Youth in the Kaleidoscope : Civic Participation Types in Estonia and the Czech Republic 1

This paper presents an exploratory typological analysis of young people (aged 15–30) as political and civic actors in Estonia and the Czech Republic. We compare youth civic engagement patterns in these two East European countries, sharing similar socio-historical contexts, and analyse the sociodemographic and attitudinal profiles of the resulting participation types. The study draws on Estonian and Czech data sets collected from November to December 2016 within the Horizon 2020 project ‘CATCH-EyoU – Constructing AcTive CitizensHip with European Youth: Policies, Practices, Challenges and Solutions’. Two independent methods (latent class analysis and cluster analysis) demonstrated shared patterns in the political and civic activities employed by the Estonian and Czech participants, suggesting the existence of four clearly distinguishable types of young citizens. A more detailed analysis revealed that the socio-demographic and attitudinal profiles of active young people, and therefore, the factors of political socialization, differed quite substantially in the two countries.


Introduction
Forms of civic and political participation are shifting.The very concept of civic engagement involves highly different activities ranging from voting to participation in voluntary associations, and from political discussions to civil disobedience and many more (Berger 2009;Ekman and Amnå 2012).Such a wide variety of manifestations results in a lack of consensus on what constitutes civic engagement.
Sociální studia / Social Studies 2/2018.Pp. 11-33. ISSN 1214-813X. 1 The research reported in this paper was funded by the European Union, Horizon 2020 Programme, Constructing AcTiveCitizensHip with European Youth: Policies, Practices, Challenges and Solutions (www.catcheyou.eu)Grant Agreement No 649538, and by the institutional research funding (IUT20-38) from the Estonian Ministry of Education and Research.The authors would like to thank Katrin Kello, Ragne Kõuts-Klemm, Lucie Lomičová, Petr Macek, Peter Noack, Signe Opermann and Andu Rämmer for the data which were collected for use in this paper.The views and opinions expressed in this publication are the sole responsibility of the authors and do not necessarily reflect the views of the European Commission.
Despite the lack of agreement on definitions, many authors are concerned about the declining civic and political participation rates among young people (Banaji and Buckingham 2013;Sloam 2014).Banaji (2016) suggests that the near moral panic about the democratic deficit among young people is the result of giving too much credit to voting rates and forgetting about the other options for civic and political participation.Some scholars claim that although young people are less engaged in politics than older age groups, young people are more likely to become involved and engage themselves outside institutionalised politics (Stolle and Hooghe 2011).
Even in reaction to that, different typologies of civic engagement and civic passivity exist, considering involvement outside institutionalised politics to a very different extent.Mostly, the existing instances of problematising young people's civic and political participation and the typologies of engagement are based on empirical observations in established Western democracies, while the conceptualization of youth participation in post-socialist Central and Eastern European countries is less common.The initial research on political activism and civic participation in the new liberal democracies in CEE countries, mushrooming in the 2000s (e.g.Howard 2002;Rose and Munro 2003; see Kalmus et al. 2018, for an overview), has not been followed by a similar wave of studies on new trends and developments.In particular, the levels and patterns of the political and civic engagement of the new generation, socialized in the political context of the enlarged EU and in the digital era, deserved more attention from researchers.
To contribute to filling this gap, this article focuses on the participation patterns of young people in Estonia and the Czech Republic.Due to the scarcity of previous cross-national research on youth participation in CEE countries, this study employs an exploratory strategy with a comparative aim.The two countries selected for this analysis, Estonia and the Czech Republic, are similar in some principal aspects of the political and economic context (both rather small post-socialist countries, accessed the EU in 2004, and since then, have faced no major political or economic turmoil), while differing in some relevant macro-level characteristics (e.g.ethnic composition and e-state developments).We expect the macro-level similarities and differences to be reflected in youth civic engagement patterns.
This paper aims to explore and compare types of young people (aged 15-30) as political and civic actors in Estonia and the Czech Republic.Furthermore, the paper analyses the socio-demographic background of more active young people versus passive or alienated ones, and explores the relationships between participation types and political attitudes.
First, we present an overview of relevant typologies of youth civic engagement.Second, we compare selected aspects of the political, economic, technological and cultural-historical contexts of civic engagement in Estonia and the Czech Republic to set the research questions and formulate hypothetical assumptions for our exploratory analysis.After introducing the methods of data collection and analysis, we present our typology, based on two different methods of clustering, and describe the socio-demographic and attitudinal characteristics of the types in both countries.The discussion concentrates on interpreting the similarities and differences in the patterns of youth civic participation in the respective national contexts, and on considering the factors contributing to higher levels of the overall political and civic engagement in the two countries.

