Katharina Eckstein, Peter Noack, Philipp Jugert Pathways to Active Citizenship in Adolescence and Young Adulthood

Polskie Forum Psychologiczne, 2015, tom 20, numer 2, s. 165-183DOI: 10.14656/PFP20150202

PATHWAYS TO ACTIVE CITIZENSHIP IN ADOLESCENCEAND YOUNG ADULTHOOD *Katharina Eckstein1, Peter Noack1, Philipp Jugert21University of Jena, GermanyUniversity of Leipzig, Germany

2

Summary. Considering the importance of active citizenship, the goal of the present research was to identify meaningful predictors of intentions to participate inpolitics in adolescence (Study 1) and young adulthood (Study 2). Based on theassumptions of the civic voluntarism model, three main predictors were examined:Resources (educational level), experiences in social networks (club membership,important others’ attitudes toward political behaviors, political discussions), andindividual characteristics (attitudes toward political behaviors, internal politicalefficacy). Despite the differences in age, both studies identified a strikingly similarpattern of results: Especially experiences in social networks predicted changes inyoung people’s intentions to participate in politics. These effects, however, weremediated by the examined individual characteristics. While the effect of politicaldiscussions was largely mediated by the young people’s sense of internal politicalefficacy, the effect of important others’ attitudes was mediated by their own attitudes toward political behaviors.Key words: political participation, adolescence, young adulthood, civic voluntarism model, internal political efficacy

The call for young people’s active involvement in the public domain is anything but new and has often been repeated by politicians, researches, and teachersover the last decades. An integral part of active citizenship is political participation.Generally, research distinguishes between conventional and unconventional formsof political participation. While the former includes traditional means of involvement (i.e., joining political parties, supporting a political candidate), the latter comprises activities that go beyond the scope of traditional governmental and partypolitics (i.e., taking part in demonstrations, volunteering for charity; cf. Barnes et al.,1979). The most prominent and frequent form of voicing one’s political opinion is voting in elections. Other conventional activities were found to be less common andwere increasingly replaced by unconventional activities. These changes could beespecially observed among young people throughout the last decades (Torney-Purta et al., 2001; Youniss et al., 2002; Syvertsen et al., 2011). Adolescence and youngadulthood are critical and formative periods for political development, since youngpeople’s political understanding, their sense of identity, and their experiences withthe political domain develop significantly at that time (cf. impressionable years hypothesis; Sears, Levy, 2003). Given these manifold changes on the one hand andthe public call for young people’s political participation on the other hand, a betterunderstanding of factors contributing to active citizenship during these periods inlife is of particular importance.One prominent theoretical approach that allows for a thorough investigation offactors underlying political participation was introduced by the civic voluntarismmodel (CVM; Verba, Schlozman, Brady, 1995). According to the model three mainpredictors explain political behaviors: Resources, the access to social networks, andindividual characteristics. Hence, in response to the question of what keeps youngpeople from becoming politically engaged, Verba, Schlozman and Brady (1995) answered: “because they can’t; because they don’t want to; or because nobody asked”(p. 269).Concerning resources, research confirmed that people with higher material(e.g., income) as well as educational achievements are more likely to participate inpolitics – not only in adulthood (Verba, Schlozman, Burns, 2005; Gallego, 2007), butalso in adolescence (Smith, 1999; Torney-Purta et al., 2001). Other socio-structuralvariables, such as age and gender, have also been considered as resources predicting participation. However, the empirical evidence is rather mixed. That is, whilesome studies found age- and gender-specific patterns, others found no such differences. Generally, the effects of age and gender were found to be less consistent thanthe effects of income and education and seem to be dependent on the examinedforms of political participation (for age, see Watts, 1999; for gender see, Kuhn, 2010).The second factor CVM identifies as predictor of political behaviors are experiences and interactions in different social contexts (i.e., social networks). Amongthese, the family and peer group are of particular relevance. This was supportedby a number of empirical studies in the field of political socialization. Correspondingly, political discussions with important others, their approval of political behaviors, as well as the existence of social role models have shown to further youngpeople’s political participation (cf. Fletcher, Elder, Mekos, 2000; Oswald, Schmid,2006; Zaff, Malanchuk, Eccles, 2008). Other important social contexts in adolescenceand young adulthood are extracurricular activities and youth organizations. Consequently, young people are more likely to become politically active, when they areinvolved in organizational networks (Youniss, Yates, 1997; Smith, 1999; McFarland,Thomas, 2006; Schmidt, Shumow, Kackar, 2007). In order to account for a broadspectrum of social influences, we examined both experiences in primary social networks (i.e., family, peer group) and organizational networks (i.e., club memberships).

