Keywords

1 Introduction

Within Europe, numerous political, economic, social, and cultural changes brought about by globalization and Europeanization have challenged and transformed young people’s sense of citizenship and identity. An important aspect of good citizenship is attitudes of tolerance towards others (Almond and Verba 1963; Sherrod and Lauckhardt 2009), which includes positive attitudes towards the equal rights of others (Green et al. 2006). Yet, in recent times, we have witnessed rising levels of intolerant and xenophobic attitudes, due, in part, to negative perceptions over increasing flows of migration, immigration, refugees, and asylum seekers (Green et al. 2006).

According to Heater (1999), the feeling of citizenship points to the fact that how a person behaves in the political sphere is related to the ideas they have about “being a citizen.” This feeling can, according to Heater, be a result of the person’s identification with specific levels of political organization and, in a second vein, with their idea of civic virtue regarding the concept of a “good citizen.” It is “citizenship” in the latter sense that we are interested in. Five profiles of citizenship norms (i.e., comprehensive, socially engaged, duty-based, monitorial, and anomic) were developed by Torres Irribarra and Carrasco (see Chap. 3) using latent class analysis on the International Association for the Evaluation of Educational Achievement (IEA) International Civic and Citizenship Education Study (ICCS) 2016 data. Drawing on their work, in the present chapter we used multiple group multinomial logistic regression models to investigate how these different profiles relate to European adolescents’ political tolerance, i.e., their attitudes towards equal rights for others (immigrants and women). We used the ICCS 2016 data from 51,040 grade 8 students (aged 14) across 14 European countries.

This chapter is divided into five sections. Following this introduction, in the second section we provide a brief overview of the study’s conceptual background, noting in particular the link between citizenship and tolerance. In section three, we describe the study’s method in terms of the variables, the data used, and how the data was analysed. We present the results of the analysis in section four, showing interesting patterns regarding the association between different citizenship norms profiles and young people’s level of tolerance (i.e., support for the equal rights of others). We end with discussion of the study’s key conclusions in section five, noting theoretical and policy implications.

2 Conceptual Background

Within Europe, globalization and Europeanization have resulted in increasingly multicultural and ethnically diverse societies and challenged the concept of citizenship. National borders are increasingly blurred (Brodie 2004; O’Sullivan and Pashby 2008; Reid et al. 2010; Schattle 2012; Torres 2002) and citizenship has taken on new meanings beyond the nation state to include European and global dimensions. This, in turn, has challenged how young Europeans see themselves as citizens, particularly in terms of how they feel about their identities and their ideas about their roles and behaviors at the local, national, European, and global levels.

In addition to how they see themselves, how young Europeans see “others” is also being challenged. Europe has witnessed a large rise in the number of refugees and migrants from the Middle East and Africa attempting to enter the European Union. European surveys have highlighted increasingly negative public attitudes and insecurity over this perceived immigration “crisis,” partly due to the rise in populist, far-right parties displaying anti-immigration rhetoric (Mylonas 2012). This public anxiety has resulted in the “othering” of minorities, a rise in racist attacks, and urban unrest that has been attributed to the cultural difference and deviancy of young migrants (Schierup and Ålund 2011, p. 56). It has also contributed to anti-European integration sentiment that, among other issues, resulted in Brexit.

The societal changes outlined have been perceived of as posing a challenge to existing value systems, making it difficult to accept new citizens, thereby complicating their integration process (Freitag and Rapp 2015). In terms of the inclusion of newcomers, some countries have chosen to implement restrictive citizenship legislation, while others have taken a more liberal approach (Midtbøen 2015). In addition, there has been a focus on citizenship initiatives that promote integration, social cohesion, and values such as tolerance and respect for cultural diversity, inclusion, equal rights, and human rights (Sampermans et al. 2017). However, efforts towards promoting these at the national level have varied considerably. For example, in England, the focus has been on learning about Fundamental British Values (FBVs) in an effort to prevent radicalization and extremism following the rise in terrorist attacks. This initiative has been criticized for potentially alienating and radicalizing students who do not see themselves as truly British (Bolloten and Richardson 2015).