Typologies of youth civic engagement
One way to classify civic and political participation is to identify active citizenship positionings (types of activists or politically passive citizens).Banaji's (2016) typology, identifying six active citizenship positionings of young people, is an excellent example of this.Banaji suggests that young people can be either (1) almost entirely disenfranchised/excluded, (2) generally apathetic/inactive/passive/disengaged, (3) generally active in conformist ways, (4) generally active in anti-democratic and authoritarian ways, (5) generally active in prodemocratic, anti-authoritarian and non-conformist ways, or (6) always active in prodemocratic, anti-authoritarian and non-conformist ways.Several important issues concerning Banaji's typology (2016) are worth emphasising.First, she makes the distinction between conformist and non-conformist participation activities, and states that neither of these is exclusively reserved for one or two types of civic and political participation.Young people who are generally active in conformist ways can occasionally be questioning or critical on a particular issue, whereas young people who are always active in pro-democratic, antiauthoritarian and non-conformist ways often have to take part in actions that seem extremely conventional and dull (such as logging media events, attending multiple public meetings, or doing extensive community work).Second, it is important to recognize that not all youth political and civic engagement is necessarily pro-democratic and anti-authoritarian.Banaji notes that young people that are generally active in anti-democratic and authoritarian ways, question the democratic rights and the value of tolerance and equality.Finally, Banaji distinguishes between different types of civic and political apathy among young peoplesome young people are entirely excluded all the time, whereas others are generally passive but may occasionally participate by making a minimal effort (e.g. by casting a vote).
Banaji ( 2016) is, of course, not the first to emphasise that different forms of civic and political passivity may exist.Ekman and Amnå (2012), for example, have distinguished between passive non-engagement (referring to citizens who are not interested in politics and do not follow political and civic affairs) and active non-participation (referring to citizens who feel disgusted with political issues and who actively avoid political discussions).Later, Amnå and Ekman (2013) suggested that at least three distinctive forms of "political passivity" should be considered: "standby citizens", unengaged, and disillusioned citizens."Standby citizens" appear to be passive but they actually keep themselves informed about politics and are prepared for political action if needed (Amnå and Ekman 2013;Amnå 2010).
To our knowledge, these typologies, which are based on observations in well-established democracies, have been not tested in new democracies.Therefore, this article aims to explore whether the active and passive citizenship types among the youth in two new democracies resemble the participation typologies of old democracies.

