Characteristics and Consequences of Having a Political Reputation in Class

The study examined the individual characteristics and consequences of psychological group processes that may lead some students to gain a reputation for being politically knowledgeable and verbal in class (a political reputation). Two normal samples of Swedish students were used, 13- year- olds (N = 835) and 16- year- olds (N = 795). Longitudinal data over one year were analyzed and showed that youths’ political reputation in class is established very early. Cross- sectional predictions showed that political interest predicted political reputation in class positively, and social fear predicted political reputation negatively in both cohorts. In addition, having a political reputation predicted increased political interest and political efficacy over one year. Further, mediation analyses showed that youths’ political predispositions, their political interest and political efficacy at T1, significantly operated on interest and efficacy at T2 via the political reputation. This suggests that political reputation partly functions as a booster of youths’ initial political predispositions over time. Future research is needed into the long- term consequences of having a political reputation.

To understand the political development of young people, knowledge is needed about how young people's experiences in their different life settings can increase their political interest and give them a sense of confidence in their political actions. This study examines one of these settings-their school classrooms. The school is commonly seen as an environment that can have a substantial impact on young people's political development-by providing them with knowledge about and motivation to attend to information about society and politics and by supporting classroom discussions and engagement in civic activities (Schulz et al., 2017). Some students can gain a political reputation among classmates for being well-informed about society and politics and initiate discussions in class on political issues that concern them. In this study, we examine the characteristics of these students and the effects this political reputation has for them. We propose that the classroom setting can be particularly valuable [Correction added on 13 May 2022, after first online publication: CRUI-CARE funding statement has been added.] for the students who gain a political reputation among classmates. Such a political reputation may increase their political interest and political efficacy. Thus, the school might act as a training ground for the political development of students whom classmates perceive as politically informed and verbal.
There is a large literature, dating back at least to the late 1950s, about the role that the school plays in adolescents' political development (Torney-Purta, 2002). In most democracies, schools have the task of providing civic education (Sundström & Fernández, 2013), but such education takes many forms in different countries (Lin, 2015;Schulz et al., 2017). Classroom discussions about society and the world are often seen as fostering both enlightenment and engagement (Parker, 2006). This study was conducted in Sweden, where civic education is not an independent subject but, in terms of its teaching, is integrated into many different subjects. In addition to content knowledge about issues related to societal and political institutions, civic education also concerns democratic values, civic skills, and engagement (Arensmeier, 2018).

Reputations in Social Groups
People gain information about others from direct observations, but they also spend a lot of time in social communication exchanging information about others (Foster, 2004). Indeed, perhaps in the majority of conversations with others, people engage in exchanging social information about others (Dunbar, 2004). They engage in gossip, which has been described as a form of cultural learning (Baumeister et al., 2004). Over time a certain communality of opinion develops; people in one's social group are recognized as possessing certain characteristics and reputations. Perhaps the most elaborated theoretical model for personal reputations was developed by Emler (1990Emler ( , 2013Emler ( , 2019. He proposed that people's reputations are social constructions that are based on social processes. In a group of people who meet continuously, group members will, over time, assign reputations to others, based on their observable behaviors and through interactions with others in the group. They will also acquire reputations based on their own behaviors. Most people are attentive to their reputations (Cavazza et al., 2015). In Emler's model, the individual will aim to generate behaviors that are in accordance with his or her social identity, which can easily be interpreted by others, and may lead to some degree of consensus about that social identity. Note that in identity theory (McCall & Simmons, 1978;Stryker & Burke, 2000), the question arises as to how individuals perceive their social identities and how this affects their behavior. In social identity theory, the question concerns how peoples' perception of belonging to a particular group or social category affect their behavior (Stets & Burke, 2000;Tajfel, 1982). Emler's personal reputation theory instead focuses on how other people in a group perceive a person' social identity and how this affects the behavior of the individual concerned.