Finally, research has identified a variety of attributes and skills affecting youngpeople’s political participation, such as their political knowledge, interest, or values.They are summarized in CVM’s third factor – individual characteristics. Accounting for all possible and diverse individual factors in this study, however, was notfeasible. In order to have a parsimonious, yet potentially strong set of predictors, weselected our indicators of individual characteristics by referring to the assumptionsof expectancy-value theory (e.g., Atkinson, 1957; Eccles, Wigfield, Schiefele, 1998).According to the theory, two factors explain a person’s decision to show a certainbehavior: First, the subjective value the person places on the behavior and, second,the belief how well he or she will do on the activity (i.e., expectancy of success;Wigfield, Eccles, 2000). In applying these considerations to political behaviors, weaccounted for young people’s approval of political activities (i.e., attitudes towardpolitical behaviors) as an indicator of values and their sense of internal politicalefficacy (i.e., perceived competencies to participate effectively in politics; cf. Niemi,Craig, Mattei, 1991) as an indicator of expectancies. Both variables have shown tobe meaningful predictors of political participation in adolescence and young adulthood (for attitudes, see Jülisch, 1996; Pancer et al., 2007; for internal political efficacy, see Vecchione, Caprara, 2009; Eckstein, Noack, Gniewosz, 2013).Although the effects of resources, experiences in social networks, and individual characteristics are mutually related, most studies based on the CVM have examined its three main predictors either as independent units or in additive designs.Thus, even though previous research has shown that the CVM is a valid framework to predict political participation (e.g., Verba, Schlozman, Brady, 1995; Barkan,2004; Gabriel, 2004; McIntosh, 2006), little is known about the underlying (developmental) processes. Following contextual models of human development, resourcesand experiences in social networks might be considered to be rather distal factors,meaning that they do not affect political behaviors directly, but indirectly throughtheir influence on the proximal determinants, such as individual characteristics. Forexample, growing up in a well-off and politically active family environment, havingopportunities to take responsibilities in a youth organization, or seeing that friendsare politically engaged might heighten young people’s political awareness, suchas the approval of political activities or confidence in their own political abilities.Eventually, this should result in a higher readiness to participate in politics.

Current StudyBased on the above-summarized assumptions and findings, the goal of thepresent research was to test theory-based predictors of intentions to participate inpolitics in adolescence (Study 1) and young adulthood (Study 2). These two agegroups were intentionally chosen, since both have shown to be critical periods forpolitical development. Furthermore, we aimed at qualifying the interplay betweenpredictors of political participation as suggested by the CVM. More precisely, wehypothesized that the effects of resources (i.e., educational level) and experiencesin social networks (i.e., club membership, important others’ attitudes toward polit

ical behaviors, political discussions) on young people’s intentions to participate inpolitics were mediated by individual characteristics (i.e., young people’s attitudestoward political behaviors, internal political efficacy beliefs). Finally, apart from education, age and gender might also be considered as resources predicting youngpeople’s readiness to participate in politics. However, given the rather inconsistent pattern of empirical evidence, we took an exploratory approach. That is, weincluded age and gender as covariates in our research, thereby controlling for itsinfluences.

Study 1MethodSample. The empirical data of Study 1 were taken from a more comprehensive project on adolescents’ political development. Data were collected in 36 publichigh schools in the federal state of Thüringen in Germany1. Participating schoolswere randomly selected and were, thus, equally located in urban and rural areas.Overall, 704 adolescent high school students from three different age cohorts (Cohort 1: 6th grade, n = 293; Cohort 2: 8th grade, n = 288; Cohort 3: 10th grade, n = 123)were surveyed twice with a one-year time lag in between (Time 1: summer/ fall2004, Time 2: summer/ fall 2005). The first time point was primarily included inthe analyses to control for the stability of adolescents’ intentions to participate inpolitics (i.e., premeasure). Yet, changes in the outcome variable were predicted byvariables from the second measurement point. Therefore, the sample characteristicsconsidering age and gender summarized below correspond to the latter time point.Overall, adolescents’ average age was 15.35 years (SD = 1.20, age range: 13-19 years)and girls and boys were approximately equally distributed (ngirls = 379, 53.8%; nboys= 325, 46.2%). Students in the sample attended two different school types: A higher,college-bound school track (n = 481, 68.3%) and a lower, more practically orientedtrack (n = 223, 31.7%).Measures. Adolescents’ intentions to participate in politics were assessed bysix items (i.e., “I would … join a political party, participate in an environmentalgroup, work for a student council, vote in elections, demonstrate, donate money”; α = .60; Noack, Gniewosz, 2008). Hence, in Study 1, a rather broad measureof intentions to participate in politics was applied which combined conventionaland unconventional forms of participation. Given the scope of activities taken intoaccount, further exploratory factor analyses were carried out in order to examinewhether the scale’s diversity might be reflected in different dimensions of intentions to participate in politics. Principal component analysis (oblimin rotation)yielded a one-factorial solution. Factor loadings ranged between .43 (intention toMost schools in the federal state of Thuringia in Germany are public and therefore freeof tuition fees. Since all participating schools were randomly selected, they can be consideredas fairly representative for the state of Thuringia concerning school equipment, teachers’professional experience, and characteristics of the student body.1