It is clear that an important aspect of good citizenship is the attitude of tolerance towards others (Almond and Verba 1963; Sherrod and Lauckhardt 2009). Tolerance is a multidimensional concept, which includes a wide range of attitudes towards different groups that may take various forms (Green et al. 2006; Isac et al. 2018). For example, political tolerance refers to giving different groups in society democratic and political rights, whereas social tolerance is related to contact with “others” (e.g., inter-ethnic friendships) (Isac et al. 2018). According to Gibson (2007), tolerance can be defined as: “the willingness to put up with disagreeable ideas and groups in order to peacefully coexist.” Without tolerance, not all groups in society would be able to defend their interests in the same way, which in turn could challenge the existence of a true democracy. Moreover, without tolerance, inter-group conflict would become practically inevitable, thus heightening even more the importance of tolerance as a fundamental democratic attitude (Hanh 1998). By definition, tolerance contains an internal paradox of allowing the ideas or interests one disagrees with, dislikes, or abhors, while also giving equal rights to people or groups regardless of whether you agree with their opinion or behaviour (Sullivan and Transue 1999; Sullivan et al. 1981; Vogt 1997). For example, as a citizen you can disagree with an idea because of political, religious, cultural, or social beliefs. However, you still allow people to express this idea and endorse their fundamental equal rights.

In light of this background and given the importance of tolerance for good citizenship and well-functioning democracies (Council of Europe 2010), we operationalize tolerance in terms of positive attitudes towards the equal rights of (a) women and (b) immigrants. In the next section, we explain the methods used to determine how different profiles of citizenship norms amongst European adolescents relate to attitudes towards these two groups.

3 Method

This section describes the methods and data used for the analysis presented in this chapter. Because of our focus on Europe, we used only the data from the 14 European countries that participated in ICCS 2016 (Bulgaria, Croatia, Estonia, Latvia, Lithuania, Slovenia, Denmark, Finland, Norway, Sweden, Italy, Malta, Belgium (Flemish), the Netherlands) (Schulz et al. 2018a). The sample is representative of the population of grade 8 students (average age 14) in each country and included a total of 51,040 students.

We used multiple group multinomial logistic regression models to determine how different profiles of citizenship norms relate to European adolescents’ attitudes towards equal rights for immigrants and women. The dependent variable was the student profiles of “good citizenship” developed in Chap. 3. Using latent class analysis, Torres Irribara and Carrasco (the present volume) used ICCS 2016 data to look at how students endorse particular citizenship norms (see Chap. 3 for a full list of the different norms) classifying them into five types of citizenship norms profile: (a) comprehensive, (b) socially engaged, (c) duty-based, (d) monitorial, and (e) anomic.

Students classed within the comprehensive citizenship profile (who were in fact the largest group) are those who value different forms of civic engagement (Ekman and Amnå 2012) including manifest forms of participation, such as voting, extra parliamentary actions such as peaceful protests, and social involvement such as helping in the local community. Students classed as socially engaged consider it important to protect the environment, protect human rights, and participate in activities that benefit the local community. They also highly value obedience to the law and respect for government representatives although they consider it less important to participate in political discussions or join a political party. Those who fall within the duty-based profile find it important to obey the law, work hard, and respect government authorities. However, they score lower on other characteristics such as participation in non-institutionalized forms of political participation such as activities to protect the environment, protect human rights, and benefit people in the local community. They also consider it less important to join a political party, participate in political discussions, and engage in peaceful protests. The monitorial group represent a mix of valuing non-conventional forms of political participation, while disregarding engaging in political parties. They value participation in elections as well as non-institutionalized forms of political participation, such as peaceful protest against unjust laws, participation in activities that benefit the local community, promotion of human rights, and protection of the environment. The anomic group expressed the lowest endorsement to all the citizenship norms, i.e., they consistently had the lowest probability of answering “important” across all the items. The most valued items for the anomic group are obeying the law and working hard, although the rate of endorsement is too low to be considered typical of this group.

The main independent variables were students’ attitudes towards equal rights for immigrants and women. These variables correspond to scales created by replicating the procedure used by Isac et al. (2019). We created these scales instead of using those included in the ICCS 2016 dataset, because the modifications proposed by Isac et al. ensure cross-cultural comparability of the resulting scores across the 14 countries included in the analysis. We used three indicators to construct the scale of attitudes towards equal rights for women. An example of such an indicator includes equality of opportunity in the labour market. We used four indicators to construct the scale of attitudes towards equal rights for immigrants, an example of which is equality of opportunity for political participation (see Table 1 for the full list of indicators).

Table 1 Indicators in the analysis of attitudes towards women and immigrants
Full size table

The fit indices of the scalar model largely comply with the model fit evaluation criteria (see Brown 2014; Wang and Wang 2012), both in terms of overall fit indices (RMSEA = 0.043; CFI = 0.985; TLI = 0.987), as well as relative fit indices (Metric vs Configural; ΔRMSEA = 0.010; ΔCFI = 0.005). The full procedures and results of the Multi Group Confirmatory Factor Analysis (MG-CFA) models used to create and test the invariance of the two scales, can be consulted in Isac et al. (2019).

