Advertisement

Journal of Youth and Adolescence

, Volume 47, Issue 4, pp 731–748 | Cite as

How is Civic Engagement Related to Personal Identity and Social Identity in Late Adolescents and Emerging Adults? A Person-Oriented Approach

  • Lyda Lannegrand-Willems
  • Basilie Chevrier
  • Cyrille Perchec
  • Alexia Carrizales
Empirical Research

Abstract

Adolescence and emerging adulthood are periods in life when individuals both question and define their place in society and form their identity. Meanwhile, active youth civic engagement represents a challenge for each democracy. The purpose of this study was to analyze the different forms of civic engagement among late adolescents and emerging adults and how they are related to personal identity and social identity, while adopting an integrative perspective through the lens of a person-oriented approach. The participants were 1217 (62.3% female) 16–24 year-old French students (M age  = 19.17; SD age  = 1.83). First, derived from cluster analyses, the findings emphasized diversity in civic engagement, from strong civic participation (in different formal and informal ways) to various forms of passivity. Diversity was also highlighted for personal identity and social identity profiles. Second, a Configural Frequency Analysis revealed a typical pattern associating passivity in civic engagement, personal carefree diffusion and rejection of social identity. Overall, these findings highlight an absence of general youth disaffection and provide a meaningful specific pattern for the understanding of passivity in political and civic matters in late adolescence and emerging adulthood.

Keywords

Civic engagement Personal identity Social identity Adolescence Emerging adulthood Person-oriented approach 

Introduction

In democratic countries, youth civic participation currently constitutes an important societal issue. For instance, the European Union (EU) aims to “encourage young people to be active citizens and participate in society in order to ensure that they have a say in the democratic processes that shape Europe’s future” (European Commission 2014, p. 1). This issue represents a challenge for every democracy, while youth disaffection in political participation, such as a low level of party membership, has been pointed out in several countries (e.g., Mycock and Tonge 2012). However, it has been argued that there are channels of political participation invested by youth beyond the traditional ones, leading to a broader typology of political participation and civic engagement (e.g., Ekman and Amnå 2012). The different forms of civic engagement should be given greater attention in current research.

Adolescence and emerging adulthood are periods when individuals both question and define their place in society, and form their identity (Arnett 2004; Lannegrand-Willems and Barbot 2015). Some scholars have focused on how “youth develop a sense of themselves as community members and citizens” (Crocetti et al. 2012, p. 523). In other words, the literature highlights relations between civic engagement and identity, since civic engagement can be viewed as self-expression (Amnå 2012) and as a domain of identity formation (Flanagan et al. 2012). When researchers study civic identity, they refer to the sense of self in civic life, which is a domain of personal identity, and to the sense of membership, which is a component of social identity (Hart et al. 2011). However, few empirical studies have looked deeper into the relations between civic engagement, encompassing its different forms, and identity, including its personal and social facets. The aim of our study was to investigate the various forms of civic engagement and personal and social identity construction during late adolescence and emerging adulthood, and to analyze how civic engagement is related to identity construction, while adopting an integrative perspective through the lens of a person-oriented approach. Indeed, our study focuses on individual patterns in order to develop a holistic and interactionist view of the individual in terms of civic engagement and identity.

Civic Engagement: Identifying Different Forms

Civic engagement refers to a broad construct that includes civic attitudes, knowledge, skills, and behaviors. It comprises the following: political participation, with conventional forms such as being a member of a political party, and non-conventional forms such as attending protests or signing petitions; civic participation (i.e., school-based community service, membership of a community organization, voluntary activities, etc.), and psychological engagement, such as paying attention to political or civic events (Barrett and Zani 2015). Non-participation - or disengagement - should be included among the forms of civic engagement, as it is not the simple reverse of engagement. It rather can be viewed as an expression and a position in itself when dealing with political and civic questions (Amnå 2012; Ekman and Amnå 2012). Disengagement comprises active antipolitical forms such as rejection of or disgust with politics, and passive apolitical forms like disinterest. All of these aspects provide an overview of the different forms of civic engagement, including political participation (formal and non-conventional, such as activism), civil participation, psychological involvement and disengagement (antipolitical and apolitical).

Identifying and studying various forms of civic engagement, including disengagement, is important in adolescence and emerging adulthood because young people are less and less committed to formal types of engagement (like being active within a party or a trade union), whereas they may be highly involved in other types of civic participation (like non-political youth organizations or volunteer activities) and psychological engagement (e.g., Crocetti et al. 2014). Moreover, in addition to a seeming lack of engagement that could be viewed as passivity or disengagement, some researchers have identified different forms of what might be called passivity. For instance, with a sample of Swedish adolescents and combining measures of participation and interest in politics and societal issues, Amnå and Ekman (2014) identified several profiles of civic and political engagement, one active and various passive ones. The active group was characterized by both high interest and participation. Regarding the passive groups, one was unengaged (both low interest and low participation), another was disillusioned (lowest interest and low participation), and the last one was characterized by the highest interest with an average manifest participation and labeled the “standby” citizens group. This latter group refers to adolescents who keep abreast of politics and societal issues and could be willing to participate if needed. This empirical study emphasized the diversity of civic engagement among adolescents by combining participation and psychological involvement. In our study, we extended this objective by identifying different profiles of civic engagement among a wide sample of late adolescents and emerging adults. We also aimed to explore the links with identity formation as it constitutes a core development task in adolescence and emerging adulthood (Arnett 2004; Erikson 1968).

Identity Formation: A Psychosocial Issue in Adolescence and Emerging Adulthood

Identity is a major concept in social sciences and is one of the most studied (Vignoles et al. 2011), while also being polysemic. A common definition is that identity refers to the “organization of self-understandings that define one’s place in the world” (Schwartz Montgomery and Briones 2006, p. 5). Nevertheless, even when limiting the field to psychology, there are multiple definitions and theoretical backgrounds to study identity. We can distinguish at least two levels of analysis: an individual one regarding personal identity, and a social one regarding social identity (for more details, see Vignoles et al. 2011). In this study, we focused on these two complementary levels: personal identity, which deals with the formation of personal values, goals and beliefs according to a neo-Eriksonian perspective (e.g., Luyckx et al. 2006b), and social identity, which refers to the sense of belonging to social groups according to a psychosocial perspective (e.g., Félonneau et al. 2013; Tajfel 1981).

Although personal identity and social identity have been studied within separate literatures, they are connected and integrated into the individual’s self. From their wide literature review on identity, Vignoles and colleagues supported an “integrative view of identity” (2011, p. 1), emphasizing that identities are “both personal and social” (p. 5). In another recent review, Schwartz, Luyckx and Crocetti focused on the concept of “intersectionality” (2015, p. 552) in order to highlight the convergence of personal and social aspects of identity. Empirical results support these assumptions, showing connections between personal attributes and social identities in self-structuration (e.g., Reid and Deaux 1996). Following these recommendations, we aimed to include both personal identity and social identity in our study.

Personal identity approach

Following on from Erikson’s assumptions on personal identity (Erikson 1968), Marcia (1966) specified two processes of identity formation: exploration, which refers to the consideration of various potential identity alternatives, and commitment, or the adoption of firm personal convictions. By combining exploration and commitment (and the presence vs. absence thereof), Marcia (1966) (Marcia et al. 1993) defined four identity statuses: identity achievement (exploration and commitment), moratorium (exploration but commitment still vague), foreclosure (no exploration but commitment), and identity diffusion (no exploration and no commitment).

In the last decade, broader process-oriented models of personal identity have been proposed and have extended Marcia’s conceptualization in order to better capture identity formation. One of these is the dual-cycle model (Luyckx et al. 2006a, b, 2008) which includes five identity processes. Two derive from Marcia’s identity paradigm: exploration in breadth and commitment making. Three other processes have been added. Exploration in depth concerns the extent to which one actively evaluates and maintains one’s current choices. Identification with commitment refers to the way that one feels certain about one’s choices and identifies with them. Finally, in order to capture the maladaptive process of identity formation, ruminative exploration indicates a locked cycle of repetitive exploration that hinders commitment formation (e.g., Luyckx et al. 2008). The identity processes are studied here at a global level that transcends content domains (Luyckx et al. 2006b). This five-process-oriented model has led to the introduction of new empirical identity statuses (e.g., Luyckx et al. 2008; Zimmermann et al. 2015): identity achievement (high on exploration in breadth, on exploration in depth and on commitment dimensions; low on ruminative exploration), moratorium (high on exploration in breadth; moderate to high on ruminative exploration; low to moderate on other dimensions), foreclosure (low on exploration dimensions; high on commitment dimensions), an undifferentiated status (moderate on all the dimensions) and two different forms of diffusion (both low on exploration in breadth, on exploration in depth and on commitment dimensions). These latter two forms have been labeled “diffused diffusion” and “carefree diffusion” because of their difference in terms of ruminative exploration (high on diffused diffusion but low on carefree diffusion, indicating that the individual is troubled vs. untroubled by the absence of personal commitments in the former and in the latter statuses respectively).