The macro-societal context of civic and political participation in Estonia and the Czech Republic
About two decades ago, several comparative studies characterised post-communist societies in Europe as a rather coherent group (see e.g.Howard 2002;Inglehart 2006).These studies concluded that despite having different religious and cultural heritage, these societies shared a common powerful experience of life under communist rule, and nearly the same length of such domination imprinted all societies in a similar way.Based on the argument that the institutional systems had not yet stabilised enough to allow a more distinctive categorisation, authors of the typologies of welfare state regimes (e.g.Hofäcker 2006) also classified post-socialist states as one group.Some other authors, however, claimed that post-socialist countries had already developed in different directions: Estonia, together with the other Baltic states, into a neo-liberal type, and the Czech Republic, as a Visegrad state, into an embedded neo-liberal type (Bohle and Greskovits 2007).
The latter differentiation in the chosen path of socio-political development has, at least partly, contributed to the fact that Estonian society is somewhat less egalitarian than the Czech counterpart: for instance, the GINI index is higher in Estonia (in 2015, 32.7 versus 25.9 in the Czech Republic; The World Bank 2018), and the same holds for the youth unemployment rate (in 2017, 13.9 % of 15-24 year-olds were unemployed in Estonia compared to 8.3 % in the Czech Republic; modelled ILO estimate, The World Bank 2018).While the levels of educational attainment are equally high in both countries (in 2015, 89.7 % and 89.8 % of the population aged 25+ had at least completed upper secondary in Estonia and the Czech Republic, respectively; UNESCO Institute for Statistics, The World Bank 2018), tertiary school enrolment is higher in Estonia (in 2015, 72.1 % versus 64.5 % in the Czech Republic; UNESCO Institute for Statistics, The World Bank 2018), suggesting that social polarization may be sharper in Estonia.
The argument about greater social differentiation in Estonia, compared to the Czech Republic, is supported by the fact that Estonia is ethnically divided: the titular group, ethnic Estonians, forms 69 % of the population, and Russians as the largest group among ethnic minorities form 25 % (Põder 2015).When the former Soviet Union disintegrated and the Estonian Republic was restored as an independent state, the socio-political status of the Russians receded overnight.Previous research has shown that Estonian Russians have faced a double challenge of coping and self-determination: both in terms of socioeconomic and cultural transition and the Estonian nation-state (Vihalemm and Kalmus 2008), which presumably has an effect on overall social cohesion as well as on political and civic engagement.
In particular, Uslaner and Brown (2005) have demonstrated that inequality is a strong determinant of generalised social trust, and that trust has an effect on communal participation.Using standardized data from over a dozen cross-national surveys of the world's rich democracies, Solt (2008) provides further empirical evidence that greater economic inequality and greater political inequality go hand in hand.His results suggest that in countries where income and wealth are more concentrated, power is more concentrated as well, leading the less affluent to be more likely to withdraw from discussing political matters and deciding that participation is not worth their effort.Therefore, the higher levels of inequality in Estonia should result in lower civic participation than in the Czech Republic.However, the destructive effect of higher levels of inequality in Estonia may be reduced by the high generalised social trust levels, as Estonia is one of the most trusting countries not only in Europe but also worldwide (Beilmann and Realo 2018), and high levels of trust is, according to Uslaner and Brown (2005), one pathway to participation.
A shared characteristic of Estonia and the Czech Republic stems from the fact that the collapse of communism was followed by diminishing participation in politics and low levels of organisational membership everywhere in Central and Eastern Europe (Howard 2002;Inglehart 2006).The relative weakness of the civil society was often described as a common feature of post-communist societies in Europe (ibid.).According to Sztompka (2004), post-communist civil societies suffered from similar kinds of post-totalitarian trauma.In a similar vein, Inglehart (2006: 67), revealed that "most central and East European publics rank substantially lower on survival/self-expression values -a syndrome of tolerance, trust, well-being, and emphasis on self-expression that is closely linked with democracy".However, he admitted that "large differences exist between the value systems of the historically Catholic or Protestant ex-communist societies of Central and Eastern Europe, and the historically Orthodox ex-communist societies" (Inglehart 2006: 67).The position of Estonia and the Czech Republic in Inglehart's value system is, therefore, not only determined by the impact of communist rule but also by the Protestant cultural heritage in Estonia and the Catholic cultural heritage in the Czech Republic, which may make both societies more responsive to the development of civil society.
From the historical perspective, the post-communist civil society in the Czech Republic was based on a relatively well-developed informal and dissident civil society that evolved in the period of late communism in the late 1970s and 1980s (Pospíšilová 2011).During the 1990s, Czech civil society swiftly shifted from informal activities to greater professionalization and gained some self-confidence (Pospíšil, Navrátil and Pejcal 2015).This progress made Czech civil society one of the most developed in post-communist Europe, but still less developed when compared to Western European countries (Celichowski 2008;Rakušanová 2005).In comparison, Estonian civil society institutions remained weak in the 1990s and the early 2000s when Lagerspetz (2001) was concerned that with a low number of active voluntary associations, Estonian civil society did not form a significant counterbalance to party politics.
The technological preconditions for the realization of the potential of digitally networked participation are rather good in both countries: the proportion of internet users among the general population is 88 % in Estonia and 78.8 % in the Czech Republic, while almost all young people (99 % in both countries) use the internet (Ait 2017;Czech Statistical Office 2018).According to the most recent country classification by the EU Kids Online network (Helsper et al. 2013), Estonia and the Czech Republic belong to the same cluster (with Bulgaria, Cyprus, Poland and Romania), characterised by children's moderate use of online opportunities, their high levels of experiencing online risks, and ineffective parental support.However, in the Czech Republic, the implementation of online media and e-government at the national and local level has been complicated by inconsistencies in policies and by other various shortcomings (Špaček 2015), while in Estonia digitization has been a government priority and one of the central symbols of the rapidly changing society leading to a widely held perception of the country as one of the leading e-states (Runnel, Pruulmann-Vengerfeldt and Reinsalu 2009).