A Political Reputation
This study is about the political reputations of students in their classes. By "political reputation" we mean the extent to which students are perceived by their classmates to be politically wellinformed and verbal in class. A classroom is normally a social environment that has high continuity over time in terms of structure, activity, membership, and interpersonal relationships. Each class member will form impressions of most of his or her classmates since he or she will be both repeatedly exposed to the classmates' behaviors and will have repeated opportunities to validate his or her judgments against those of others. In accordance with the requirement of intergroup stability in the reputation literature (Emler, 1990), the regularity of interactions between students in a class should maximize the likelihood that the students will reach some consensus about who has a particular social identity. Over many weeks, students acquire information about their classmates' behaviors, both through direct observation and indirectly from the reports of others in the class, a process that develops confidence in the impressions formed. This should be true for many facets of reputation including political reputation in the class.
Studying social reputations in class via peer nominations has a long history in research on child and adolescent adjustment (Bukowski & Newcomb, 1984;Hallinan, 1981;Masten et al., 1985). Social characteristics of students as judged by their classmates, such as being popular, prosocial, a leader, peer isolated, withdrawn, or aggressive, according to their classmates, predict adjustment, social functioning, work and academic careers much later in life (Cowen et al., 1973;Curlee et al., 2019;Gest et al., 2006;Morison & Masten, 1991;Wentzel et al., 2021). However, there is no previous research on young peoples' political reputations in class.
The research question as to which characteristics of students are associated with a political reputation in class needs to consider the stability of classmates' reports. Already in late childhood, classmates' nominations of social behaviors reflecting such characteristics as aggression, popularity, achievement orientation, and immaturity have shown stability coefficients above .50 over a one-year interval (Bukowski & Newcomb, 1984), while aggression, shyness, cooperativeness, and leadership have demonstrated stability coefficients in the .60 to .70 range over two years (Aleva et al., 2017), and peer victimization showed a stability coefficient of .50 over two years (Liu et al., 2014). For single items, the stabilities in the study by Aleva and coworkers (2017) ranged between .30 and .70. In short, peer nominations of social behaviors have substantial stability already by age 9 to 10. Potentially this is due to the aggregation of the opinions of all the students in a class which yield a highly reliable measure of an individual student's characteristics in the classroom. Whether the same high level of stability is also true for peers' nominations of their classmates' political knowledge and verbal behavior in class is unknown and is investigated in the study.
This first attempt focuses on some basic questions to describe students with a political reputation in class and to identify the consequences this may have for those students. We focus particularly on youths' political interest and internal political efficacy as these are central markers of youths' general political engagement (Barrett & Brunton-Smith, 2014) and are strong predictors of their political participation (Levy & Akiva, 2019;Strömbäck & Shehata, 2010). Attention on adolescents is particularly important because political interest and internal political efficacy become temporally stable later on in early adulthood (cf. Prior, 2019). Our basic study hypothesis is that political predispositions, like political interest and internal political efficacy, are important predictors of youths' political reputation in class, and that political reputation in class can function as a booster of the same predispositions over time. Hence, we attend to: 1. The characteristics of the students who have a political reputation. We propose that political interest and internal political efficacy are key predictors of having a political reputation in class. Of interest too are (1) other relevant personal characteristics that promote or preclude youths with regard to initiating political issues in class, (2) youths' self-perceived activity on political issues in the classroom, and (3) the role of social factors, like SES, gender, and immigrant background. 2. The role of a political reputation in the development of political interest and internal political self-efficacy. We assume that a political reputation in class contributes to increases in students' political interest and internal political efficacy over time.
In short, the study attends to the factors that predict political reputation in class and the effects that having a political reputation has for students.

Characteristics of Students with a Political Reputation
According to Emler (1990, p. 190) reputations are "social representations of the self" and should reflect individuals' identities and self-concepts. For students with a political reputation, a political interest is likely to be part of their motivations, identities, and self-concept. Having a political interest, that is, being attentive to political issues (Levy & Akiva, 2019) or having an ongoing positive evaluation of politics (Prior, 2010), may be the most obvious individual characteristic of students with a political reputation in class. Students' internal political efficacy, that is, their beliefs in their own ability to understand and participate effectively in politics (Craig et al., 1990;Miller et al., 1980), is also likely to impact political reputation. Beliefs in one's ability to discuss politics should be associated with being politically active in class and increase the likelihood that a student acquire a political reputation among classmates. Also, having a political reputation among classmates (an outsider perspective) might be particularly true for the youths who perceive themselves to be politically active in class (an insider perspective). Finally, social visibility in a group makes it easier to gain a reputation (Anderson & Shirako, 2008). In contrast, social fear, that is, fear of speaking in front of the class, may prohibit some adolescents from daring to express their political views and become politically visible among peers (Musso et al., 2020). We selected these four individual characteristicspolitical interest, internal political efficacy, perception of own political activities in class, and social fear-as potential determinants of students' political reputation in class. In addition, we examine the role that social characteristics, like SES, gender, and immigrant status, play regarding political reputation in class.