vote in elections) and .69 (intention to join a political party). Since excluding singleitems with low loadings led to a further decline in Cronbach’s alpha, we decided toinclude this rather broad indicator of intentions to participate in politics that combined conventional and unconventional activities.The three main predictors of the CVM were operationalized as follows:First, school track (0 = lower track, 1 = higher track) and parents’ educational level(1 = no degree, 2 = 9th grade, 3 = 10th grade, 4 = high school diploma, 5 = university degree)were considered as indicators of educational resources. Second, the amount of political discussions with important others (4-item-scale; e.g., “Do you discuss national politics with your parents/ peers”; α = .80), important others’ attitudes towardpolitical behaviors (4-item scale; e.g., “My parents think we should take the chanceto participate in politics”; Fischer, Kohr, 2002; α = .73), and club membership (0 = nota member, 1 = member) were chosen as indicators of experiences in social networks.Club membership included different forms of organizational involvement, such asparticipation in fire brigades, the Red Cross, church clubs, and charity clubs. Andthird, individual characteristics were assessed by internal political efficacy (4-itemscale; e.g., “I consider myself as qualified to participate in politics”; Krampen, 1991;α = .78) and attitudes toward political behaviors (4-item-scale; e.g., “We should takethe chance to participate in politics”; Fischer, Kohr, 2002; α = .68). If not stated differently, response options ranged from 1 = I do not agree at all to 4 = I totally agree.Finally, age and gender (0 = male, 1 = female) served as control variables.Procedure. The main analyses were conducted by means of latent structuralequation modeling (SEM) using Mplus 6 (Muthén, Muthén, 1998-2010). Besides theadvantage to account for measurement errors (Byrne, 2006), the application of SEMallows investigating more complex relationships between several independent anddependent variables. To reduce model complexity, item parcels were constructedfor each scale. All latent variables were measured by two parcels. Items were assigned to parcels according to their factor loadings so that each parcel had a similarrelation to the latent construct (cf. Little et al., 2002).Two models were specified to test our assumptions: In a first step, young people’s intentions to participate in politics were predicted by educational resourcesand experiences in social networks (Model 1). In a second step, individual characteristics were added to the analyses (Model 2). To test for mediation, we examinedthe significance of the indirect effects of resources and experiences in social networks on young people’s intentions to participate in politics via individual characteristics. Moreover, we used bootstrapping to determine the confidence intervals ofthe indirect effects. In doing so, new data sets are generated, each containing an estimate of the indirect effect (NSamples = 1000; cf. Efron, Tibshirani, 1993). A significantindirect effect exists, if zero is not included in the 95% confidence interval.

ResultsThe means, standard deviations, and zero-order correlations of all variablesare shown in table 1. As the correlational relationships indicate, all predictors weresignificantly positively related to adolescents’ intentions to participate in politics.In a first step, adolescents’ intentions to participate in politics were predictedby their behavioral intentions from the previous measurement point (premeasure),school track, parental education (i.e., educational resources), club membership, political discussions with others, as well as important others’ attitudes toward politicalbehaviors (i.e., experiences in social networks). Moreover, the effects of the twocovariates – age and gender – were taken into account. Correlations among the predicting variables were freely estimated (Model 1.1). The model showed a good fit tothe data, χ²(33, N = 704) = 58.44, p = .004, CFI = .976, TLI = .951, RMSEA = .033, andSRMR = .026. Controlled for adolescents’ intentions to participate in politics at theprevious measurement point (β = .34, SE = .11, p = .003), political discussions withothers (β = .30, SE = .10, p = .003) and important others’ attitudes toward politicalbehaviors (β = .25, SE = .06, p < .001) explained changes in adolescents’ intentions toparticipate in politics. Club membership had no significant effect (β = .08, SE = .05,p = .093). From the considered indicators of educational resources, school track hada significant effect (β = .14, SE = .05, p = .003), while the effect of parental educationwas not significant (β = .01, SE = .05, p = .879). The effects of the two covariates –age and gender – were also not significant (βage = -.02, SE = .05, p = .716; βgender = .07,SE = .05, p = .211). Overall, the examined predictors explained 46.0% of the variancein adolescents’ intentions to participate in politics.In a second step, only those indicators of educational resources and experiences in social networks that had a significant effect on adolescents’ behavioral intentions were considered (i.e., school track, political discussions, important others’attitudes). Moreover, adolescents’ attitudes toward political behaviors and theirinternal political efficacy beliefs were included as mediating variables (Model 1.2,see figure 1 for a graphical depiction). The model fit the data well, χ²(45, N = 704)= 88.22, p = .000, CFI = 0.985, TLI = 0.972, RMSEA = .039, and SRMR = .025. Whenaccounting for adolescents’ attitudes toward political behaviors (β = .33, SE = .08,p < .001) and internal political efficacy (β = .31, SE = .09, p = .001), the effects of important others’ attitudes (β = .03, SE = .07, p = .683) and political discussion withothers (β = .03, SE = .10, p = .810) were no longer significant. Tests of indirect effectsshowed that the latter were mediated by the examined individual characteristics.While the effect of important others’ attitudes on intentions to participate in politics was mediated by students’ own attitudes toward political behaviors (β = .24,SE = .06, p < .001; 95%CI [0.10, 0.31]; 89.89 % of total effect), the effect of politicaldiscussions was largely mediated by adolescents’ sense of internal political efficacy(β = .22, SE = .07, p = .001; 95%CI [0.08, 0.28]; 67.74% of total effect). Furthermore,the effect of political discussions was also partially mediated by students’ attitudestoward political behaviors (β = .07, SE = .02, p = .001, 95%CI [0.02, 0.09]; 22.58% oftotal effect). The effect of school track, however, was not mediated by the examinedindividual characteristics. More precisely, while school track predicted young peo

ple’s intentions to participate in politics (β = .15, SE = .04, p < .001), the indirect effects via attitudes toward political behaviors (β = .01, SE = .01, p = .240; 95%CI [-0.01,0.03]) and internal political efficacy beliefs were not significant (β = -.01, SE = .01,p = .382; 95%CI [-0.04, 0.01]). Overall, the predictors explained a substantial amountof variance in adolescents’ intentions to participate in politics, as the R²-value of .54indicates2.