As assumed by Erikson (1968), achievement allows integration of the individual into society. Overall, the literature emphasizes that achieved personal identity contributes to positive development (e.g., Luyckx et al. 2008). Empirical studies have indeed demonstrated that achieved identity status is related to a high level of psychosocial adjustment while diffusion is related to the lowest level (e.g., Côté and Schwartz 2002). Specifically, in a large sample of emergent adult students, Schwartz and colleagues (2011) showed that the two identity diffusion profiles (namely diffused diffusion and carefree diffusion) were characterized by the lowest scores in positive psychosocial functioning dimensions and the highest scores in negative psychosocial functioning dimensions. Participants from these two profiles were similarly high on internalizing problems but the carefree diffusion was higher on externalizing problems. This study highlights the meaningful distinction between diffused diffusion and carefree diffusion, as carefree diffusion seems to have more antisocial tendencies.

Personal identity evolves during adolescence and emerging adulthood. In a systematic review of longitudinal studies on identity development published between 2000 and 2010, Meeus (2011) highlighted the fact that personal identity developed globally from diffusion in early adolescence to moratorium and foreclosure, and to achievement in emerging adulthood. Longitudinal designs are needed for the understanding of identity development but we consider that studying personal identity during adolescence and emerging adulthood within a cross-sectional design may remain relevant when it is related to the other main aspects of psychosocial development.

Social identity approach

In a psychosocial approach, social identity refers to a part of self, based on the individual’s sense of belonging to social groups (Tajfel 1981). An individual can belong to different types of social group, such as nationality, ethnicity, gender, family, which can be at complementary, overlapped or nested levels (Amiot et al. 2007). In our study, we focused on the sense of belonging to social places or territories (e.g., Félonneau et al. 2013), which are nested from proximal levels of social groups (neighborhood, city) within more distal levels (national and supranational).

One issue consists in understanding how the sense of belonging to different levels of community is shaped. Multiple nested social identities can be managed at different degrees of inclusion or exclusion (Félonneau et al. 2013) that are more complex and diverse in the context of a pluralistic society (Brewer 2001) or globalization (Jensen et al. 2011). The literature exploring the context of migration, specifically the model of adaptation to a bicultural context proposed by Berry (1997), can be useful in order to identify the different strategies used by individuals in the context of globalization. Jensen et al. (2011) proposed to expand this model in order to describe four different strategies in a globalized world where people may refer to their local identity and/or global identity: strategies of integration (development of both local and global identity), separation (development of local identity), assimilation (development of global identity) and marginalization (rejection of both local and global identity). We decided to apply this proposition to four different nested levels of community that characterize the sense of belonging to social territories, as carried out by Félonneau et al. (2013), namely neighborhood, city, national and supranational levels. Considered inclusively, young people may simultaneously develop a sense of belonging at different levels (i.e., strategy of integration). Others may select one social identity level, either reinforcing a local social identity (i.e., strategy of separation) or identifying with a supranational social identity level (i.e., strategy of assimilation). Finally, others may feel disconnected from all socio-territorial groups (i.e., strategy of marginalization), experiencing identity confusion (Jensen et al. 2011). However, to our knowledge, the ways that adolescents and emerging adults manage these different nested social identities have not been empirically tested.

The combination of multiple social identities may develop from adolescence to adulthood. Amiot et al. (2007) argued that the different levels of social identity moved from “fragmentation and differentiation” during adolescence “toward an increased integration of this diversity” (p. 377). A cross-sectional study has already pointed out that levels of socio-territorial belonging seemed to be more integrated among emerging adults than adolescents (Félonneau et al. 2013). Using a variable-oriented approach, the results of this study showed less contrast between the different levels in emerging adults compared to adolescents. Moreover, it was shown that integration was linked to positive psychological adaptation (Phinney et al. 2001). We aimed to test this assumption in greater depth by investigating how adolescents and emerging adults manage multiple nested social identities, using a person-oriented approach.

Why Civic Engagement Could be Related to Identity?

Identity is considered as a “developmental asset” (Luyckx et al. 2008, p. 616) that contributes to positive development and may influence behavioral outcomes, such as civic engagement. Concerning the relations between personal identity and civic engagement, empirical studies have shown robust links between identity achievement (or identity formation processes such as commitment and in-depth exploration) and various forms of active civic engagement (e.g., Crocetti et al. 2012). In a longitudinal study investigating the bi-directionality of relations between different styles of processing personal identity issues and civic engagement in adolescence, findings revealed stronger effects of identity on civic engagement than the reverse and negative relations between the diffused-avoidant style and civic participation (Crocetti et al. 2014). Moreover, using a clustering approach, a study identified four distinct groups of adolescents in terms of their involvement in political and community activities, and the group which was most involved presented a higher level of achievement and a lower level of diffusion than the uninvolved group (Pancer et al. 2007). Concerning relations between social identity and civic engagement, the absence of a sense of belonging to a community can lead to the avoidance of any civic behavior (Hart et al. 2011). The study of behavioral outcomes, such as civic engagement, related to personal and social identities appears to be particularly relevant for living together in inclusive and democratic societies (e.g., Flanagan and Christens 2011). However, the way that civic engagement and identity are related in a globalized context is still a core issue (Jensen et al. 2011).

Integrative Perspective According to a Person-Oriented Approach: From Dimensions to Profiles and Configurations

It is very common in psychological research to focus mainly on relations between variables at the level of the entire sample. However, many scholars have defended a modern person-oriented approach in which “the individual is regarded as a dynamic system of interwoven components that is best understood in terms of whole-system properties and often best studied by methods that retain these properties as far as possible, such as those that focus on individual patterns of information” (Bergman and Andersson 2010, p. 155). In our case, focusing only on the general or mean pattern that emerges for civic engagement does not make sense as we already know from the literature the diversity of forms that characterize civic engagement. This point is even more obvious for personal identity as the establishment of identity statuses (profiles) from identity processes (dimensions) is fundamentally embedded in the theoretical models dealing with identity formation. In the same way, we can better understand the complexity of social identity by considering the sense of belonging to different social groups jointly rather than separately. In line with such a holistic-interactionistic system view of the individual (Bergman and Andersson 2010), we adopted an integrative perspective that consisted in first establishing profiles from dimensions of civic engagement, personal identity, and social identity respectively. Second, we considered the individual through his membership of configurations that cross these profiles. In both cases, the methods of the person-oriented approach were favored in order to focus on individual patterns of information in a systemic perspective.

Current Study

Analyzing the way that civic engagement is related to identity in the French context within an integrative perspective and according to a person-oriented approach constitutes the rationale of the current study. In France, formal political participation by youth corresponds to new specific forms related to standby citizens. For instance, in 2017, two elections were held in France (presidential and legislative). Systematic voting accounted for 17.7% of 18–24 year-olds compared to 42.2% of 40–74 year-olds, while intermittent voting (i.e., voting in only one round when there are two rounds of voting) represented 62.2% (vs. 49.6%) and systematic abstention 20.2% (vs. 8%). Systematic voting has decreased among 18–24 year-olds since 2007 (31.3%) and 2012 (25.9%), the years in which the last two presidential elections were held. In the meantime, most young people have taken steps to register on the national list of electors (85% vs. 89% for the whole population aged above 18), attesting to their interest in political matters. However, many people do not seem to trust national political parties, nor EU politics. While the European Union argues for a European integration process, in the last European elections in 2014, one third of young French people voted for the extreme right-wing political party, and thus against EU institutions (Lannegrand-Willems and Barbot 2015). Furthermore, it should be emphasized that for many years civic education was not a core objective in France. In 2015, a national program of moral and civic education was defined and applied in all the French primary and secondary schools with the aims of conveying a basis of common values and preparing young people for the exercise of citizenship.

In this study, we pursued two complementary objectives. First, we aimed to investigate the various forms of civic engagement, personal identity statuses, and the combinations of different senses of belonging to nested socio-territorial levels during late adolescence and emerging adulthood. We expected a cluster solution to emerge for each construct, in line with the literature. We hypothesized that, as in Amnå and Ekman’s study (2014), various civic engagement profiles would emerge: active ones such as manifest participation combining high levels of several forms of civic and/or political engagement, passive ones such as disengagement, and a standby profile characterized by latent forms of civic engagement with high interest. Regarding personal identity, we expected to validate the six-cluster solution, well established in previous studies and based on the dual-cycle model (e.g., Zimmerman et al. 2015), so that the following identity statuses could be identified: achievement, moratorium, foreclosure, diffused diffusion, carefree diffusion, and undifferentiated status. Regarding social identity, we hypothesized that four different profiles could validate the assumptions of Jensen et al. (2011) on local and global identity in the globalized context, applied in our study to the combinations of different senses of belonging to four nested socio-territorial levels: integration (high on all socio-territorial levels), separation (high on proximal levels, i.e., city and neighborhood), assimilation (high on the supranational level) and marginalization (low on all socio-territorial levels). In addition, we expected differences in the distribution of adolescents and emerging adults across different clusters, showing development in civic engagement (i.e., emerging adults overrepresented in an active profile compared to adolescents, attesting to the active role in society while growing up), in personal identity (i.e., emerging adults overrepresented in achievement), and in social identity (i.e., emerging adults overrepresented in an integrated profile).

Second, we aimed to analyze how profiles of civic engagement, personal identity and social identity are related. Derived from results of various earlier studies (Crocetti et al. 2014a, b; Pancer et al. 2007), we expected to identify convergent configurations between civic engagement profiles, personal identity statuses, and social identity profiles, showing that the active civic participation profile is linked to an optimal development of personal and social identity profiles, respectively identity status of achievement and integration profile. Conversely, we hypothesized that the civic disengagement profile would be related to both personal identity diffusion (specifically, carefree diffusion because of its antisocial tendencies—Schwartz et al. 2011) and the social identity profile of marginalization (because of the link between the absence of a sense of belonging to a community and avoidance of any civic behavior—Hart et al. 2011).