Apart from attitudes towards equal rights for immigrants and women, other independent variables used in the analysis were student gender, home literacy resources, civic knowledge, and immigrant status (see Table 2 for the main characteristics of these variables). These variables were included as covariates in the analysis since they have been shown to be important predictors of the independent variable in previous studies (see, for example, Hooghe et al. 2015, 2016).

Table 2 Independent variables
Full size table

We report the odds (i.e., the relative probabilities) of belonging to each of the profiles with reference to each of the independent variables. In the present model, we use the largest group, i.e., the comprehensive citizen, as the reference category. The fitted model assesses the change in odds, for one unit of every covariate included in the model. In particular, we are interested in how students endorse citizenship norms, conditional to their attitudes towards equal rights for immigrants and for women. For example, the model tells us if the odds of being socially engaged are higher than being comprehensive, when students present higher levels of support for equal rights for women. No constraints between countries were added, so the relationship between the variables of interest are freely estimated.

All estimates include the complex sample design of ICCS 2016. Taylor Series Linearization was used for variance estimation, including school stratification and schools as the primary sampling unit (Stapleton 2013). Survey weights were scaled as up to 500 observations to ensure the equal contribution of each country to the results (Gonzalez 2012). Civic knowledge plausible values were included as imputed data, to account for its measurement error in all calculations (Rutkowski et al. 2010). Data preparation was carried out using the IEA IDB Analyzer (IEA 2017) and IBM SPSS (IBM 2015). All analyses were performed in Mplus 8.2 (Muthén and Muthén 2017). The full information maximum likelihood (FIML) method was implemented to handle missing data. Only 297 cases with missing data on all variables were not included in the analysis.

4 Results

The main focus of the present chapter is the relationship between students’ attitudes towards equal rights for women and for immigrants, and their associations with students’ citizenship norms endorsement. We present the estimates for the associations of these two variables for each country, while controlling for students’ gender, SES, civic knowledge, and immigrant status (see Tables 3, 4, and 5). Because of space constraints, in these three results tables we present the logits and odds for students’ attitudes towards equal rights for women and for immigrants’ only. The coefficients of the other predictors included in the model are available as an appendix (see Appendix C).

Table 3 Results of the multiple group multinomial logistic regression model: Central and Eastern countries
Full size table
Table 4 Results of the multiple group multinomial logistic regression model: Nordic and Southern countries
Full size table
Table 5 Results of the multiple group multinomial logistic regression model: Western countries
Full size table

The analyses of the data show some interesting general patterns regarding attitudes towards equal rights for immigrants and women. First, in the case of support for equal rights for women, the strongest associations between this variable and the different citizenship profiles were found in Sweden and the Netherlands. In both countries, the odds of being socially engaged were higher than the odds of being comprehensive. In Sweden this was 37%, and in the Netherlands 36%.

We now describe the results for each of the profiles with a focus on the relationship between students’ attitudes towards equal rights for women over and above the other covariates i.e., student gender, socioeconomic status, civic knowledge, and immigrant status (see Tables 3, 4, and 5).

The odds of students being classified as socially engaged (as compared to comprehensive) were significantly higher as their support for equal rights for women increased in half of the countries included in the analysis (i.e., Bulgaria, Lithuania, Finland, Norway, Sweden, Malta, and the Netherlands). This association was not statistically significant in the rest of the countries (Croatia, Estonia, Latvia, Slovenia, Denmark, Italy, and Belgium (Flemish)).

In terms of the comparison with comprehensive students, the odds of students being classified as duty-based were significantly higher as their support for equal rights for women increased in the Netherlands. Conversely, in Latvia, the odds of students being classified as duty-based were significantly higher as their support for equal rights for women decreased. In the rest of the countries, this relationship was not statistically significant.

Using the comprehensive students as a reference group, the odds of students being monitorial were significantly higher as their support for equal rights for women decreased in Croatia, Denmark, Norway, Italy, and Belgium (Flemish). On the contrary, in Bulgaria, students have higher odds of being monitorial when their support for equal rights for women is higher than that of comprehensive students. In the rest of the countries, this association is not statistically significant.

The odds of students being classified as anomic rather than comprehensive are significantly higher as their support for equal rights for women increased in Bulgaria. In Belgium (Flemish), by contrast, the odds of students being classified as anomic (as compared to comprehensive students) were significantly higher as their support for equal rights for women decreased. In the rest of the countries, this relationship was not statistically significant.