Aims and research questions
This exploratory article takes a closer look at the participation patterns of young people in Estonia and the Czech Republic and the relationships between the participation types, socio-demographic characteristics and selected political attitudes.We set the following research questions: (1) How can young Estonians and Czechs be divided into types of political and civic participation?What are the main similarities and differences between the typological patterns of political and civic activities emerging in the two countries?(2) How can the types of political and civic participation be characterised in terms of socio-demographic characteristics and political attitudes?What are the main similarities and differences between the profiles of the types in the two countries?Based on the common macro-level characteristics of the two countries (mainly, the shared experience of life under communist rule, followed by post-totalitarian trauma; the approximate position in Inglehart's value system, determined by the comparable political history and cultural heritage, and equally high levels of educational attainment), we expect to find substantial concurrences in the overall structure of youth civic and political engagement in the two countries.By referring to several differences between Estonia and the Czech Republic (Estonian society being somewhat less egalitarian and ethnically more divided than Czech society; in comparison to the Czech Republic, digitization and e-state developments being politically high priority in Estonia), we assume that some variations in the profiling aspects of the participation types in the two countries exist.

Method
The study draws on data collected as part of an international project dealing with the development of active citizenship among European youth Constructing AcTive CitizensHip with European Youth: Policies, Practices, Challenges and Solutions -CATCH-EyoU.We employ a data set (N=2,419) collected concurrently -in November-December 2016in the Czech Republic (N=1,346) and Estonia (N=1,073).In both countries, we conducted fully comparable surveys with some questions (e.g.educational path) tailored for adolescents (aged 14-18) and young adults (ages 19-30).Therefore, we may treat the sample as consisting of four sub-samples (see sample characteristics in Table 1).Due to the country specifics, the sub-samples differ in terms of the sampling procedures applied.
• Czech Republic, older (N=814): Data were collected in five Czech regions (Prague, Pardubice, Vysočina, South Moravia, Moravian-Silesian Region) using computerassisted personal interviews (43 %) or computer aided web interviewing; that is, online questionnaires (57 %), and applying quota sampling for each region (residency, gender, age, economic activity).Sampling and data collection in this group was conducted by a professional research agency that employed its established network of interviewers and research contacts.• Czech Republic, younger (N=532): All participants were high school students.
Schools were randomly sampled, based on the official register of public and private schools from the Ministry of Education, Youth and Sports, in the five aforementioned regions.In these schools, all available year 11 and 12 classes were tested.
Overall, 54 % of participants attended academically-oriented schools, while 46 % attended vocationally-oriented schools.To avoid potential practical obstacles (e.g.no available room in classrooms equipped with computers), students completed written questionnaires in their classrooms.• Estonia, older (N=499): Respondents were recruited in different educational institutions (universities, colleges and vocational schools), army service units, and local youth organisations across Estonia.• Estonia, younger (N=574): Respondents were recruited in various locations across Estonia (the capital city of Tallinn, the cities of Tartu and Narva, and six smaller towns) in different educational institutions (primarily senior secondary schools but also vocational schools).The recruitment procedure was the same in both Estonian age groups.A member of the research team visited the lesson, lecture or meeting, introduced the study, and asked people to fill in the consent forms.After that, links to online questionnaires (in Estonian or Russian, according to their choice) were sent by email to people who agreed to participate in the study.

Participation
There were 18 indicators of civic and political participation in the questionnaire (Table 2).With the following question: "People can express their opinions regarding important local, ecological or political issues.They do so by participating in different activities.Have you done any of the following in the past 12 months?"-respondents were asked to indicate how often, if ever, they have done any of the listed civic activities.A five-point response scale included "no", "yes -rarely", "yes -sometimes", "yes -often", and "yes -very often".The 18 indicators of participation were designed to cover protest activities (e.g.signing petitions), volunteering and charity, online participation (e.g.sharing political content), illegal participation (e.g.political graffiti), and institutionalised participation (e.g.working for a political party or a candidate).

Other variables
The independent variables include the socio-demographic variables of gender, age, ethnic majority/minority status (only in Estonia and differentiated according to the preferred language of the survey -Estonian or Russian, respectively), mother's highest completed level of education, 2 school track (lower and higher) (in the case of high school students), 2 The respondents had to choose between seven options -(1) Didn't finish any school (less than the 9th grade), (2) Completed 9 years of schooling, (3) Completed 10 years of schooling, (4) Completed 12/13 years of schooling, (5) Bachelor's degree / before pre-diploma, (6) Master's degree (Diploma), ( 7) More than master's degree.
occupational status 3 and student status 4 (in the case of older young adults) and place of residence. 5Using the question: "Are you still in education?".