Method

Participants

The sample for this study consisted of 1217 student participants from the urban area of Bordeaux (in South-West France; M age  = 19.17; SD age  = 1.83; 62.36% women; 15.28% of the participants’ parents had advanced degrees). Of the total sample, 387 were late adolescents from grades 10–12 recruited from five public high schools (M age  = 17.36; SD age  = 1.16; three general and two vocational high schools) and 830 were emerging adults from Bachelor’s and Master’s Studies from one university (M age  = 20.02; SD age  = 1.43; 33.05% were enrolled in human sciences and health studies; 27.72% in law and political sciences; 25.43% in sciences and technology; 13.80% in economics and management). In terms of ethnic background, the sample was composed of an absolute majority of French participants (94%). Data were gathered through paper questionnaires for the vocational high-school students during school hours, and through an online version for the general high-school and emerging adult students. All participants completed the self-report questionnaires voluntarily and anonymously.

Measures

Civic engagement

The Civic Engagement Scale was developed for the present study based on the theoretical typology of Ekman and Amnå (2012) and its applications. We included all the dimensions of the Participatory Behaviors Scale (PBS, Talò and Mannarini 2015; the dimensions were: disengagement, civil participation, formal political participation, and activism), and three dimensions of the Political Socialization Program (PSP, Amnå et al. 2010; the dimensions were: online activities and civic engagement, individual political interest and societal interest, and youth participation). A pool of 42 items was generated. First, a principal component analysis was conducted: seven items were deleted (saturations with an absolute value lower than .30) and nine factors were selected. With this solution, 63% of the variance was explained. Second, a confirmatory factor analysis with diagonally weighted least squares estimation1was performed to evaluate the fit of the 9 factors solution. According to the usual criteria (e.g., Hooper et al. 2008), the fit was acceptable: χ 2 (524) = 3210.09, CFI = .96, RMSEA = .06 [.064–.069], TLI = .96, and WRMR = 2.09. Finally, our Civic Engagement Scale was a 35-item questionnaire. Participants indicated to what extent they recognized each listed behavior as their behavior. Each item was rated on a 5-point Likert scale ranging from 1 (don’t apply to me at all) to 5 (totally apply to me) assessing nine civic engagement forms: disengagement (4 items; Cronbach’s alpha = .61; e.g., “Is unconcerned by politics”), information-discussion (5 items; Cronbach’s alpha = .78; e.g., “Is interested in political issues and events”), community organization involvement (3 items; Cronbach’s alpha = .71; e.g., “Volunteers in a social/civic/religious organizations”), voting (2 items; Cronbach’s alpha = .59; e.g., “Votes in elections or referenda”), formal participation (5 items; Cronbach’s alpha = .84; e.g., “Is a member of a party, syndicate or political organizations”), life-style related involvement (2 items; Cronbach’s alpha = .55; e.g., “Adopts a lifestyle with a clear social orientation (e.g., vegetarianism, anti-consumerism, punk subculture, etc.)”), legal activism on the Internet (5 items; Cronbach’s alpha = .78; e.g., “Is connected to a Facebook group (or the like) that is concerned with societal issues”), legal activism (4 items; Cronbach’s alpha = .77; e.g., “Organized a protest or boycott”) and illegal activism (5 items; Cronbach’s alpha = .84; e.g., “Painted political messages or graffiti on walls”).

Personal identity

The Dimensions of Identity Development Scale (DIDS) is a 25-item questionnaire (5 items for each dimension) developed by Luyckx et al. (2008) and rated on a 5-point Likert scale ranging from 1 (completely disagree) to 5 (completely agree). It assesses five identity processes (CM: commitment making, e.g., “I have decided on the direction I am going to follow in life”; IC: identification with commitment, e.g., “ I sense that the direction I want to take in life will really suit me”; EB: exploration in breadth, e.g., “I think actively about different directions I might take in my life”; ED: exploration in depth, e.g., “I think about the future plans I already made”; RE: ruminative exploration, e.g., “I worry about what I want to do with my future”). In the present study, we used a short form of the French version (Zimmermann et al. 2015) comprising three items per dimension. A confirmatory factor analysis with diagonally weighted least squares estimation was performed to evaluate the fit of a final version including 13 items, with 3 items for each dimension except for RE and IC (2 items). Cronbach’s alphas were .88, .81, .74, .64, and .86, respectively. The model suggests an adequate fit to the data: χ² (55) = 363.21, CFI = .99, RMSEA = .07[.062-.075], TLI = .99 and, WRMR = 1.67.

Social identity

The Social Identity measure consists of a 4-item questionnaire that assesses the sense of belonging to four socio-territorial levels. We used the procedure proposed by Félonneau et al. (2013). For the item “I feel I am… European; French; from my city; from my neighborhood”, participants answered for each level according to a 5-point Likert scale ranging from 1 (not at all) to 5 (entirely).

Plan of Analyses

Three cluster analyses were conducted using the same two-step procedure for each construct of this study: civic engagement, personal identity and social identity. The first step consisted in a hierarchical cluster analysis using Ward’s method and squared Euclidean distances. In the second step, the initial cluster centers obtained from this hierarchical analysis were used as nonrandom starting points in an iterative k-means analysis. The final number of clusters was determined according to three criteria (e.g., Luyckx et al. 2008): substantive theorizing, parsimony, and explanatory power (i.e., the most variance explained in each constituent dimension). For cluster interpretation, we used Cohen’s (1988) conventional criteria: an absolute value of 0.2 SD as a small effect, 0.5 SD as a moderate effect, and 0.8 SD as a large effect. Furthermore, differences between late adolescents and emerging adults in the three cluster solutions (civic engagement, personal identity and social identity) were investigated by Chi-square tests (with an examination of the standardized residuals with an absolute value greater than 2 to determine which cells significantly differ from the hypothesis of independence).

We then analyzed the configurations combining civic engagement, personal identity and social identity using Configural Frequency Analysis (CFA) (Lienert and Krauth 1975; von Eye 2002). Each configuration is a unique pattern crossing memberships in the cluster solutions obtained previously (e.g., participants who are in the first civic engagement cluster and in the first personal identity cluster and in the first social identity cluster are in the “1,1,1” configuration). CFA has been demonstrated to be particularly well suited to the analysis of multivariate cross classification of categorical variables in a person-oriented approach (Stemmler 2014; von Eye et al. 2013). Readers unfamiliar with CFA can find a detailed presentation of the application of CFA and a comparison with variable-oriented strategies in von Eye et al. (2006) in the context of parental attitudes toward alcohol consumption in adolescence. The aim of the CFA is to identify configurations (cells in the cross classification) in which the observed frequency is significantly higher (type) or lower (antitype) than the expected frequency derived from a base model.2Here, we used the first-order CFA base model that accounts for all main effects of variables included. As CFA involves multiple significance tests on the same data (one test per configuration), the significance threshold (α) must be protected to accommodate the increased risks that come with capitalization on chance and dependency of tests (von Eye et al. 2013). To establish this protected threshold (α*), we used the conservative Bonferroni procedure (α* = 0.05/r, where r is the total number of configurations). The number of participants may vary modestly from one analysis to another due to missing data, as we used all available information for each specific analysis without replacing the missing data. Thus, the sample to be analyzed varies from 1144 to 1217, depending on the statistics considered.

Results

Descriptive Statistics

Descriptive statistics (means and standard deviations) and bivariate correlations between all variables of civic engagement, personal identity and social identity are reported in Table 1. Regarding civic engagement dimensions, as expected, disengagement was negatively related to several dimensions of civic and political involvement and participation. Conversely, all those dimensions were positively interrelated, except voting which was positively associated with information-discussion, formal participation and life-style related involvement, and negatively with illegal activism. Between personal identity variables, correlations are in line with previous studies: high positive relations between commitment variables (commitment making and identification with commitment) and exploration variables (exploration in breadth and exploration in depth), and negative relations between ruminative exploration and other personal identity dimensions except exploration in depth. Regarding social identity, all four variables were positively interrelated. The correlations were higher between proximal levels: European and French, French and city, city and neighborhood.
Table 1

Descriptive statistics and correlations between civic engagement, personal and social identity dimensions

Scales

M

SD

α

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

Civic engagement

1. Disengagement

2.73

0.88

.61

−.41***

.04

−.13***

−.20***

−.06

−.12***

−.12**

−.03

−.07

−.11**

−.11**

−.09

.08

−.05

−.03

.01

.01

2. Information-discussion

3.12

0.94

.78

 

.15***

.16***

.39***

.30***

.43***

.30***

.16***

.03

.07

.19***

.14***

.05

−.02

−.02

−.01

−.10*

3. Community organization involvement

2.51

1.17

.71

  

.04

.27***

.26***

.25***

.38***

.21***

.06

.06

.14***

.08

.01

−.04

−.08

−.04

.01

4. Voting

4.07

1.07

.59

   

.10***

.02*

.00

.01

−.14***

.01

−.03

.09

.11***

.03

.07

.21***

.06

−.02

5. Formal participation

1.50

0.76

.84

    

.25***

.38***

.42***

.30***

.04

.07

.14***

.08**

.01

.02

−.01

−.02

−.05

6. Life-style related involvement

2.32

1.24

.55

     