Figure 1. Graphical depiction of Model 1.2: Standardized coefficients for the structural equation model predicting changes in intentions to participate in politics from resources and experiences in social networks and mediated by individual characteristics(Study 1, N = 704)Note. Non-significant paths are dashed. Latent variables are shown with ovals* p < .05; ** p < .01; *** p < .001

DiscussionBased on the assumptions of the civic voluntarism model, the present researchexamined the effects of educational resources, experiences in social networks, andindividual characteristics on adolescents’ intentions to participate in politics. Weconsidered educational resources and experiences in social networks to be ratherdistal variables, the effects of which were mediated by individual factors. In lineIn a subsequent analysis we examined whether our findings were moderated by adolescents’ age or gender. We found no indication that the examined relationships of Model 1.2differed among girls and boys or among younger and older participants, Δχ² (20) ≤ 29.18,p ≥ .085.2

with our expectations, the results of our final model revealed that the effects of political discussions and important others’ attitudes toward political behaviors weresignificantly mediated by the examined individual factors. As our findings indicate there might be two different pathways to active citizenship: While the effect ofpolitical discussions was largely mediated by adolescents’ internal political efficacy beliefs, the effect of important others’ attitudes toward political behaviors wascompletely mediated by adolescents’ own attitudes. The effect of school track, incontrast, was not mediated by the examined individual characteristics.When interpreting these findings, some methodological aspects of this studyneed to be addressed. A first aspect concerns the assessment of adolescents’ intentions to participate in politics, which was based on a rather broad measure. Besidesits methodological drawbacks (low reliability of the scale and therefore a more conservative bias of the results; cf. Reis, Judd, 2000), it would be interesting to examinewhether these findings could be confirmed in a separate examination of conventional and unconventional forms of participation. Moreover, this study focused onparental attitudes only as an indicator of important others’ attitudes toward political behaviors. Yet as adolescents get older, peers become more and more a reference group from which they seek advice and approval (Syvertsen, Flanagan, 2006).A joint examination of parents’ and peers’ attitudes toward political behaviorscould therefore provide a more comprehensive picture of social influences (cf. Jugert et al., 2013). Finally, so far these conclusions are based on a single study. Yet,the question arises to what extent the findings might be a characteristic of the examined age group or forms of political participation. In order to address some ofthese questions in more detail, we conducted a second study that focused on youngadults and conventional political behaviors.

Study 2MethodSample. Data from Study 2 were derived from a longitudinal survey on political orientations in young adulthood. Students from a medium-sized universitylocated in the federal state of Thüringen in Germany3 were surveyed over a periodof two measurement points, divided by a six-month time lag (Time 1: spring 2009,Time 2: fall 2009). As in Study 1, data from the previous time point were primarilyincluded to control for the stability of young people’s intentions to participate inpolitics (i.e., premeasure). The study comprised 433 students. More female (n = 293,3In terms of demographic characteristics, such as students’ age or socio-economic background, the student body of this university is comparable to other German universities. MostGerman universities are public and therefore require no or only a small amount of tuitionfees, which reduces selection effects due to students’ socio-economic background. For somesubjects, such as psychology or medicine, however, there is a restricted admission, whichvaries between universities. Criteria of selection are mostly students’ Abitur grades (highschool diploma). Within this study, we tried to include students from different subjects inorder to reflect a broad student body, thereby reducing possible selection effects.

67.7%) than male (n = 140, 32.3%) participants took part in the survey and students’mean age was 21.79 years (SD = 2.52, age range: 17-33 years). More than half of thestudents (n = 242, 56.7%) were in their third semester. The overall number of completed semesters ranged between one and twelve (M = 4.00, SD = 2.10). Participantscame from different fields of studies, such as economics, humanities, natural science, and social science, in order to reflect a broad student body.Measures. In contrast to Study 1, this study concentrated on the examinationof conventional political behavioral intentions only (4-item-scale; e.g., “I would …work for a political party, visit political debates, support a political candidate, contact politicians”; α = .77). Parental education, as indicator of educational resources,was operationalized as in Study 1. Since the participants were exclusively university students and consequently shared a similar educational background (i.e., successful completion of the higher school track), no further indicator of educationalresources was included in the analyses. Indicators of experiences in social networkswere the amount of political discussions with others (4-item-scale; e.g., ”I oftendiscuss political issues with … my parents, siblings, friends, romantic partner”;α = .79), important others’ attitudes toward political behaviors (4-item-scale; e.g.,“Most people who are important to me think that it is important to participate inpolitics”; Ajzen, 2002; α = .79), and club membership (0 = not a member, 1 = member).Similar to Study 1, club membership included a broad array of organizational memberships (e.g., charity clubs, student networks). Finally, individual characteristicswere assessed by the same measures used in Study 1 (i.e., internal political efficacy,α = .88; attitudes toward political behaviors, α = .73). Age and gender (0 = male,1 = female) served as covariates. If not stated otherwise, response options rangedfrom 1 = I do not agree at all to 6 = I totally agree.