5
The respondents had to choose between five options -(1) a big city, (2) the suburbs or outskirts of a big city, (3) a town or small city, (4) a village, and (5) a farm home or home in the countryside.
The second set of independent variables are variables indicating the participants' trust (social and institutional trust), interest (political interest) and attitudes to politics and political institutions (internal political efficacy, alienation, authoritarianism, support for democracy).All of the variables (except in the case of political interest) indicating attitudes used a response scale ranging from "strongly disagree" (1) to "strongly agree" (5).
Social trust was measured by one item: "I feel that most people can be trusted".
Political interest was measured using 4 items: "How interested are you in politics?";"How interested are you in what is going on in society?"; "How interested are you in topics related to the European Union?" and "How interested are you in national politics?"(Czechs: α =.90; Estonians: α =.88). 6emocracy.Support for democracy was measured using one item "Democracy is the best system of government that I know".
Alienation was measured using 4 items.Two of them addressed the EU level of government, the other two the national level -at both levels, one item addressed personal and one item institutional alienation, where the higher the score the higher the level of alienation ("People like me do not have opportunities to influence the decisions of the European Union"; "It does not matter who wins the European elections, the interests of ordinary people do not matter"; "People like me do not have opportunities to influence the decisions of the national parliament"; "It does not matter who wins the national elections, the interests of ordinary people do not matter").The score was computed with averaged items (younger Czechs: α = .80;older Czechs: α = .87;younger Estonians: α = .87;older Estonians: α = .88).
Internal political efficacy was measured using 3 items: "If I really tried, I could manage to actively work in organisations trying to solve problems in society"; "If I really tried, I could manage to help to organise a political protest"; and "If I really tried, I could manage to take part in a demonstration in my hometown".The final score was computed using averaged items (younger Czechs: α = .74;older Czechs: α = .74;younger Estonians: α = .82;older Estonians: α = .84).
Authoritarianism was measured using two items: "Instead of needing civil rights and freedoms, our country needs one thing only: law and order"; "Obeying and respecting authority are the most important values that we should teach our children".The final score was computed using averaged items (younger Czechs: α = .49;older Czechs: α = .68;younger Estonians: α = .51;older Estonians: α = .60).

Analysis
To analyse youth participation patterns in the two countries, we employed a typological analysis, which is a strategy for descriptive (quantitative or qualitative) data analysis "whose goal is the development of a set of related but distinct categories within a phenomenon that discriminate across the phenomenon" (Ayres and Knafl 2012).We conducted the analysis separately for Estonia and the Czech Republic, and used two independent methods to crossvalidate our results.First, we conducted latent class analysis (LCA) in Mplus 7.4 (Muthén andMuthén 1998-2015), using a robust maximum likelihood estimator (MLR) and automatic starting values with random starts.The final number of classes was chosen from solutions involving two to six classes based on the Bayesian information criterion (BIC), relative entropy and likelihood-ratio tests (LR).A lower level of the BIC, a higher level of relative entropy and a significant difference from the solution with less classes were considered preferable.Second, we re-analysed our data using the two-step cluster procedure in IBM SPSS 22. Clustering was based on log-likelihood distances, which are appropriate for noncontinuous variables, and the number of clusters was specified based on the previous results of the LCA.
Both the LCA and cluster analysis were conducted on the 18 participation variables.Considering the response scale ranging from "no" to "yes -very often", we treated the participation variables as ordinal.We also dichotomized these variables before the analyses [0=no, 1=yes (at least rarely)] due to extremely low numbers of people in some response categories for some items (e.g.only four people in Estonia and four in the Czech Republic painted or stuck political messages or graffiti on walls "very often").Although the dichotomization means a somewhat rougher categorisation of participants (i.e.people who did an activity only once and several times fall into the same category), it was a necessary step to prevent problems with model estimation.Moreover, dichotomous measures of participation are commonly also used in other large European surveys, such as the European Social Survey or the European Values Study.
After the classes (clusters) representing different participation patterns were established, we used one-way ANOVAs with Bonferonni post-hoc tests to determine whether they differed in terms of their socio-demographic profiles and attitudes.