.38***

.36***

.31***

−.09*

−.05

.13***

.02

.14***

−.11**

−.15***

−.11**

−.12***

7. Legal activism on the Internet

1.78

0.87

.78

      

.50***

.38***

−.02

.05

.15***

.10*

.10**

−.03

−.13**

−.07

−.07*

8. Legal activism

2.25

1.02

.77

       

.56***

.01

.03

.14***

.12***

.09*

.03

−.07

−.02

.00

9. Illegal activism

1.53

0.84

.84

        

−.05

.02

.02

.01

.10

−.03

−.13***

−.04

−.01

Personal identity

10. Commitment making

3.37

1.16

.88

         

.57***

.30***

.45***

−.57***

.10

.09

.08

.08

11. Identification with Commitment

3.26

1.19

.81

          

.26***

.44***

−.45***

.10*

.04

.08

.08

12. Exploration in breadth

4.06

0.90

.74

           

.40***

.03*

.05

.03

.03

−.03

13. Exploration in Depth

3.61

0.94

.64

            

−.20***

.08

.08

.04

.04

14. Ruminative Exploration

3.10

1.41

.86

             

−.01

−.03

−.09

−.13***

Social identity

15. European

3.85

1.36

              

.49***

.31***

.26***

16. French

4.31

1.17

               

.42***

.27***

17. City

3.86

1.34

                

.61***

18. Neighborhood

3.21

1.58

                 

α = Cronbach’s alpha

*p < .05; **p < .01; ***p < .001

Regarding the relations between civic engagement and personal identity, correlations show that there was a majority of significant relations with identity exploration variables. Disengagement was negatively associated with identification with commitment and exploration in breadth. All forms of engagement or participation (information-discussion; community organization involvement; formal participation; life-style related involvement; legal activism on the Internet; legal activism) were positively associated with exploration in breadth. Information-discussion, formal participation, voting, legal activism on the Internet and legal activism were associated with exploration in depth. Further, life-style related involvement, legal activism on the Internet and legal activism were positively related to ruminative exploration. Finally, illegal activism was not linked to personal identity dimensions.

Regarding the relations between civic engagement and social identity, voting was only positively associated with national identity. Conversely, legal activism on the Internet and illegal activism were negatively linked to national identity. Further, life-style related involvement was negatively associated with all the dimensions of social identity.

Cluster Analysis on Civic Engagement

Combining the 9 civic engagement dimensions, the cluster analysis yielded a seven-cluster solution presented in Fig. 1. This cluster solution accounted for 40.3% of the variance in disengagement, 38.9% in information-discussion, 26.3% in community organization involvement, 46.8% in voting, 61.9% in formal participation, 53% in life-style related involvement, 46.5% in legal activism on the Internet, 47.6% in legal activism and 53.8% in illegal activism. A discriminant function analysis supported this final cluster solution (Wilks’ lambda = .02; χ2(54) = 4524.7; p < .0001; 91.88% of cross-validated grouped cases correctly classified). The seven clusters presented different patterns of civic engagement. The formal participation cluster scored very high on formal participation, and moderately high on information-discussion and community organization involvement. The disengagement cluster scored very high on disengagement and low on information-discussion. The standby cluster had a median score on information-discussion, scored moderately high on voting, and moderately low on the other dimensions. The both formal and informal participation scored very high on all the dimensions except disengagement and voting. The informal involvement and participation cluster scored very high on life-style related involvement, legal activism on the Internet, legal activism and illegal activism. The life-style involvement cluster scored very high on life-style related involvement only. Finally, the passivity and nonvoting cluster scored very low on voting and moderately low on the other dimensions except disengagement.
Fig. 1

Cluster solution for civic engagement. z-scores for disengagement, information-discussion, community organization involvement, voting, formal participation, life-style related involvement, legal activism on the Internet, legal activism and illegal activism. N = 1170

The distribution of civic engagement clusters was significantly different between late adolescents and emerging adults [χ2(N = 1170, df = 6) = 75.52, p < .0001]. A detailed examination of the standardized residuals indicated that, compared to adolescents, emerging adults were overrepresented in formal participation, disengagement and life-style involvement, and underrepresented in passivity and nonvoting (see Table 2).
Table 2

Distribution of late adolescents and emerging adults in civic engagement, personal identity and social identity clusters

 

Late adolescents % (ASR)

Emerging adults % (ASR)

χ2(df)

Civic engagement clusters (N = 1170)

  

72.52(6)*

 Formal participation

11.41 (−2.2)

16.43 (2.2)

 

 Disengagement

15.21 (−2.1)

20.53 (2.1)

 

 Standby

22.51 (0)

22.58 (0)

 

 Both formal and informal participation

5.85 (0.5)

5.19 (−0.5)

 

 Informal involvement and participation

10.52 (−0.5)

11.47 (0.5)

 

 Life-style involvement

7.02 (−3.5)

14.38 (3.5)

 

 Passivity and nonvoting

27.48 (7.9)

9.42 (−7.9)

 

Personal identity clusters (N = 1197)

  

49.81 (5)*

 Achievement

28.45 (3.7)

18.82 (−3.7)

 

 Foreclosure

15.99 (3.8)

8.56 (−3.8)

 

 Moratorium

14.63 (0)

14.60 (0)

 

 Diffused diffusion

9.76 (−5.6)

23.52 (5.6)

 

 Carefree diffusion

12.74 (0.7)

14.23 (0.7)

 

 Undifferentiated

18.42 (0.7)

20.27 (0.7)

 

Social identity clusters (N = 1200)

  

48.66 (4)*

 National and supranational identity

20.00 (1.9)

25.06 (1.9)

 

 Moderate national identity

4.86 (−4.8)

14.58 (4.8)

 

 Local identity

13.78 (0.6)

15.18 (0.6)

 

 Rejection of social identity

13.25 (0.9)

15.30 (0.9)

 

 Integration of social identity

48.11 (6.1)

29.88 (−6.1)

 

ASR adjusted standardized residuals; bold adjusted standardized residuals reflect significant over—or underrepresentation

* p < .001

Cluster Analyses on Personal Identity and Social Identity

Personal identity

A final six-cluster solution emerged by combining the 6 personal identity processes (see Fig. 2). It accounted for 65.5% of the variance for commitment making, 57.5% for identification with commitment, 57.9% for exploration in breadth, 51.4% for exploration in depth, and 67.7% for ruminative exploration. A discriminant function analysis supported this final cluster solution (Wilks’ lambda = .037; χ2(25) = 3911.50; p < .0001; 94.7% of cross-validated grouped cases correctly classified). As expected, the achievement cluster scored high on all the dimensions except ruminative exploration. The foreclosure cluster scored moderately high on both commitment dimensions, moderately low on exploration in depth, and very low on exploration in breadth and ruminative exploration. The moratorium cluster scored intermediate on both commitment dimensions and high on all other dimensions. The diffused diffusion cluster scored low on both commitment dimensions, intermediate on exploration in breadth and in depth, but very high on ruminative exploration. The carefree diffusion cluster scored very low on both commitment dimensions, on exploration in breadth and in depth, and intermediate on ruminative exploration. Finally, the undifferentiated cluster scored intermediate on all the dimensions.
Fig. 2

Cluster solution for personal identity. z-scores for commitment making, identification with commitment, exploration in breadth, exploration in depth, and ruminative exploration. N = 1197

The distribution of personal identity statuses differed significantly between late adolescents and emerging adults [χ2(N = 1198, df = 5) = 49.81, p < .0001]. A detailed examination of the standardized residuals indicated that, compared to adolescents, emerging adults were overrepresented in diffused diffusion and underrepresented in foreclosure and achievement (see Table 2), which is an unexpected result.

Social identity

From the cluster analysis combining the different senses of belonging to four socio-territorial levels, a five-cluster solution was retained (see Fig. 3). This solution explained 69% of the variance for European, 77.7% for French, 51.2% for city, and 68.5% for neighborhood. A discriminant function analysis supported this final cluster solution (Wilks’ lambda = .045; χ2(12) = 3705.50; p < .0001; 96.5% of cross-validated grouped cases correctly classified). The five clusters could be clearly distinguished. The national and supranational identity profile scored moderately high on European and French, and moderately low on city and neighborhood. The moderate national identity cluster scored moderately high on French, and low on European, city and neighborhood. The local identity cluster scored moderately high on city and neighborhood, intermediate on French, and low on European. The rejection of social identity cluster scored low on all the dimensions, the lowest being French. Finally, the integration of social identity cluster scored high on all the dimensions.
Fig. 3

Cluster solution for social identity. z-scores for European, French, city and, neighborhood. N = 1200

The distribution of social identity clusters was significantly different between late adolescents and emerging adults [χ2(N = 1200, df = 4) = 48.66, p < .0001]. A detailed examination of the standardized residuals indicated that, compared to adolescents, emerging adults were overrepresented in the moderate national identity and underrepresented in the integration of social identity (see Table 2), which are unexpected results.