The Role of Political Reputation in the Development of Political Interest and Political Efficacy
The clearest evidence that the classroom can act as a reinforcing ground for students with a political reputation would be if these students' political interest and internal political efficacy increased over time more than the interest and efficacy of other students. Over-time correlations for individuallevel political interest are high (Prior, 2019), and political interest tends to stabilize as early as late adolescence to early adulthood (Russo & Stattin, 2017). This study examines whether political reputation may be one of the early social conditions that reinforces adolescents' political interest, before the point in time when only very high levels of stability in political interest can be expected. Specifically, if students perceive that their classmates consider them to be politically well-informed and able to articulate their political positions, this may further enhance their political interest. This argument agrees with Emler's (1990) findings that if youths advocated a particular political position, and had a reputation for this, they enhanced this position so as to be in line with their reputation. We also asked whether internal political efficacy is reinforced by developing a political reputation in class. Increased political efficacy can be expected on the grounds that a political reputation among classmates is a peer-group verification that the person has the capacity to discuss political issues publicly (Burke & Stets, 1999). More specifically, concerning the overall longitudinal links, we examine whether political reputation functions as a booster of youths' political predispositions, such that the links between initial levels of political interest and internal political efficacy and later levels of the same political predispositions are mediated by political reputation in the class. This mediational hypothesis is tested in the study.
We assume that, due to continual classroom interactions and comments shared among classmates, it should be obvious to students in a specific class which classmates are politically informed and verbal and which classmates seldom raise their voices about political matters. Political reputation in class in this study is measured through classmates' reports. However, a rival hypothesis is that it is the students' own perceptions of their political impact in the classroom, rather than their political reputation among others, that is associated with increased political interest over time. This contrasts political activity from an insider perspective with that of an outsider perspective, the collective judgement of social identity (Emler, 1990). This difference in perspectives offers the possibility of asking whether it is the person's reputation as politically active in class or instead the persons' own view of being politically active in class that is associated with future political interest and political efficacy.
Here, both the political reputation and the students' self-image of having input into political discussions in class will be considered in tandem.

The Present Study
In this longitudinal study, two samples of adolescents, 13-and 16-year-olds, were followed over one year. The 13-year-olds were students in compulsory school, and the 16-year-olds were students in upper-secondary (Swedish high) school. These two age cohorts were included not only to examine whether the proposed findings are robust across ages but also to see whether distinct developmental predictions can be identified at these two periods of life.
First, the study examines whether the political interest of students is the primary predictor of students' political reputation in class. The study also investigates whether other individual characteristics, such as students' internal political efficacy, perception of their own political activities in class, or social fear, predict their political reputation in class. Because we had no previous information about the temporal stability of political reputation in class and how this might affect results, we decided to perform two types of analyses. First, we employed a cross-sectional design to analyze the role that political interest, political efficacy, self-perception of one's political activities in class, and social fear have concurrently for political reputation. Next, we used a longitudinal design to examine the predictive power of the same individual characteristics for change in political reputation over one year. Of note is that most of the young people in this study remained in the same school and class over time. Hence, they were in the same intact student group over the one-year period and can be considered to know each other well.
Second, we examine whether having a political reputation among classmates tends to increase students' political interest and internal political efficacy over time. Here, we compare the role of political reputation with the students' own perception of having an input on political discussions in class. More generally in terms of what happens over time, we test a mediated model to examine the possibility that political reputation boosts youths' initial political interest and efficacy over time.
Finally, since there are mixed findings in the literature as to whether boys or girls are more politically interested and active (Cicognani et al., 2012;Schulz et al., 2017), gender is included as an active variable in the main analyses. Immigrant status was also controlled for in the analyses. Immigrant background is associated with less civic knowledge in many countries (Schulz et al., 2017), but for political involvement the pattern is more mixed. Previous research also suggests that parental education is a predictor of adolescents' political engagement (Torney-Purta et al., 2001). To control for the educational resources of the family, all analyses were replicated with subsamples of youths for whom parental reports about their education level were available.