Results
Based on the BIC, relative entropy and LR tests, we selected the LCA solutions with four classes as the most adequate in both countries (Table 3).These solutions were characterised by the lowest levels of BIC, they fitted significantly better (p < .05)than the solutions with three classes, and they had higher levels of relative entropy than the solutions with more classes (although they were inferior to the solutions with fewer classes).In Estonia, one item had to be excluded from the analysis due to estimation problems ("Painted or stuck political messages or graffiti on walls").Table 4 shows probabilities of taking part in specific activities in the four classes.The patterns found using the LCA were replicated by the results from the cluster analysis, even though absolute numbers of class/cluster members and the clarity of the profiles slightly differed (Table 4).Note: Probabilities of taking part in the activity are reported for the LCA.Proportions of participants who took part in the activity are reported for the cluster analysis.
The two methods applied in the Estonian and Czech samples in this study identified small groups of generally active young citizens (group 1), larger groups of generally passive young citizens (group 4) and two intermediate subgroups of moderately active young citizens (groups 2 and 3) in both countries.In both Estonia and the Czech Republic, generally active young citizens (General activists) were more likely to participate in almost all activities than were young people from other groups, while Passive young citizens were unlikely to practice any of the activities.Moderately active young citizens were more clearly profiled in their choice of civic activities.Compared to General activists, they were less likely to practice "openly political" activities linked to offline protests or addressing the field of institutionalised politics, or illegal activities.
At the same time, the Estonian and Czech samples differed in terms of practices employed by the two groups of moderately active citizens.In the Czech Republic, the first subgroup of moderately active young citizens (group 2) was more focused on volunteering and charity (thus, we may call them Voluntary benefactors), while the second subgroup of moderately active young citizens (group 3) was more often engaged in online activities (thus, we may label this group Digital activists).
In Estonia, the profiles partly differed as one group of moderately active young citizens (group 2) practiced both voluntary/charity and online activities (for reasons explained below we call this group Positively engaged activists), whereas the other group of moderately active young citizens (group 3) was clearly focused solely on donating money with other activities being less likely than in the previous group (thus, we may call this group Benefactors).
We examine the groups further in terms of the detail of their socio-demographics and political attitudes (Tables 5 and 6), focusing from now on only on the LCA results.In Estonia, the smallest group of General activists -when compared to the other three Estonian groups -can be described as the youngest, dominated by males with lower economic and cultural capital, and tending to be recruited from the population in rural areas and small towns.The group comes with a higher portion of the ethnic minority (Estonian Russians) and is characterised by the highest inclination toward authoritarianism, the lowest support for democracy (though they do not deprecate democracy), and the highest level of social trust.At the same time, General activists in Estonia are rather moderate in terms of political interest, internal efficacy, alienation or institutional trust.
The Czech General activists -in comparison to the remaining Czech groups -are also dominated by young males but they tend to come from the urban population.On average, they express the highest interest in politics and internal efficacy, the lowest preference for authoritarian attitudes, and the lowest level of alienation in the Czech sample.
The two intermediate Estonian groups of moderately active citizens, unlike the group of General activists, primarily consist of women, and both can be characterised as rather urban, higher educated groups with a low proportion of the Russian minority.Both groups share the strongest support for democracy and the lowest support for authoritarianism in the Estonian sample.However, these two groups differ from each other in some political attitudes.Positively engaged activists express by far the highest levels of internal political efficacy and interest in politics, and the lowest alienation in the Estonian sample (which motivates our choice of the label for this group).By contrast, the group of Benefactors is more alienated from politics, and its members express significantly lower interest in politics and internal efficacy (the levels of these variables are comparable to those of the Estonian General activists).Institutional trust 3.17 3.24 3.13 2.94 5.898 <0.01 f a Significant differences between all groups, except General activists and Passive young citizens; between Benefactors and Positively engaged activists (p<0.05,Bonferroni post-hoc test).b Significant differences between all groups, except General activists and Positively engaged activists (p<0.05,Bonferroni post-hoc test).c Significant differences between all groups, except General activists (p<0.05,Bonferroni post-hoc test).d Significant differences between all groups, except Benefactors and Positively engaged activists (p<0.05,Bonferroni post-hoc test).e Significant differences between General activists and other groups (p<0.05,Bonferroni post-hoc test).f Significant differences between Passive young citizens and Benefactors, between Passive young citizens and Positively engaged activists (p<0.05,Bonferroni post-hoc test).b All groups were significantly different from each other, except General activists and Digital activists (p<0.05,Bonferroni post-hoc test).c All groups were significantly different from each other (p<0.05,Bonferroni post-hoc test).d Significant differences between General activists and other groups (p<0.05,Bonferroni post-hoc test).e Significant differences between General activists and Digital activists, and between Digital activists and Passive young citizens (p<0.05,Bonferroni post-hoc test).f Significant differences between Passive young citizens and Digital activists (p<0.05,Bonferroni post-hoc test).
The Czech groups of moderately active young citizens also differ from each other and from the other two Czech groups, though their distinction follows different paths.Voluntary benefactors might be described as rather detached from institutionalised politics: the group with the lowest mean age in the sample, with the highest portion of females and with the lowest proportion from the urban population comes with a relatively low interest in politics and internal efficacy (nevertheless, both interest and efficacy are still higher than in the group of Passive young citizens).The group of Digital activists, on the other hand, consists almost equally of both genders and comes with significantly greater interest in politics and internal efficacy, and, compared to the other groups, relatively low social trust.The levels of alienation, authoritarianism and institutional trust are moderate in both groups.
The Estonian group of Passive young citizens -demographically rather younger, with the highest proportion of ethnic Russians in the Estonian sample -is, unsurprisingly, characterised by the lowest interest in politics, internal efficacy and institutional trust, and the highest alienation.Similarly, the Czech group of Passive young citizens typically expresses the lowest interest in politics and internal efficacy in the Czech sample, although their level of alienation or institutional trust does not deviate substantially from the overall average.In contrast to the other Czech groups, Passive young citizens are demographically distinct mainly in terms of having the highest proportion of employed members (and the lowest proportion of students), and by the lowest proportion of young people whose mothers have a higher education.