Configural Frequency Analysis (CFA) Crossing Civic Engagement, Personal Identity and Social Identity

Using CFA, we analyzed the distribution of participants in the 7 × 6 × 5 cross-classification (i.e., a total of 210 configurations) of the variables civic engagement (1 = formal participation, 2 = disengagement, 3 = standby, 4 = both formal and informal participation, 5 = informal involvement and participation, 6 = life-style involvement, 7 = passivity and nonvoting), personal identity (1 = achievement, 2 = foreclosure, 3 = moratorium, 4 = diffused diffusion, 5 = carefree diffusion, 6 = undifferentiated) and social identity (1 = national/supranational identity, 2 = moderate national identity, 3 = local identity, 4 = rejection of social identity, 5 = integration of social identity). Table 3 displays the observed and expected cell frequencies of the Civic Engagement × Personal Identity × Social Identity cross-classification of the participants’ cluster memberships. As the LR-χ² for the first-order base model was significant, LR-χ² = 313.71 (df = 194, p < .001), we can continue to identify configurations with frequencies significantly different from what was expected. The Bonferroni-protected α* = 0.0002 (0.05 / 210) led to the identification of one type:3(7,5,4). This significant type indicates that the configuration “passivity and non-voting × carefree diffusion × rejection of social identity” was more frequently observed than expected (fo = 10, fe = 3.33, z = 3.66, p < α*). The number of participants whose civic engagement is characterized by passivity and non-voting, whose personal identity status is carefree diffusion and who reject social identity is three times higher than expected by chance.
Table 3

Observed and expected frequencies for CFA crossing civic engagement, personal identity and social identity. N = 1144

Civic engagement

Personal identity

Social identity

  

1

2

3

4

5

1

1

10 (8.73)

6 (4.35)

8 (5.51)

5 (5.31)

22 (12.94)

 

2

1 (4.31)

0 (2.15)

2 (2.72)

1 (2.62)

6 (6.39)

 

3

9 (5.84)

5 (2.91)

1 (3.69)

2 (3.56)

15 (8.66)

 

4

8 (8.16)

5 (4.06)

3 (5.15)

4 (4.97)

8 (12.10)

 

5

3 (5.70)

3 (2.84)

1 (3.60)

1 (3.47)

6 (8.45)

 

6

6 (8.01)

3 (3.99)

8 (5.06)

5 (4.88)

15 (11.89)

2

1

3 (11.06)

5 (5.51)

5 (6.98)

4 (6.73)

15 (16.41)

 

2

10 (5.46)

3 (2.72)

2 (3.45)

0 (3.33)

15 (8.10)

 

3

9 (7.40)

0 (3.69)

1 (4.67)

4 (4.51)

11 (10.98)

 

4

14 (10.34)

12 (5.15)

5 (6.52)

6 (6.29)

14 (15.33)

 

5

7 (7.22)

3 (3.60)

11 (4.56)

7 (4.40)

10 (10.71)

 

6

9 (10.16)

8 (5.06)

6 (6.41)

5 (6.18)

14 (15.07)

3

1

15 (13.09)

7 (6.52)

10 (8.26)

3 (7.97)

28 (19.42)

 

2

5 (6.46)

4 (3.22)

5 (4.08)

3 (3.94)

12 (9.59)

 

3

13 (8.76)

1 (4.36)

5 (5.53)

3 (5.33)

14 (13.00)

 

4

11 (12.23)

7 (6.09)

9 (7.72)

5 (7.45)

13 (18.15)

 

5

11 (8.55)

5 (4.26)

6 (5.39)

1 (5.20)

20 (12.68)

 

6

9 (12.02)

6 (5.99)

4 (7.58)

4 (7.32)

19 (17.83)

4

1

4 (3.15)

0 (1.57)

3 (1.98)

4 (1.92)

9 (4.67)

 

2

0 (1.55)

2 (0.77)

0 (0.98)

0 (0.95)

2 (2.30)

 

3

5 (2.11)

1 (1.05)

0 (1.33)

3 (1.28)

3 (3.12)

 

4

1 (2.94)

2 (1.46)

0 (1.86)

3 (1.79)

3 (4.36)

 

5

1 (2.05)

0 (1.02)

0 (1.30)

2 (1.25)

1 (3.05)

 

6

4 (2.89)

0 (1.44)

1 (1.82)

5 (1.76)

3 (4.28)

5

1

5 (6.49)

1 (3.23)

4 (4.10)

5 (3.95)

7 (9.63)

 

2

0 (3.21)

0 (1.60)

2 (2.02)

2 (1.95)

5 (4.76)

 

3

8 (4.35)

2 (2.17)

3 (2.74)

4 (2.65)

8 (6.45)

 

4

8 (6.07)

2 (3.02)

10 (3.83)

7 (3.70)

7 (9.00)

 

5

2 (4.24)

3 (2.11)

6 (2.68)

4 (2.58)

2 (6.29)

 

6

6 (5.96)

2 (2.97)

1 (3.76)

5 (3.63)

7 (8.85)

6

1

6 (7.15)

3 (3.56)

5 (4.51)

3 (4.36)

4 (10.61)

 

2

2 (3.53)

2 (1.76)

0 (2.23)

2 (2.15)

4 (5.24)

 

3

6 (4.79)

1 (2.39)

1 (3.02)

2 (2.92)

7 (7.10)

 

4

13 (6.69)

6 (3.33)

6 (4.22)

5 (4.07)

11 (9.92)

 

5

2 (4.67)

1 (2.33)

6 (2.95)

2 (2.84)

5 (6.93)

 

6

10 (6.57)

8 (3.27)

4 (4.15)

8 (4.00)

6 (9.74)

7

1

7 (8.37)

7 (4.17)

6 (5.28)

4 (5.10)

12 (12.42)

 

2

4 (4.13)

1 (2.06)

5 (2.61)

5 (2.52)

14 (6.13)

 

3

0 (5.60)

0 (2.79)

4 (3.54)

7 (3.41)

6 (8.31)

 

4

6 (7.82)

4 (3.90)

3 (4.94)

3 (4.76)

5 (11.61)

 

5

8 (5.47)

2 (2.72)

4 (3.45)

10 (3.33)

4 (8.11)

 

6

10 (7.69)

2 (3.83)

5 (4.85)

7 (4.68)

10 (11.40)

Expected frequencies in parentheses. Configuration with an observed frequency that differs from the expected frequency in bold

Civic engagement clusters: 1 = formal participation, 2 = disengagement, 3 = standby, 4 = both formal and informal participation, 5 = informal involvement and participation, 6 = life—style involvement, 7 = passivity and nonvoting; Personal identity clusters: 1 = achievement, 2 = foreclosure, 3 = moratorium, 4 = diffused diffusion, 5 = carefree diffusion, 6 = undifferentiated; Social identity clusters: 1 = national/supranational identity, 2 = moderate national identity, 3 = local identity, 4 = rejection of social identity, 5 = integration of social identity

Discussion

Adolescence and emerging adulthood constitute two periods in life when individuals both question and define their place in the world, and form their identity (e.g., Arnett 2004; Lannegrand-Willems and Barbot 2015). Meanwhile, active youth civic engagement represents a challenge for democracies. In this study, we first highlighted the various profiles of civic engagement, personal identity and social identity in late adolescence and emerging adulthood, using a person-oriented approach. Our results revealed a meaningful cluster solution for each construct, emphasizing in particular the diversity of forms of civic engagement, comprising different profiles of formal and/or informal participation, a standby profile, a disengaged profile and a passivity and nonvoting profile. We then explored the configurational relationships between forms of civic engagement, personal identity statuses and social identity profiles. Our findings showed a meaningful pattern for the understanding of youth passivity in political and civic concerns that was typically combined with personal carefree diffusion and rejection of social identity.

Various Profiles of Civic Engagement

As expected, our findings revealed different forms of civic engagement among late adolescents and emerging adults. These findings show an absence of general youth disaffection, contribute to the literature on the diversity of civic engagement and support previous assumptions and findings in other western countries (e.g., Barrett and Zani 2015; Ekman and Amnå 2014). The analysis of the correlations between civic engagement dimensions enhances the findings regarding diversity, since most of the civic engagement dimensions were positively interrelated (for instance, we observed positive links between formal participation and informal involvement). Among the seven profiles identified, four were active, representing 43.6% of our sample. Late adolescents and emerging adults can be actively involved in civic engagement through formal participation, through both formal and informal participation, through informal involvement and participation and through life-style involvement. These results contribute to specifying the diversity of active youth civic engagement. We assume that some of these active profiles depend on the national context. Specifically, the two groups of informal participation (i.e., both the formal and informal participation group and the informal involvement and participation group) characterized by a low level of voting and a high level of legal and illegal activism (legal activism on the Internet, such as organization of online protest; legal activism, such as participation in legal demonstrations or strikes; illegal activism, such as participation in an illegal action) may be specific to the French context. For instance, compulsory voting exists in some countries, but not in France; furthermore, collective mobilization constitutes a unique French issue (Lannegrand-Willems et al. 2011). These contextual expressions of active civic engagement profiles deserve further cross-cultural investigations. Conversely, three profiles of what could be called passivity were identified. First, a passivity and nonvoting group was characterized by a clearly passive attitude toward political and civic issues. Second, a disengagement group was contrarily characterized by an opposite attitude toward political issues and hence this group does not constitute a passive group. Third, a standby group seemed to be concerned with politics and societal issues and was willing to participate if needed, as highlighted by Amna and Ekman (2014). These distinctions are highly meaningful, since an apparent passivity can hide potential active profiles. Finally, we observed that, compared to adolescents, emerging adults were underrepresented in passivity and nonvoting. This result is in line with the evolution of civic engagement during the transition to adulthood (Flanagan and Levine 2010). Emerging adults were more engaged in three profiles: formal participation and life-style involvement (which could be indicators of progress in formal and informal political and civic involvement), and disengagement. This reinforces the assumption that disengagement can be viewed as an expression and a position in itself when dealing with political and civic questions (Amnå 2012; Ekman and Amnå 2012).