ResultsAs in Study 1, latent structural equation modeling was applied. Again, we useditem parceling to reduce the number of parameters to be estimated. The intercorrelations, means, and standard deviations of all variables are summarized in table 2.In a first step young adults’ intentions to participate in politics were predictedby parental education, club membership, political discussions with others, and important others’ attitudes toward political behaviors while controlling for the effectsof behavioral intentions at the previous measurement point (i.e., premeasure) as wellas the covariates age and gender. All predictors were allowed to covary (Model 2.1).The model fit the data well, χ²(29, N = 433) = 47.61, p = .016, CFI = .987, TLI = .973,RMSEA = .038, and SRMR = .026. Controlled for young adults’ behavioral intentions at the previous time point (β = .48, SE = .06, p < .001), both important others’attitudes (std. estimate = .18, SE = .06, p = .004) and political discussions with others(β = .29, SE = .08, p < .001) predicted changes in young adults’ intentions to participate in politics. Yet, neither parental education (β = .01, SE = .04, p = .791) nor clubmembership (β = -.06, SE = .04, p = .124) had an additional effect. Also, the effects ofthe covariates – age and gender – were not significant (βage = .06, SE = .04, p = .121;βgender = -.04, SE = .05, p = .339). The predictors explained 62.3% of the variance inyoung adult’s intentions to participate in politics.In a second step, we examined whether the effects of experiences in social networks that had significantly explained students’ intentions to participate in politicswere mediated by their internal political efficacy and own attitudes toward politicalbehaviors (Model 2.2, see figure 2 for a graphical depiction). Again, this model fitthe data well, χ²(39, N = 433) = 64.51, p = .006, CFI = 0.989, TLI = 0.982, RMSEA = .039,and SRMR = .031. Both individual characteristics significantly predicted changes instudents’ intentions to participate in politics (βattitudes = .32, SE = .07, p < .001; βefficacy= .30, SE = .08, p < .001). While important others’ attitudes were no longer a significant predictor of young adults’ political behavioral intentions (β = -.03, SE = .06,p = .667), the effect of political discussions with others was diminished, yet still significant (β = .21, SE = .09, p = .017). As in Study 1, we tested for the significance ofindirect effects. The results showed that the effect of important others’ attitudes wasmediated by students’ own attitudes toward political behaviors (β = .20, SE = .05,p < .001; 95%CI [0.10, 0.29]; 83.33% of total effect). The effect of political discussionswith others, in turn, was partially mediated by students’ sense of internal politicalefficacy (β = .26, SE = .07, p = .001; 95%CI [0.11, 0.41]; 63.41% of total effect). Witha R²-estimation of .72, the predictors explained a substantial amount of variance inyoung adults’ intentions to participate in politics4.

In a subsequent analysis we examined whether these findings were moderated byyoung adults’ age or gender. We found no indication that the examined relationships ofModel 2.2 differed among female and male or among younger and older participants,Δχ² (14) ≤ 11.61, p ≥ .637.4

Figure 2. Graphical depiction of Model 2.2: Standardized coefficients for the structural equation model predicting changes in intentions to participate in politics from resources and experiences in social networks and mediated by individual characteristics(Study 2, N = 433)Note. Non-significant paths are dashed* p < .05; ** p < .01; *** p < .001

DiscussionDespite differences in age and forms of political participation, the analyses ofStudy 2 revealed a strikingly similar pattern of results to that of our first study.Again, the examined indicators of experiences in social networks were found toexplain changes in young people’s intentions to participate in politics. While theeffect of important others’ attitudes on young adults’ behavioral intentions wascompletely mediated by their own attitudes toward political behaviors, the effectof political discussions was partially mediated by their internal political efficacybeliefs. In contrast to Study 1, however, political discussions with others also directly explained young adults’ intentions to participate in politics. One possible explanation for this finding might lie in the different forms of political activities thatwere examined. That is, unlike in Study 1, only students’ intentions to participatein conventional politics were assessed. It can be assumed that processes underlyingpolitical behavioral intentions differ slightly depending on the types of activitiestaken into account. Moreover, so far we only focused on two specific individualcharacteristics. Yet, other variables, such as young people’s political knowledge andinterest, have also shown to be related to political discussions with others and levelsof political participation (Torney-Purta et al., 2001; cf. Richardson, 2003). In orderto address this issue in more detail, future studies should take a more thorough

look at further individual characteristics as well as its relation to different types ofpolitical activities.Similar to Study 1, neither parental education, which was included as an indicator of resources, nor club membership explained changes in young adults’ intentions to participate in politics. While the former might be a characteristic of oursample (university students only and therefore less variance in parents’ educationalbackground), the non-significant effect of club membership might be related to theway of assessing this variable. Given its similarities to Study 1, we will address bothfindings in more detail in the general discussion.