Discussion
At the first sight, the analysis delivered an analytically comforting picture of the political and civic activities of young people in Estonia and the Czech Republic, demonstrating obvious shared patterns in the types of activities employed by the Estonian and Czech participants.In both countries, our analysis provided four clearly distinguishable clusters, suggesting the existence of four main types of young citizens in terms of their political and civic agency: a small group of generally active young citizens, a larger group of generally passive young citizens, and two intermediate subgroups of moderately active young citizens who are more clearly profiled in their preference of civic activity.We may say that moderately active young people prefer latent forms of participation (Ekman and Amnå 2012), and engage primarily in social as opposed to political methods.
However, the cross-national differences identified in the in-between zone, dividing generally passive young citizens from those few that are active in all measured aspects, seem to be crucial -these variations in particular suggest a distinct position of "new" activities, linked with the use of online media, in the respective countries.The clusters suggest that while in Estonia, online political activities are increasingly adopted along with the increasing intensity of all activities (from the most inactive Passive young citizens to the most active group of General activists), in the Czech Republic the adoption of networked participation leads to the articulation of a specific type that might, in comparison with the other cluster of moderately active young citizens (Voluntary benefactors), be understood as "digital citizens or activists" (Digital activists).In Estonia, such a more or less clear distinction between "digital" and "offline" moderately active young citizens does not appear, suggesting that, unlike in the Czech Republic, online activities are more firmly adopted in the repertoires of action among young Estonians.Nevertheless, as we have shown elsewhere (Kalmus et al. 2018), new opportunities for digitally mediated participation have not created completely new forms or patterns of political or civic engagement in Estonia.
This study embraces both conventional as well as new and, in this respect, "unconventional" forms of participation.However, we tend to agree with Ekman and Amnå (2012) that it is problematic to classify an activity as "unconventional" participation because the forms of participation are shifting, and many forms of participation, once considered unconventional, have now become rather mainstream.Furthermore, what may seem "unconventional" to older generations may well be rather "conventional" for young people, and vice versa.Therefore, we may speculate that sharing political content online -on Facebook, for example -may be much more "conventional" for young people than working for a political party or contacting a policy maker.The practices employed by generally and moderately active young citizens in both countries indicate that for a considerable proportion of young people digital participation is the primary, and therefore most conventional form of civic and political engagement.In particular, "unconventional" online participation seems to be a routine part of the repertoires of Estonian groups of General activists and Positively engaged activists, and Czech groups of General activists and Digital activists.Therefore, this study provides further support for the claims that the forms of civic participation are changing and younger generations often combine online with offline participation activities.
A more detailed look at the groups that, regarding their preferred practices, can be conceived as national counterparts of each other, further complicates the typological picture.Our analysis shows that the respective groups differ in socio-demographic indicators and political attitudes.Although some differences in the sampling procedures in Estonia and the Czech Republic as well as the specifics of cluster analysis prevent us from a more precise and conclusive comparison of the outlined groups across the Estonian and Czech samples, the characteristics of the groups provide quite differing profiles of the youth's participation in the two countries.In other words, the same sets of interrelated practices are linked to different political attitudes and socio-demographic characteristics in young people in the two national contexts: Estonian General activists cannot be considered the same as Czech General activists, as they comprise socio-demographically different people with dissimilar political attitudes, and the same applies to the rest of the crossnationally comparable pairs of groups.With the exception of gender, Czech General activists, in their political attitudes and socio-demographics, are more similar to the Estonian group of Positively engaged activists.As these two groups form the most well-off segments of the Czech and Estonian samples, respectively, we can say that a more advantaged socioeconomic position is a favourable precondition of more conventional engagement.
Furthermore, our analysis demonstrated that in terms of their parents' education, type of residence, and economic and cultural capital, Estonian General activists are rather disadvantaged compared to other groups.They also demonstrate the lowest support for democracy and the highest level of authoritarianism.All in all, these features suggest that this small group of Estonian youth seems to be disappointed in the mainstream politics, and their high level of general activism rather rises from protest-mindedness.Therefore, as the case of Estonian General activists illustrates, disillusionment with the established political system and protest-mindedness can, under certain circumstances, lead to higher levels of participation.This interpretation is partly in accord with the results of Amnå and Ekman (2013), who also demonstrated that active young people in Sweden perceived themselves to be worse off than passive young people (e.g. the standby, unengaged, and disillusioned groups).However, their analysis did not demonstrate young activists from worse-off families to be less supportive of democracy than other young people.In that regard we can rather see some similarities with the group of young people being generally active in anti-democratic and authoritarian ways in Banaji's (2016) typology.Interestingly, the tendency that the group of Estonian General activists is characterised by a lack of different capitals does not apply to the Czech sample: Czech General activists are definitely not disadvantaged, quite the opposite.
Therefore, based on theoretical assumptions and previous findings, Estonian General activists represent an interesting case, as these rather disadvantaged youths would be rather unlikely candidates as civic activists.If we consider the claims of Uslaner and Brown (2005) and Solt (2008) that greater inequality leads to less civic participation, it is surprising that in Estonia, which displays greater macro-level inequalities than the Czech Republic, we actually find outstanding levels of civic participation among the small group of relatively disadvantaged youth.However, one possible explanation for this contradictory finding could also be provided by Uslaner and Brown (2005), who suggest that inequality affects civic participation negatively mainly through generalised social trust.As an inversion of this theory, Estonian General activists are characterised by high levels of generalised social trust.Consequently, we can speculate that high levels of trust may buffer them from alienationas Estonian General activists generally trust other people and also state institutions, they do not despair, believing that participation would not be worth their effort; instead, they intend to change society in their desired direction.For this purpose, they employ many of the same conventional methods of participation as the more advantaged youth.In addition, they also engage in more radical methods, which may be more facilely available to them, as Gallego (2008) has demonstrated that protest activities seem to be somewhat more easily adaptable for disadvantaged groups.In summary, the case of Estonian General activists presents an intriguing example of controversial protest-minded young activists, who, while living in a rather unequal society, hold high trust in other people and state institutions.We have not come across protest-minded activists, who exhibit high levels of trust toward state institutions in the literature.
In general, we suggest that without considering the dynamics of national political contexts and the social and demographic structures of particular societies, focusing on the mere intensity and form of political and civic activities provides only a blind measure.Practically speaking, our results suggest that contextual specifics must be taken seriously when designing tools to boost civic participation in young people.Even though our analysis focused on two countries with a similar socio-historical background, the profiles of civically active young people and the factors or paths of political socialization leading to particular types differed quite substantially.While some aspects of previous typologies were applicable in the interpretation of our findings, it seems to be a challenge for us to elaborate a universal, context-independent model of youth participation.