Personal Identity Statuses

As in the literature and in empirical findings on personal identity, we identified six identity statuses: achievement, moratorium, foreclosure, diffused diffusion, carefree diffusion and undifferentiated status. The observed distribution was in line with results from previous studies, including a French adolescent and emerging-adult sample (e.g., Zimmermann et al. 2015), stressing a relatively high proportion of diffused diffusion (19.2% in our sample) characterized by a very high level of ruminative exploration. Additionally, the particularity of the carefree diffusion profile (i.e., not so carefree) was identified in those previous studies. These findings support the assumption that diffusion may increase in our Western societies (e.g., Côté 1996) and that the context of crisis and insecurity in France might contribute to this particular identity formation (Zimmermann et al. 2015). Further, we observed that emerging adults were overrepresented in diffused diffusion and underrepresented in achievement and foreclosure. This result is unexpected since it is the opposite of the maturation principle of personal identity development (e.g., Meeus 2011). We can consider that this result is due to cohort effects as there are cross-sectional data. However, the French national context may also contribute to explaining it, as it is characterized by a high rate of young people entering university (87.9% obtained the required baccalaureate diploma to enter university in 2017) associated with a high rate of youth unemployment (24.6% in France/18.7% in the European Union in 2016) compared to the overall unemployment rate (9.5% in France/7.9% in the European Union) (Direction de l’Evaluation, de la Prospective et de la Performance [DEPP] 2017; Organisation for Economic Co-operation and Development [OECD] 2017). This is an insecure context for emerging adult students and it may reinforce ruminative exploration and inhibit commitment processes (Lannegrand-Willems et al. 2016). Thus, our results support the assumption that identity constitutes an adaptation to the social context (Baumeister and Muraven 1996) and that empirical findings on personal identity should be systematically interpreted with regard to the socio-cultural context.

Social Identity Profiles

Regarding social identity, we found five different clusters that might be interpreted in line with the assumptions using Berry’s model expanded in the context of globalization (Jensen et al. 2011). The national and supranational identity group may be viewed as using a strategy of assimilation. According to Jensen et al. (2011), in assimilation young people embrace global culture instead of local culture. In the socio-territorial literature, local culture refers to proximal levels such as the city and the neighborhood (e.g., Félonneau et al. 2013). In our results, since within this profile national identity was associated with supranational identity, we assume that these two levels represent global levels in contrast to local ones. An opposing group, namely local identity characterized by the prevalence of sense of belonging to the city and neighborhood, can be seen as using a strategy of separation. The moderate national identity group constitutes an intermediate group between the two previous ones, and cannot be directly referred to Berry’s model applied to the context of globalization. However, the distinction between local identity and national identity groups is in line with results from a previous French study (Félonneau et al. 2013). It was shown that a local identity was prevalent among youth from a region with strong cultural identity, whereas a national sense of belonging prevailed among youth from a region with less cultural specificity. The rejection of social identity profile (characterized by no sense of belonging to any level) is akin to a strategy of marginalization. Finally, the integration of social identity corresponds to a strategy of integration of the four levels. This latter profile was the most frequent in our study (35.5%), followed by the national and supranational identity profile (23.5%), which can be viewed as two different inclusive strategies regarding nested levels of social identity (where distal levels like national and supranational include the proximal levels). Contrary to what we previously expected, emerging adults were overrepresented in the moderate national identity profile and underrepresented in the integration profile. This is not in line with a developmental hypothesis, in which we expected to observe progress toward integration of the four levels (e.g., Amiot et al. 2007). However, in a previous study in the French context (Félonneau et al. 2013), even though levels of socio-territorial belonging were more integrated among emerging adults than adolescents, the prevalence of the national level was observed. As we have previously mentioned, many young French people have little trust in EU politics. This can hinder the sense of belonging to this supranational level. The contextual aspects should then be taken into account in understanding the dynamics of social identity. Above all, the implementation of Berry’s model in the context of globalization (Jensen et al. 2011) to expand the potential different levels of sense of belonging contributes to the understanding of these dynamics.

Configurations of Civic Engagement, Personal Identity and Social Identity

By investigating the distribution of participants in the configurations crossing civic engagement, personal identity and social identity, a typical pattern was identified, combining: passivity and nonvoting, carefree diffused identity, and social marginalized identity (i.e., rejection of social identity). This result is in line with our expectations, specifying passivity as a problematic pattern for the individual and for society. Indeed, this specific configuration combines the problematic facets of each construct. According to the literature, passivity characterizes people who are unengaged in any political or civic concerns (Amnå and Ekman 2014); carefree diffusion is associated with antisocial tendencies (e.g., Schwartz et al. 2011); and marginalization refers to people who may feel disconnected from all social identity groups, experiencing identity confusion (Jensen et al. 2011). At the individual level, it can be assumed that this pattern is associated with negative psychosocial adjustment. At the societal level, this pattern represents a threat to the democratic process.

Moreover, this result emphasizes that no profile of active civic engagement was typically and simultaneously associated with one or several personal identity statuses and social identity profiles. Beyond the fact that the analyses we performed are highly conservative, the absence of typical configurations here stresses the diversity of active civic engagement profiles that cannot correspond to specific personal and social identity profiles. This argument is enhanced by examining the correlations between civic engagement and identity dimensions. Personal identity exploration processes (i.e., exploration in breadth and exploration in depth) were related to several forms of civic participation. In other words, personal identity development processes may have various active behavioral civic outcomes. Regarding social identity, national level of identity was positively linked to voting. However, since voting characterized three civic engagement profiles (voting mean scores were similar in formal participation, disengagement and standby groups), no specific type could subsequently be identified.

Practical Implications

These findings may have important social implications. In western and globalized societies where personal identity diffusion may increase (Côté 1996) and the combination of nested social identities becomes more complex (Jensen et al. 2011), we observe a diversity of personal identity, social identity, and civic engagement profiles. Civic engagement is related to personal and social identity construction. The literature on identity development processes has already emphasized their bidirectional relations and their mutual reinforcement (e.g., Crocetti et al. 2012; Hardy et al. 2010). In our cross-sectional study, we showed that a configuration combining passivity and nonvoting, carefree diffusion, and marginalization, could be particularly at-risk in terms of individual and societal concerns. Therefore, we should simultaneously promote the development of personal identity and social identity and encourage various forms of civic engagement in adolescence. For instance, in civic education, Youniss (2011) recommends that critical discussions and debates should be organized in the classroom and that civic participation and community action by youth should be encouraged. We assume that these combined activities may enhance identity exploration processes (exploration in breadth and in depth, which were significantly associated with civic and political engagement in our study), favor inclusive social identity (in order to inhibit rejection of any social identity and marginalization) and reinforce civic engagement in its diversity of expressions. This implies adopting openness to the democratic debate in education (Youniss 2011), promoting a sense of belonging to an inclusive society, and conceiving civic engagement in a broad perspective (Jensen and Flanagan 2008).

Limitations and Strengths of this Study and Suggestions for Future Research

This study has two major limitations. First is the cross-sectional design. The way civic engagement and identity profiles evolve over time needs to be examined in longitudinal studies from late adolescence to emerging adulthood. Second, although the study comprises a broad sample of late adolescents and emerging adults, it does not include migrant youth or ethnic minority youth for whom identity formation is a more complex process as it is both a developmental and an acculturative task (Motti-Stefanidi et al. 2012; Motti-Stefanidi and Masten 2017). Moreover, there are specific ways in which immigrant youth may engage in civic involvement (e.g., Ballard et al. 2015; Eckstein et al. 2015) that need to be deepened according to the sociocultural context. Future studies might examine how societies include or exclude migrant groups and ethnic minority groups (Flanagan and Levine 2010), and how these groups may develop personal identity, social and cultural identities, and may get involved and participate in various forms of civic engagement.

Nevertheless, the strength of the study is that, using a person-oriented approach which considers the person in a holistic-systemic view, the findings highlight meaningful relations between civic engagement, personal identity and social identity. They provide a better understanding of potential combinations between civic engagement and identity, particularly an at-risk configuration combining passivity and nonvoting, carefree diffusion and marginalization. They may also lead to relevant practical applications for civic education and avenues for future research. For instance, this study could be followed up with a focus on passivity and identity confusion, as identity confusion is linked to psychological problems, taking into account the context of globalization (Jensen et al. 2011). Future studies should also include comparisons between several western and democratic countries in order to analyze the role of the sociocultural context in the emergence of various forms of civic engagement, disengagement and passivity and the formation of personal identity and social identity. For instance, France is a country that promotes a model of universalist integrative identity, while North American countries have adopted a model of multicultural integrative identity (Lannegrand-Willems and Barbot 2015). Such contrasts between sociocultural contexts and their influence on civic engagement, personal identity and social identity development should be taken into account.

Conclusion

This study used a person-oriented approach to better understand the relations between civic engagement and identity in late adolescence and emerging adulthood. The study’s findings emphasize diversity in civic engagement, from strong civic participation (in different formal and informal ways) to disengagement, standby and passivity. An equally remarkable diversity is highlighted in personal identity and social identity profiles. This study also underscores the importance of taking into account personal and social identity in order to understand civic engagement at the individual level. A meaningful specific pattern of passivity in civic engagement emerged, associated with problematic personal identity and rejection of social identity. This configuration combining civic passivity, carefree diffusion and marginalization constitutes an at-risk pattern for the individual and for society, and leads us to recommend a focus on the simultaneous promotion of civic engagement, personal identity and social identity in civic education. Future research should analyze how political and civic passivity related to identity develops from adolescence to emerging adulthood, while taking into account the characteristics of the country in which it takes place.