General DiscussionThe adolescent and young adulthood years provide a broad spectrum of developmental changes, experiences, and challenges that shape young people’s politicalviews and behaviors. These developments, however, do not automatically lead toactive citizenship, which is, after all, an essential part of a vital democracy. It wastherefore the goal of the present research to gain a more elaborate understanding ofprocesses underlying young people’s intentions to participate in politics. By qualifying the interplay between different sets of predictors of political behaviors assuggested by the civic voluntarism model, we could show that the effects of political discussions and important others’ attitudes were largely mediated by individual characteristics. More precisely, our findings indicate that there are different pathways to active citizenship: Engaging successfully in political discussionsseems to affect young people’s intentions to participate in politics by enhancingtheir perceived political capabilities (i.e., internal political efficacy). This finding fitsthe existing literature underlining the importance of a constructive and supportive climate of discussion in which young people are encouraged to express theirown opinions und therewith have the chance to collect experiences with fundamental democratic principles – such as negotiating and compromising (Torney-Purtaet al., 2001; Hess, 2002; e.g., Campbell, 2008). According to Richardson (2003), theseinteractions are one example of vicarious experiences for they “allow […] the participants to learn more about political participation without necessitating actual involvement” (p. 29). Initiating discussions of current societal or political issues – beit in the classroom or within the family or peer context – might therefore presentone promising approach to stimulate young people’s political awareness and confidence (cf. Hahn, 1996; McDevitt, Chaffee, 2000).The perception of important others’ attitudes toward political behaviors, inturn, affected young people’s own attitudes. Eventually, this positive evaluationof political behaviors made young people more willing to take action themselves.As this finding further underlines, young people’s political development is boundto their social environment. That is, on their way to active citizenship they seek approval from close persons, adapt their actions and attitudes to existing role models,and are affected by prevailing social values and norms. Increasing the visibility ofpolitical activities or providing opportunities to get in contact with politically engaged peers could therefore represent another possible pathway to further youngstrona 178

people’s readiness to take an active role in the public sphere. We found this patternin our sample of adolescent high school students as well as in our sample of youngadult university students. The availability of two datasets with comparable designsdid not only allow us to consider a broader age range but also to examine different forms of political participation, lending additional value to our study. Yet, ourmeasures of political discussions, important others’ attitudes, and intentions to participate in politics differed in Study 1 and 2. To draw more sustained conclusions,further research using identical instruments across studies is needed.Contrary to our expectations, club membership – which was included as another indicator of experiences in social networks – had no additional effect on youngpeople’s intentions to participate in politics in both studies. The zero-order correlations between club membership and intentions to participate in politics werealso small (see table 1 and 2) and associations became non-significant when all predictors were taken into account in the model estimation. One possible explanationfor this finding might be the way of assessing club memberships. So far we onlyassessed whether young people were members or not. Yet, since membership cantake many different forms and shapes, future studies should include a broader indicator of membership that also considers time spent in a club, the kind of club, orthe quality of experiences (cf. mastery experiences; Bandura, 1997).Besides the effects of young people’s experiences in social networks, the examined indicators of educational resources either had a direct effect (cf. school track,Study 1) or were not related to young people’s intentions to participate in politics(cf. parental education, Study 1 and 2). Regarding the latter finding, it could behypothesized that the effects of educational resources and experiences in social networks are not independent from each other. That is, the availability of educationalresources might facilitate access to social networks, which, in turn, affects individual characteristics and eventually young people’s intentions to participate in politics.Empirically, however, we found no evidence for this assumption given the predominantly non-significant zero-order correlations between educational resources andexperiences in social networks (see table 1 and 2). Moreover, when interpreting thisfinding one should take into account that we only included educational level as anindicator of resources. In addition to intellectual aspects, however, the latter mightalso comprise material aspects, such as income or occupational status. The availability of books and newspapers at home represents another resource underlyingpolitical behaviors (Verba, Schlozman, Brady, 1995). Therefore, a broader index ofresources would be desirable in future studies in order to add to this study.Some limitations of the present research should be noted. Both studies werecarried out among high school as well as university students from the federal stateof Thüringen in former East Germany. This might raise the question of representativeness – especially against the backdrop of recent German history (for an overview of the consequences of German unification, see Silbereisen, 2005). In order tomake more sustained statements about the generalizability of our findings, furtherresearch should comprise participants from different regional and national contexts. Moreover, based on the study’s design, we cannot rule out that the examined

mediation processes operate in the opposite direction. That is, internal political efficacy might lead to higher levels of political discussions with important others.Likewise, young people who consider political activism as important might projecttheir attitudes onto their friends or parents and may therefore perceive the latters’attitudes as more positive. Longitudinal mediation analyses are therefore neededin future studies to investigate the relation between resources, social networks, individual characteristics and young people’s intentions to participate in politics. Inorder to examine changes in the outcome variable, we controlled for young people’sintentions to participate in politics at the previous measurement point. Therefore,this kind of modeling still provided more information than traditional cross-sectional designs. Finally, we considered young people’s behavioral intentions so far.Even though intentions have shown to predict political behaviors (e.g., Eckstein,Noack, Gniewosz, 2013), it is important to additionally include the amount of actualpolitical activities. After all, it is people’s real actions that make a difference and notonly their readiness to do so.In sum, by comparing two different age groups the present research providedsome valuable findings concerning the prediction of young people’s intentions toparticipate in politics. Overall, the civic voluntarism model has shown to be a helpful framework in order to examine meaningful predictors. Moreover, our studyindicated that there are different pathways to active citizenship: While politicaldiscussions affected young people’s intentions to participate in politics largely viatheir sense of internal political efficacy, the effect of important others’ attitudes wasmediated by young people’s own attitudes toward political behaviors. Given theoften-criticized lack of political involvement, especially among young people, a better understanding of the processes underlying intentions to participate in politics isof particular importance, as it might provide some answers to the frequently raisedquestion of what keeps young people from becoming politically engaged.ReferencesAjzen, I. (2002). Constructing a TPB questionnaire: Conceptual and methodological considerations. Retrieved from http://www.people.umass.edu/aizen/pdf/tpb.measurement.pdfAtkinson, J.W. (1957). Motivational determinants of risk-taking behavior. Psychological Review, 64, 359-372, doi: 10.1037/h0043445Bandura, A. (1997). Self-efficacy: The exercise of control. New York, NY: W.H. Freeman.Barkan, S.E. (2004). Explaining public support for the environment: A civic volunteerism model. Social Science Quarterly, 85, 913-937, doi: 10.1111/j.00384941.2004.00251.xBarnes, S.H., Kaase, M., Allerbeck, K.R., Farah, B.G., Heunks, F., Ingelhart, R., Jennings, M.K., Klingemann, H.D., Marsh, A., Rosenmayr, L. (1979). Political action: Mass participation in five Western democracies. Beverly Hills, CA: Sage.Byrne, B.M. (2006). Structural equation modeling with EQS: Basic concepts, applications,and programming. Mahwah, NJ: Erlbaum.strona 180