Limitations
Although our analysis delivered a picture of the differences in young people's political and civic activities in Estonia and the Czech Republic that is quite easy to interpret, one must keep in mind that due to some differences in the sampling procedures in the two countries, the cross-national variation in youth civic participation patterns may partly result from the composition of the respondent groups in the respective countries.
Another issue to consider is typological analysis as a specific method: different methods and algorithms of clustering may provide partly different outcomes.We tested different ways of clustering our respondents, reaching, indeed, somewhat different outcomes.Nevertheless, despite minor differences, the general pattern that small groups of generally active young citizens, larger groups of generally passive young citizens, and moderately active intermediate subgroups remained the same across different analyses.Therefore, we are fairly confident that the presented typology represents the broad patterns of youth participation in the Czech Republic and Estonia quite well.
Typologies, nonetheless, are constructions created by researchers.Qualitative insights are needed to obtain a more detailed and varied picture of youth civic participation types.Future outcomes of the CATCH-EyoU project and further studies may provide information about youth civic and political engagement as constructed and interpreted by young people themselves.

Table 1 :
Sample characteristics and descriptive statistics for the Czech Republic and Estonia

Table 2 :
The percentage of young people in both countries who have done (at least rarely) any activity from the 18 indicators of civic and political participation 4

Table 4 :
Results of latent class analysis (LCA) and cluster analysis

Table 5 :
Socio-demographic characteristics and political attitudes in the youth participation types in Estonia

Table 6 :
Socio-demographic characteristics and political attitudes in the youth participation types in the Czech Republic