Footnotes

  1. 1.

    The Diagonally Weighted Least Squares (DWLS) estimator respects the ordinal nature of the data using a polychoric correlation matrix.

  2. 2.

    As stated by von Eye et al. (2006, p. 993): “The expected frequencies are estimated under the assumption of variable independence. If this assumption is violated, variable associations must exist. However, instead of modeling these associations, we look for types and antitypes at the level of individual cells. These individual cell deviations from the assumption of variable independence carry the statement that variables are associated, at least locally.”

  3. 3.

    It should be kept in mind that the Bonferroni procedure we used is highly conservative in our case as the number of configurations is large. Few possibilities exist to contradict the model as a discrepancy between observed and expected frequencies has to be very large to be significant. We also carried out a less conservative analysis (i.e., a global Chi-square test on the three-dimensional cross-table, followed by an examination of the standardized residuals with an absolute value greater than 2 to determine which cells significantly differ from the hypothesis of independence). In line with CFA results, we found that there were more observations in the 7,5,4 configuration than expected under the null model. This less conservative approach also identified 14 other configurations in which participants were overrepresented and 2 configurations in which participants were underrepresented.

Notes

Acknowledgements

The authors would like to thank the students who kindly volunteered to participate in the study.

Authors’ Contributions

L.L.W. coordinated the conception of the study, its design and drafted the manuscript; B.C. participated in the conception, the design, and the data collection of the study, performed statistical analyses and participated in the drafting of the article; C.P. participated in the conception and the design of the study, performed statistical analyses and participated in the drafting of the article. A.C. participated in the conception and the design of the study and performed statistical analyses. All authors read and approved the final manuscript.

Funding

This research was undertaken with the partial financial support by the Erasmus + Programme of the European Union, strategic partnership project “Innovative Curriculum for Strong Identities in Diverse Europe (INSIDE)” No. 2016-1-LT01-KA203-023220 and with the financial support of the Research Department of Human and Social Sciences of the University of Bordeaux.

Data Sharing Declaration

The datasets generated and/or analyzed during the current study are not publicly available but are available from the corresponding author on reasonable request.

Compliance with Ethical Standards

Conflict of Interest

The authors declare that they have no conflict of interest.

Ethical Approval

All procedures performed in the current study were in accordance with the ethical standards of the national research committee and with the 1964 Helsinki declaration and its later amendments.

Informed Consent

Informed consent was obtained for all participants in the study.

