Feelings of unfairness have a discouraging influence on individuals’ civic activities. We suggest that this result is an expression of the high value that people place on education and equity in contemporary societies. Simply because people regard receiving a given level of education as a valuable aim, the subjective assessment of the fairness of available educational opportunities becomes a significant factor that shapes their active citizenship.

Our study comes with some limitations. Overall, it has been a real challenge to conceptualise and find sufficient reliable indicators for measuring active citizenship in different social domains. For calculating the scale of active citizenship, therefore, we used available data from the ESS 2018, which does not have a theoretically based module on active citizenship and includes only a small set of the questions from the original citizenship ESS 2002 survey. Thus, if other more recent data are available, e.g., from a survey specially focused on citizenship, the indicators and the dimensions might be enriched. We also acknowledge that the ESS data are cross-sectional and cannot be a basis for revealing causality, only associations. Our analysis refers to the situation before the recent crises, caused by the COVID-19 pandemic and the armed conflicts in Ukraine and the Middle East, which have deeply and negatively influenced all spheres of our contemporary societies.

Several directions for further research can be outlined. The first one relates to the use of a dynamic perspective and investigating how the association between subjective assessments regarding the fairness of people’s educational opportunities and active citizenship changes over time. In this regard, it is especially important to analyse how some crisis events—for example, the COVID-19 pandemic, the Russia–Ukraine war, and the Israel–Hamas war—have affected the assessment of the fairness of educational opportunities, active citizenship, and their relationship. The present study provides a basis for comparison and could serve as a reference point for studying the dynamics of the association between the subjective assessment of educational opportunities and active citizenship and for assessing the effects of recent crisis events on people’s life. Secondly, interesting findings could be expected if this association was analysed separately for different domains of active citizenship. Thirdly, it is worth studying how other macro factors (e.g., the level of inclusive economic growth or the level of individualism/collectivism) moderate the relationship between subjective fairness assessments about people’s opportunities to achieve a desired level of education and active citizenship. Fourthly, special attention needs to be paid to the question of how to consider the simultaneous effects on active citizenship of both the internal (e.g., curricula, quality of education, students’ involvement in university governance) and external (educational opportunities) characteristics of educational institutions. Fifth, although we have used quantitative data, we believe that there is a need to apply a mixed-method design and to hear people’s voices in order to better understand active citizenship. A recent qualitative study (Visser et al. 2023) shows that feelings of entitlement and a taste for politics are crucial for understanding (non)participation of less-educated citizens in social activities. Sixth, the relationship between the fairness of educational opportunities and active citizenship can be explored with the use of objective measures of fairness (e.g., see Brunori et al. 2012; Boyadjieva and Ilieva-Trichkova 2020). Lastly, all future studies of active citizenship should take into account its recent forms, which are related to the development of digital technologies and social media.

The present study has clear practical and policy implications, as the newly designed scale can be used for evaluating the effectiveness of education policies across Europe and the role of education for stimulating such an important characteristic of democratic societies as active citizenship. It shows that the education policies in relation to active citizenship should not be limited to initiatives and measures regarding curriculum and school activities. Very important in this respect is the access to different educational institutions and whether policies are in place which ensure fairness of educational opportunities for all. Our findings also demonstrate that the embeddedness of education in the wider social environment and its relations with other social spheres should always be taken into account when assessing its contributions to individual and social wellbeing, as well as in formulating policies for its future development.