Campbell, D.E. (2008). Voice in the classroom: How an open classroom climate fosters political engagement among adolescents. Political Behavior, 30, 437-454, doi:10.1007/s11109-008-9063-zEccles, J.S., Wigfield, A., Schiefele, U. (1998). Motivation to succeed. W: W. Damon,N. Eisenberg (red.), Handbook of child psychology (s. 1017-1095). New York, NY:Wiley.Eckstein, K., Noack, P., Gniewosz, B. (2013). Predictors of intentions to participatein politics and actual political behaviors in young adulthood. International Journal of Behavioral Development, 37, 428-435, doi: 10.1177/0165025413486419Efron, B., Tibshirani, R.J. (1993). An introduction to the bootstrap. New York, NY:Chapman & Hall.Fischer, A., Kohr, H. (2002). Politisches Engagement [Political engagement].W: A. Glöckner-Rist (red.), ZUMA Informationssystem. Elektronisches HandbuchSozialwissenschaftlicher Erhebungsinstrumente (Version 6.00). Mannheim, Germany: Zentrum für Umfragen, Methoden und Analysen.Fletcher, A.C., Elder, G.H., Mekos, D. (2000). Parental influences on adolescent involvement in community activities. Journal of Research on Adolescence, 10, 29-48,doi: 10.1207/SJRA1001_2Gabriel, O.W. (2004). Politische Partizipation [Political participation]. W: J.W. vanDeth (red.), Deutschland in Europa: Ergebnisse des European Social Survey 2002-2003 (s. 317-338). Wiesbaden, Germany: VS Verlag für Sozialwissenschaften.Gallego, A. (2007). Inequality in political participation: Contemporary patterns in European countries. Retrieved from http://escholarship.org/uc/item/3545w14vHahn, C. (1996). Research on issues-centered social studies. W: R.W. Evans,D.W. Saxe (red.), Handbook on teaching social issues (s. 25-41). Washington,DC: National Council for the Social Studies.Hess, D. (2002). Discussing controversial public issues in secondary social studiesclassroom: Learning from skilled teachers. Theory and Research in Social Education, 30, 10-41 doi: 10.1080/00933104.2002.10473177Jugert, P., Eckstein, K., Noack, P., Kuhn, A., Benbow, A. (2013). Offline and online civic engagement among adolescents and young adults from three ethnicgroups. Journal of Youth and Adolescence, 42, 123-135, doi: 10.1007/s10964-0129805-4Jülisch, B.R. (1996). Zwischen Engagement, Apathie und Resignation: PolitischeOrientierungen Jugendlicher in Ost- und West-Berlin [Between engagement,apathy, and resignation: Political orientations of young people in East andWest Berlin]. W: J. Mansel, A. Klocke (red.), Die Jugend von heute: Selbstanspruch,Stigma und Wirklichkeit (s. 69-87). Weinheim, Germany: Juventa.Krampen, G. (1991). Entwicklung politischer Handlungsorientierungen im Jugendalter:Ergebnisse einer explorativen Längsschnittsequenz-Studie [The development of political action competencies in adolescence: Results of a longitudinal sequentialstudy]. Göttingen, Germany: Hogrefe.Kuhn, H.P. (2010). International perspectives on political socialization and gender:An introduction. W: A. Ittel, H. Merkens, L. Stecher, J. Zinnecker (red.), Jahrbuchstrona 181