References

  1. Amiot, C. E., de la Sablonnière, R., Terry, D. J., & Smith, J. R. (2007). Integration of social identities in the self: Toward a cognitive-developmental model. Personality and Social Psychology Review, 11(4), 364–388.  https://doi.org/10.1177/1088868307304091.CrossRefPubMedGoogle Scholar
  2. Amnå, E. (2012). How is civic engagement developed over time? Emerging answers from a multidisciplinary field. Journal of Adolescence, 35, 611–627.  https://doi.org/10.1016/j.adolescence.2012.04.011.CrossRefPubMedGoogle Scholar
  3. Amnå, E., & Ekman, J. (2014). Standby citizens: Diverse faces of political passivity. European Political Science Review, 6(2), 261–281.CrossRefGoogle Scholar
  4. Amnå, E., Ekström, M., Kerr, M., & Stattin, H. (2010). Codebook: The political socialization program. Sweden: Örebro University. Youth and Society.Google Scholar
  5. Arnett, J. J. (2004). Emerging adulthood: The winding road from late teens through the twenties. Oxford: Oxford University Press.Google Scholar
  6. Ballard, P. J., Pavlova, M. K., Silbereisen, R. K., & Damon, W. (2015). Diverse routes to civic participation across ages and cultures: An introduction. Research in Human Development, 12(1-2), 1–9.  https://doi.org/10.1080/15427609.2015.1010341.CrossRefGoogle Scholar
  7. Barrett, M., & Zani, B. (2015). Political and civic engagement: Multidisciplinary perspectives. London: Routledge.Google Scholar
  8. Baumeister, R. F., & Muraven, M. (1996). Identity as adaptation to social, cultural and historical context. Journal of Adolescence, 19, 405–416.  https://doi.org/10.1006/jado.1996.0039.CrossRefPubMedGoogle Scholar
  9. Bergman, L. R., & Andersson, H. (2010). The person and the variable in developmental psychology. Zeitschrift für Psychologie / Journal of Psychology, 218(3), 155–165.  https://doi.org/10.1027/0044-3409/a000025.CrossRefGoogle Scholar
  10. Berry, J. W. (1997). Immigration, acculturation, and adaptation. Applied Psychology: An International Review, 46(1), 5–34.  https://doi.org/10.1080/026999497378467.Google Scholar
  11. Brewer, M. B. (2001). The many faces of social identity: Implications for political psychology. Political Psychology, 22(1), 115–125.  https://doi.org/10.1111/0162-895X.00229.CrossRefGoogle Scholar
  12. Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). New York, NY: Academic Press.Google Scholar
  13. Crocetti, E., Erentaite, R., & Žukauskiene, R. (2014a). Identity styles, positive youth development, and civic engagement in adolescence. Journal of Youth and Adolescence, 43(11), 1818–1828.  https://doi.org/10.1007/s10964-014-0100-4.CrossRefPubMedGoogle Scholar
  14. Crocetti, E., Garckija, R., Gabrialaviciute, I., Vosylis, R., & Žukauskiene, R. (2014b). Reciprocal associations between identity and civic engagement in adolescence: A two-wave longitudinal study. International Journal of Developmental Science, 8(3-4), 115–124.Google Scholar
  15. Crocetti, E., Jahromi, P., & Meeus, W. (2012). Identity and civic engagement in adolescence. Journal of Adolescence, 35, 521–532.  https://doi.org/10.1016/j.adolescence.2011.08.003.CrossRefPubMedGoogle Scholar
  16. Côté, J. E. (1996). Sociological perspectives on identity formation: The culture-identity link and identity capital. Journal of Adolescence, 19, 417–428.  https://doi.org/10.1006/jado.1996.0040.CrossRefGoogle Scholar
  17. Côté, J. E., & Schwartz, S. J. (2002). Comparing psychological and sociological approaches to identity: Identity status, identity capital, and the individualism process. Journal of Adolescence, 25, 571–586.CrossRefPubMedGoogle Scholar
  18. DEPP. (2017). Le baccalauréat 2017 [The “baccalaureat” diploma 2017]. Ministère de l’Education Nationale. http://www.education.gouv.fr/cid56455/le-baccalaureat-2017-session-de-juin.html.
  19. Eckstein, K., Jugert, P., Noack, P., Born, M., & Sener, T. (2015). Comparing correlates of civic engagement between immigrant and majority youth in Belgium, Germany, and Turkey. Research in Human Development, 12(1-2), 44–62.  https://doi.org/10.1080/15427609.2015.1010346.CrossRefGoogle Scholar
  20. Ekman, J., & Amnå, E. (2012). Political participation and civic engagement: Towards a new typology. Human Affairs, 22(3), 283–300.  https://doi.org/10.2478/s13374-012-0024-1.CrossRefGoogle Scholar
  21. Erikson, E. H. (1968). Identity: Youth and crisis. New York: Norton.Google Scholar
  22. European Commission. (2014). Cordis programmes: Societal and political engagement of young people and their perspectives on Europe. http://cordis.europa.eu/programme/rcn/664964_en.html.
  23. Flanagan, C., Beyers, W., & Žukauskiene, R. (2012). Political and civic engagement development in adolescence. Journal of Adolescence, 35, 471–473.  https://doi.org/10.1016/j.adolescence.2012.04.010.CrossRefPubMedGoogle Scholar
  24. Flanagan, C. A., & Christens, B. D. (2011). Youth civic development: Work at the cutting edge. San Francisco, CA, US: Jossey-Bass.Google Scholar
  25. Flanagan, C. A., & Levine, P. (2010). Civic engagement and the transition to adulthood. The Future of Children, 20(1), 159–179.CrossRefPubMedGoogle Scholar
  26. Félonneau, M., Lannegrand-Willems, L., Becker, M., & Parant, A. (2013). The dynamics of sociospatial identity: Comparing adolescents and young adults in two French regions. Applied Psychology: An International Review, 62(4), 619–639.  https://doi.org/10.1111/j.1464-0597.2012.00498.x.CrossRefGoogle Scholar
  27. Hardy, S. A., Pratt, M. W., Pancer, S. M., Olsen, J. A., & Lawford, H. L. (2010). Community and religious involvement as contexts of identity change across late adolescence and emerging adulthood. International Journal of Behavioral Development, 35, 125–135.CrossRefGoogle Scholar
  28. Hart, D., Richardson, C., & Wilkenfeld, B. (2011). Civic identity. In S. J. Schwartz, K. Luyckx, V. L. Vignoles (Eds.), Handbook of identity theory and research, vols. 1 and 2 (pp. 771–787). New York, NY: Springer. https://doi.org/10.1007/978-1-4419-7988-9_32..Google Scholar
  29. Hooper, D., Coughlan, J., & Mullen, M. (2008). Structural equation modelling: Guidelines for determining model fit. The Electronic Journal of Business Research Methods, 6(1), 53–60.Google Scholar
  30. Jensen, L. A., Arnett, J. J., & McKenzie, J. (2011). Globalization and cultural identity. In S. J. Schwartz, K. Luyckx, V. L. Vignoles (Eds.), Handbook of identity theory and research, vols. 1 and 2 (pp. 285–301). New York, NY: Springer Science+Business Media. https://doi.org/10.1007/978-1-4419-7988-9_13..Google Scholar
  31. Jensen, L. A. & Flanagan, C. A. (2008). Immigrant civic engagement: New translations. Applied Developmental Science, 12(2), 55–56.  https://doi.org/10.1080/10888690801996996.CrossRefGoogle Scholar
  32. Lannegrand-Willems, L., & Barbot, B. (2015). Challenges of adolescent psychology in the European identity context. In E. L. Grigorenko (Ed.), The global context for new directions for child and adolescent development. New Directions for Child and Adolescent Development, 147, 69–76.  https://doi.org/10.1002/cad.20091.
  33. Lannegrand-Willems, L., Perchec, C., & Marchal, C. (2016). Vocational identity and psychological adjustment: A study in French adolescents and emerging adults. Journal of Adolescence, 47, 210–219.  https://doi.org/10.1016/j.adolescence.2015.10.005.CrossRefPubMedGoogle Scholar
  34. Lannegrand-Willems, L., Sabatier, C., & Brisset, C. (2011). France. In J. J. Arnett (Ed.), Adolescent Psychology Around the World, chap. 18, (257–271). New York: Psychology Press.Google Scholar
  35. Lienert, G. A., & Krauth, J. (1975). Configural frequency analysis as a statistical tool for defining types. Educational and Psychological Measurement, 35(2), 231–238.  https://doi.org/10.1177/001316447503500201.CrossRefGoogle Scholar
  36. Luyckx, K., Goossens, L., & Soenens, B. (2006a). A developmental contextual perspective on identity construction in emerging adulthood: Change dynamics in commitment formation and commitment evaluation. Developmental Psychology, 42, 366–380.  https://doi.org/10.1037/0012-1649.42.2.366.CrossRefPubMedGoogle Scholar
  37. Luyckx, K., Goossens, L., Soenens, B., & Beyers, W. (2006b). Unpacking commitment and exploration: Preliminary validation of an integrative model of late adolescent identity formation. Journal of Adolescence, 29, 361–378.  https://doi.org/10.1016/j.adolescence.2005.03.008.CrossRefPubMedGoogle Scholar
  38. Luyckx, K., Schwartz, S. J., Berzonsky, M. D., Soenens, B., Vansteenkiste, M., Smits, I., & Goossens, L. (2008). Capturing ruminative exploration: Extending the four-dimensional model of identity formation in late adolescence. Journal of Research in Personality, 42(1), 58–82.  https://doi.org/10.1016/j.jrp.2007.04.004.CrossRefGoogle Scholar
  39. Luyckx, K., Schwartz, S. J., Goossens, L., Soenens, B., & Beyers, W. (2008). Developmental typologies of identity formation and adjustment in female emerging adults: A latent class growth analysis approach. Journal of Research on Adolescence, 18(4), 595–619.CrossRefGoogle Scholar
  40. Marcia, J. E. (1966). Development and validation of ego-identity status. Journal of Personality and Social Psychology, 3, 551–558.  https://doi.org/10.1037/h0023281.CrossRefPubMedGoogle Scholar
  41. Marcia, J. E., Waterman, A. S., Matteson, D. R., Archer, S. L., & Orlofsky, J. L. (1993). Ego identity: A handbook for psychosocial research. New York: Springer.CrossRefGoogle Scholar
  42. Meeus, W. (2011). The study of adolescent identity formation 2000–2010: A review of longitudinal research. Journal of Research on Adolescence, 21, 75–94.  https://doi.org/10.1111/j.1532-7795.2010.00716.x.CrossRefGoogle Scholar
  43. Motti-Stefanidi, F., Berry, J., Chryssochoou, X., Sam, D. L., & Phinney, J. (2012). Positive immigrant youth adaptation in context: Developmental, acculturation, and social psychological perspectives. In A. S. Masten, K. Liebkind & D. J. Hernandez (Eds.), Realizing the potential of immigrant youth (pp. 117–158). New York, NY: Cambridge University Press.CrossRefGoogle Scholar
  44. Motti-Stefanidi, F., & Masten, A. S. (2017). A resilience perspective on immigrant youth adaptation and development. In N. J. Cabrera, B. & Leyendecker (Eds.), Handbook on positive development of minority children and youth (pp.19-34). https://doi.org/10.1007/978-3-319-43645-6_2. Cham: Springer.Google Scholar
  45. Mycock, A., & Tonge, J. (2012). The party politics of youth citizenship and democratic engagement. Parliamentary Affairs, 65, 138–161.CrossRefGoogle Scholar
  46. OECD (2017). Youth unemployment rate (indicator).  https://doi.org/10.1787/c3634df7-en. (Accessed on 28 November 2017).
  47. Pancer, S. M., 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(6), 741–759.  https://doi.org/10.1002/jcop.20176.CrossRefGoogle Scholar
  48. Phinney, J. S., Horenczyk, G., Liebkind, K., & Vedder, P. (2001). Ethnic identity, immigration, and well-being: An interactional perspective. Journal of Social Issues, 57(3), 493–510.  https://doi.org/10.1111/0022-4537.00225.CrossRefGoogle Scholar
  49. Reid, A., & Deaux, K. (1996). Relationship between social and personal identities: Segregation or integration. Journal of Personality And Social Psychology, 71(6), 1084–1091.  https://doi.org/10.1037/0022-3514.71.6.1084.CrossRefGoogle Scholar
  50. Schwartz, S. J., Beyers, W., Luyckx, K., Soenens, B., Zamboanga, B. L., Forthun, L. F., & Waterman, A. S. (2011). Examining the light and dark sides of emerging adults’ identity: A study of identity status differences in positive and negative psychosocial functioning. Journal of Youth and Adolescence, 40, 839–859.  https://doi.org/10.1007/s10964-010-9606-6.CrossRefPubMedGoogle Scholar
  51. Schwartz, S. J., Luyckx, K., & Crocetti, E. (2015). What have we learned since Schwartz (2001)?: A reappraisal of the field of identity development. In K. C. McLean & M. Syed (Eds.), The Oxford handbook of identity development (pp. 539–561). New York, NY: Oxford University Press.Google Scholar
  52. Schwartz, S. J., Montgomery, M. J., & Briones, E. (2006). The role of identity in acculturation among immigrant people: Theoretical propositions, empirical questions, and applied recommendations. Human Development, 49(1), 1–30.  https://doi.org/10.1159/000090300.CrossRefGoogle Scholar
  53. Stemmler, M. (2014). Person-centered methods: Configural frequency analysis (CFA) and other methods for the analysis of contingency tables. Cham: Springer.CrossRefGoogle Scholar
  54. Tajfel, H. (1981). Human groups and social categories: Studies in social psychology. Cambridge: Cambridge University Press.Google Scholar
  55. Talò, C., & Mannarini, T. (2015). Measuring participation: Development and validation the participatory behaviors scale. Social Indicators Research, 123(3), 799–816.  https://doi.org/10.1007/s11205-014-0761-0.CrossRefGoogle Scholar
  56. Vignoles, V. L., Schwartz, S. J., & Luyckx, K. (2011). Introduction: Toward an integrative view of identity. In S. J. Schwartz, K. Luyckx, V. L. Vignoles(Eds.), Handbook of identity theory and research (pp. 1–27). New York, NY: Springer. https://doi.org/10.1007/978-1-4419-7988-9_1.Google Scholar
  57. von Eye, A. (2002). Configural frequency analysis: Methods, models, and applications. Mahwah, N.J: Lawrence Erlbaum Associates.Google Scholar
  58. von Eye, A., Bogat, G. A., & Rhodes, J. E. (2006). Variable-oriented and person-oriented perspectives of analysis: The example of alcohol consumption in adolescence. Journal of Adolescence, 29(6), 981–1004.  https://doi.org/10.1016/j.adolescence.2006.06.007.CrossRefGoogle Scholar
  59. von Eye, A., Mun, E. Y., Mair, P., & von Weber, S. (2013). Configural frequency analysis. In T. D. Little (Ed.), The Oxford handbook of quantitative methods: Statistical analysis 2, (74–105). New York, NY: Oxford University Press.Google Scholar
  60. Youniss, J. (2011). Civic education: What schools can do to encourage civic identity and action. Applied Developmental Science, 15, 98–103.  https://doi.org/10.1080/10888691.2011.560814.CrossRefGoogle Scholar
  61. Zimmermann, G., Lannegrand-Willems, L., Safont-Mottay, C., & Cannard, C. (2015). Testing new identity models and processes in French-speaking adolescents and emerging adults students. Journal of Youth and Adolescence, 44(1), 127–141.  https://doi.org/10.1007/s10964-013-0005-7.CrossRefPubMedGoogle Scholar

Copyright information

© Springer Science+Business Media, LLC, part of Springer Nature 2018

Authors and Affiliations

  1. 1.University of Bordeaux, Laboratory of Psychology EA 4139BordeauxFrance

Personalised recommendations