Procedure and Samples
Two age cohorts of adolescents, 13-and 16-year-olds, were followed up in a longitudinal project on the political socialization of young people (Amnå et al., 2009). The study was conducted in a medium-sized city in Sweden between 2010 and 2015. The 13-year-old participants were recruited from 10 schools, with a total of 38 classes involved, while the 16-year-olds were recruited from three high schools, with a total of 57 classes involved. The schools were strategically chosen to be representative of the social and demographic characteristics of the adolescents in the city. The study employed an opt-out procedure in which parents of those young people in the two cohorts were informed about the study prior to the first assessment and had the opportunity to opt out their children from the study. The school assessments took place in the participants' classrooms during regular school hours. Trained test administrators distributed a self-report questionnaire to the participants without teachers being present. At the first assessment, a contribution of 1,000 SEK (about US$120) was made to the class fund, and, at subsequent assessments, the participants were compensated with movie tickets. One of the six local ethics review boards of the Swedish Ethical Review Authority approved all aspects of the procedure.
Here, we use information from two consecutive annual assessments of the 13-and 16-year-olds. The target samples, that is, the number of adolescents who were officially enlisted in the selected schools, at Time 1 (T1) were 960 and 1,052. The number of respondents who filled out the questionnaire at T1 were 909 (95% response rate) and 870 (83% response rate). The analytical samples used in the cross-sectional analyses included students who had complete data on the measures used in this study at T1 and were in classes with eight or more students, N = 835 in the younger cohort and N = 795 in the older. Among these, 734 of the 13-year-olds and 600 of the 16-year-olds participated also at T2, one year later, and were in classes with eight or more students (analytical samples for the longitudinal analyses). The percentage of the participants both of whose parents were born outside the Scandinavian countries was 21% and 17% in the respective cohorts, which is somewhat higher than the proportion in the country as a whole. The percentage of girls was 49% in both cohorts. The city reflects the country regarding its proportion of immigrants, income, level of education, unemployment rate, and the percentage who voted in the last election (Statistics Sweden, 2010).

Political Reputation in Class (Peer Nominations)
A measure of being politically informed and verbal in class was used, based on classmates' reports. All students were asked the question: "Here, different types of people are described. Which of your classmates are like these people?" Two of the types of people described were "Keeps track of what is going on in Sweden and in the world" and "Often starts discussions about societal issues in class." The students were asked to nominate three students who matched each of the two descriptions. The number of times the students were nominated for each of the two descriptions was divided by the total number of students who participated on the day of the data collection (and multiplied by 100 to increase clarity) and the two resulting scores were averaged. The two-item social reputation measure had alpha reliabilities of .75 and .79 for the age-13 cohort at T1 and T2. The corresponding reliabilities for the age-16 cohort were .87 and .91.

Political Interest
Level of political interest was measured with two questions. The first was: "How interested are you in politics?" (see the ANES, 2014 Panel Study for the same item). The response options ranged from 1 (totally uninterested) to 5 (very interested). This direct question has become a standard measure in the literature and has the benefit of avoiding "complications related to the distinction between interest and behavioral utterances or consequences of interest" (van Deth & Elff, 2000, p. 33). Second, the participants were asked the question: "People differ in how they feel about politics. What are your feelings?" They answered on a six-item response scale ranging from (1) loathe to (6) great fun. The interest question was rescaled to range between 1 and 6, and the two items were then aggregated. The two-item political interest measure had alpha reliabilities of .77 and .85 for the age-13 cohort at T1 and T2. The corresponding reliabilities for the age-16 cohort were .87 and .88.
Internal political efficacy was measured by nine items, all referring to a domain-specific efficacy belief in one's own actions to effect change in society (Bandura, 2006). After the stem "If I really tried, I could manage to …," the participants responded to nine statements, such as "Take on responsibility in a political youth organization"; "Be an active member of a political organization"; and "Discuss politics with persons with more experience than I have." The response scale ranged from 1 (I could definitely not manage that) to 4 (I could definitely manage that). Alpha reliability was .90 in the younger cohort and .93 in the older at T1. Both reliabilities at T2 were .94.

Fear of Speaking in Front of the Class
This was a single item included in an instrument that is a modified version of the Social Phobia Screening Questionnaire (SPSQ; Furmark et al., 1999), created for adults. The current version was adapted for children and adolescents up to the age of 18 and is called the SPSQ-C (the Social Phobia Screening Questionnaire for Children); it was piloted for the current study. The specific item used was "State how much fear you usually have when you speak in front of the class." The question was answered on a 3-point scale, ranging from 1 (no fear) to 3 (a lot of fear).
Self-perceived activity on political issues in the classroom was a four-item scale with questions such as "Have you ever brought up political issues during class?" and "Has it happened that you have brought up a societal problem for discussion during class?" The response scales ranged from 1 (has never happened) to 5 (very often). The alpha reliabilities were .90 and .91 for the age-13 cohort at T1 and T2 and .88 and .89 for the age-16 cohort.
Immigrant status was a dichotomous measure differentiating between students both of whose both parents were born outside the Nordic countries (coded 1) and other students (coded 0).