Second, in the case of support for equal rights for immigrants, the strongest and more consistent associations were found with the odds of belonging to the anomic class in Finland, Italy, and the Netherlands (66%, 73%, and 64%, respectively).

Another interesting pattern is that, when compared to comprehensive students, lower support for equal rights for immigrants is associated with higher odds of belonging to almost all the other classes in the vast majority of countries. The only exceptions are: socially-engaged students in Lithuania, duty-based students in Bulgaria, and monitorial students in Denmark, where support for equal rights for immigrants does not establish a statistically significant association.

There is also a pattern indicating that the probability of belonging to the monitorial and anomic groups has a stronger (i.e., negative) association with the support for equal rights for immigrants. Support for equal rights for women, tends to establish a weaker pattern regarding the odds of belonging to the classes included in the analysis. Using comprehensive students as the reference group, it increases the odds of students being socially engaged in seven countries, decreases the odds of being monitorial in five countries, and makes a significant difference in belonging to the other two groups (duty-based and anomic) in two countries. However, for the last two groups, the direction of the association is mixed (positive in one country and negative in the other).

5 Discussion and Conclusion

Considering how young people perceive themselves interacting with democracy from a politically normative point of view—in Heater’s terminology, their “feeling” of what citizenship should/should not be like—this chapter uses the five citizenship norms profiles that are present amongst European young people (see Chap. 3). Most grade 8 students in Europe were categorized within the comprehensive citizenship profile, while socially-engaged, duty-based, monitorial, and anomic groups were also identified. In studying the relationship of these citizenship norms with the tolerance concept that was operationalized by levels of support for equal rights for women and immigrants, two clear patterns emerged.

First, regarding the support for equal rights for women, we established that by using comprehensive students as the reference group, the odds of students being socially engaged increase in seven countries, the odds of being monitorial decrease in five countries, and makes a significant difference in belonging to the other two groups (duty-based and anomic) in two countries. However, for the last two groups, the direction of the association is mixed (positive in one country and negative in the other). We can hence conclude that the relationship between the five citizenship norms profiles and the first operationalization of tolerance (i.e., support for equal rights for women) is not consistent in Europe. However, the hypothesis that especially the socially engaged are the most open to equal rights for women seems to hold for at least seven countries.

Second, looking into equal rights for immigrants, it is clear that the comprehensive group deals well with the ambivalence present in the definition of tolerance. This group scores high on all the indicators of good citizenship and seems to be able to agree to disagree with others, and hence also to work with immigrant groups towards higher social cohesion in Europe. The other groups do not seem to accept the paradox of giving equal rights to people or groups regardless of whether you agree with their opinion or behaviour. The duty-based, monitorial, anomic, and even the socially-engaged groups, show significantly lower support for equal rights for immigrants than the students classified as comprehensive. This relationship was expected for groups such as the duty-based students, while their political activities are mostly related to democracy’s social order. This can imply a very narrow interpretation of rights of non-Europeans in Europe. Also, for the monitorial and anomic groups, this outcome comes as no surprise, as these two types of citizens seem to be more focused on the local, personal level (Westheimer and Kahne 2007) rather than opening themselves to a larger, globalized, more diverse world. As for students in the socially-engaged group, who theoretically are described as being concerned about social needs, human rights, and the environment (Dalton 2008; Barber and Ross 2018), we expected more young people to be part of this group (Dejaeghere and Hooghe 2009; Hooghe and Oser 2015; Reichert 2016) and we also hypothesized a higher inclination to support equal rights for immigrants. However, it seems that this group is also more focused on their local (maybe more homogenous) community, rights, and the environment, and that they seem to “hunker down” (Putnam 2007), making them engaged in their own group, but not inclined to be open to more equal rights for immigrants than the comprehensive group.

The results of the study have both theoretical and policy implications. In terms of the theoretical implications, the results develop understanding of the mechanisms underpinning the relationship between young European adolescents’ feelings about citizenship and their attitudes towards others (specifically immigrants and women). It is worth highlighting that our results indicate there is no single citizenship profile that endorses all the democratic values analyzed in this chapter. Rather, the endorsement of different democratic values is dependent on students’ individual views regarding how a good citizen is defined. In terms of policy implications, given the current European political climate characterized by increasingly negative perceptions of immigrants, our results suggest the need to develop targeted policy interventions for the promotion of tolerance and equal rights for others amongst young Europeans.