Jugendforschung 8 (2008/2009) (s. 11-24). Wiesbaden, Germany: VS Verlag für
Sozialwissenschaften, doi: 10.1007/978-3-531-92320-8_1
Little, T.D., Cunningham, W.A., Shahar, G., Widaman, K.F. (2002). To parcel or not
to parcel: Exploring the question, weighing the merits. Structural Equation Modeling, 9, 151-173, doi: 10.1207/S15328007SEM0902_1
McDevitt, M., Chaffee, S. (2000). Closing gaps in political communication and
knowledge: Effects of a school intervention. Communication Research, 27, 259-292, doi: 10.1177/009365000027003001
McFarland, D.A., Thomas, R.J. (2006). Bowling young: How youth voluntary associations influence adult political participation. American Sociological Review, 71,
401-425, doi: 10.1177/000312240607100303
McIntosh, H. (2006). The development of active citizenship in youth. Dissertation
Abstracts International, 66 (12-B), 6951.
Muthén, L.K., Muthén, B.O. (1998-2010). Mplus user’s guide: Sixth edition. Los Angeles, CA: Muthén & Muthén.
Niemi, R.G., Craig, S.C., Mattei, F. (1991). Measuring internal political efficacy in
the 1988 national election study. American Political Science Review, 85, 1407-1413,
doi: 10.2307/1963953
Noack, P., Gniewosz, B. (2008). Schulbasierte Einflüsse auf die politisch-soziale (In-) Toleranz Jugendlicher [School-based effects on the political and social (in-)tolerance
of juveniles] (Unpublished research report, DFG Research Report FOR 481/2).
Germany: University of Jena.
Oswald, H., Schmid, C. (2006). The influence of parents and peers on political participation of adolescents in the New States of Germany. W: A. Sliwka, M. Diedrich, M. Hofer (red.), Citizenship education. Theory – research – practice (s. 85-96).
Münster, Germany: Waxmann.
Pancer, M.S., Pratt, M., Hunsberger, B., Alisat, S. (2007). Community and political
involvement in adolescence: What distinguishes the activists from the uninvolved? Journal of Community Psychology, 35, 741-759, doi: 10.1002/jcop.20176
Reis, H.T., Judd, C.M. (2000). Handbook of research methods in social and personality
psychology. Cambridge, Great Britain: Cambridge University Press.
Richardson, W.K. (2003). Connecting political discussion to political engagement:
The role of civic knowledge, efficacy and context for adolescents. Dissertation
Abstracts International, 64 (9-A), 3193.
Schmidt, J., Shumow, L., Kackar, H. (2007). Adolescents’ participation in service
activities and its impact on academic, behavioral, and civic outcomes. Journal of
Youth and Adolescence, 36, 127-140, doi: 10.1007/s10964-006-9119-5
Sears, D.O., Levy, S. (2003). Childhood and adult political development.
W: D.O. Sears, L. Huddy, R. Jervis (red.), Oxford handbook of political psychology
(s. 60-109). Oxford, Great Britain: Oxford Univ. Press.
Silbereisen, R.K. (2005). Social change and human development: Experiences from
German unification. International Journal of Behavioral Development, 29, 2-13, doi:
10.1080/01650250444000478
strona 182

Smith, E.S. (1999). The effects of investments in the social capital of youth on political and civic behavior in young adulthood: A longitudinal analysis. Political
Psychology, 20, 553-580, doi: 10.1111/0162-895X.00156
Syvertsen, A.K., Flanagan, C.A. (2006). Peer influence on political development.
W: L. Sherrod, C.A. Flanagan, R. Kassimir, A.K. Syvertsen (red.), Youth activism:
An international encyclopedia (s. 462-466). Westport, CT: Greenwood Press.
Syvertsen, A.K., Wray-Lake, L., Flanagan, C.A., Osgood, D.W., Briddell, L. (2011).
Thirty-year trends in U.S. adolescents’ civic engagement: A story of changing
participation and educational differences. Journal of Research on Adolescence, 21,
586-594, doi: 10.1111/J. 1532-7795.2010.00706.x
Torney-Purta, J., Lehmann, R., Oswald, H., Schulz, W. (2001). Citizenship and education in twenty-eight countries: Civic knowledge at age fourteen. Amsterdam, The
Netherlands: IEA.
Vecchione, M., Caprara, G.V. (2009). Personality determinants of political participation: The contribution of traits and self-efficacy beliefs. Personality and Individual Differences, 46, 487-492, doi: 10.1016/j.paid.2008.11.021
Verba, S., Schlozman, K.L., Brady, H.E. (1995). Voice and equality. Civic voluntarism in
American politics. Cambridge, MA: Harvard University Press.
Verba, S., Schlozman, K.L., Burns, N. (2005). Family ties: Understanding the intergenerational transmission of participation. W: A.S. Zuckerman (red.), The social
logic of politics: Personal networks as contexts for political behavior (s. 95-117). Philadelphia, PA: Temple University Press.
Watts, M.W. (1999). Are there typical age curves in political behavior? The “age
invariance” hypothesis and political socialization. Political Psychology, 20, 477-499, doi: 10.1111/0162-895X.00153
Wigfield, A., Eccles, J.S. (2000). Expectancy-value theory of motivation. Contemporary Educational Psychology, 25, 68-81, doi: 10.1006/ceps.1999.1015
Youniss, J., Bales, S., Christmas-Best, V., Diversi, M., McLaughlin, M., Silbereisen, R.
(2002). Youth civic engagement in the twenty-first century. Journal of Research on
Adolescence, 12, 121-148, doi: 10.1111/1532-7795.00027
Youniss, J., Yates, M. (1997). Community service and social responsibility in youth. Chicago, IL: University of Chicago Press.
Zaff, J.F., Malanchuk, O., Eccles, J.S. (2008). Predicting positive citizenship from adolescence to young adulthood: The effects of a civic context. Applied Developmental Science, 12, 38-53, doi: 10.1080/10888690801910567