Attrition Analyses
Logistic regression analyses were performed to compare the T1 participants who took part and those who did not take part in the data collection at T2 on all the study variables at T1. For both age cohorts, no significant differences emerged for any of the variables at T1. Considering the low Nagelkerke R 2 coefficients, .02 in the younger sample and .01 in the older, the T2 samples were not regarded as biased. Table 1 reports the correlations among the measures used in the study. Three results stand out. In both cohorts, moderate correlations appeared among the measures of political interest, political efficacy, and self-perceived activity on political issues in class. Political reputation was less strongly correlated with these measures in the age-13 cohort than in the age-16 cohort. Note that youths' political reputation was moderately related to the participants' own perceived activity on political issues in class in the age-16 cohort but was less so in the age-13 cohort. Apparently, the youths' perception of their influence on political discussions in class did not match up with their political reputation in the younger cohort. Second, political interest, political efficacy, and self-perceived activity on political issues in class correlated significantly with political reputation but were uncorrelated with gender. Nonetheless, girls were less likely to have a political reputation than boys. Finally, participants whose both parents were born outside the Nordic countries reported more political interest, political efficacy, and activity on political issues in class in the younger cohort compared with other students. This was not the case in the older cohort.

Stability over Time of Political Reputation in Class
Before examining the first research question, the temporal stabilities of the measures of political interest, political efficacy, and political reputation need to be considered. As shown in Table 1, as might be expected, the temporal stabilities of all three measures were higher in the older cohort than in the younger. Also note that the temporal stability of political reputation was higher than the stability of political interest and political efficacy in both cohorts (.75 vs. .54 and .52 in the younger cohort and .82 vs. .74 and .67 in the older cohort). Apparently, the temporal stability of the students' political reputations was very high already in the youngest cohort.

Characteristics of the Students with a Political Reputation
Although we do not focus on contextual features, multilevel models were used to account for the nested structure of the data (adolescents nested in classes and classes nested in schools). A full three-level model which considers the schools is technically not applicable because the computation of parameters would exceed the number of school clusters. We therefore used two-level randomintercept models with the primary goal of explaining individual-level variation in the dependent variables while controlling for the variation in the mean response across different classes. All the continuous independent variables are measured at level 1, the individual level, and were group-mean centered.
We assumed that the students' own political interest would be the strongest predictor of their political reputation in class. The other individual characteristics assessed, the students' political efficacy, perceptions of their own activities on political issues in class, and social fear, might also affect their classroom reputation. For the two age cohorts, Table 2 reports the multilevel cross-sectional predictions of political reputation in class from these individual characteristics. .82*** -.25*** .36*** .23*** .30*** .15*** −.21*** .10** . Note. The age-13 cohort is above the diagonal and the age-16 cohort below. Gender was coded 0 = girl and 1 = boy, and Immigrant background was coded 0 = both parents born outside the Nordic countries and 1 = all other parents. *p < .05 **p < .01 ***p < .001.  Table 2, of the different predictors, political interest was the strongest predictor of political reputation in both cohorts. In both cohorts, fear of speaking in class and immigrant status predicted political reputation negatively. Youths whose parents were born outside the Nordic countries had lower political reputation than other youths. In the older cohort, political efficacy was a unique significant predictor as was youths' self-perceived political activities in class and gender (political reputation in class was higher for male than female students). Altogether, political interest, social fear, and immigrant status were all independent predictors of political reputation in both cohorts. Table 2 also reports longitudinal analyses of the independent variables predicting changes in political reputation over one year. As seen in the table, there were some significant predictors, but all of them were weak. In both cohorts, social fear was a significant negative predictor. In the older cohort also political interest and self-perceived political activities in the class were positive predictors, and immigrant status was a negative predictor of changes in political reputation from T1 to T2. In both cohorts, the independent variables contributed with only 2% of the variance over and above the contribution of political reputation at T1. Apparently, this means that once established, the political reputations of students are not much affected by the individual characteristics examined in this study.

The Role of Having a Political Reputation in the Development of Political Efficacy and Interest
The possibility that students' political reputations in class increases their political interest and efficacy over time was examined with regression analyses. Table 3 shows the predictions of students' political interest and political efficacy at T2 from their political reputation in class at T1, their political interest and political efficacy, self-perceived political activities on political issues in class, and controlling for gender and immigrant status at T1. As shown in the table, the findings were similar for both age cohorts. Political reputation in class was a significant positive predictor of both political interest and political efficacy at T2. The self-perception of being active in class on political issues was not a significant predictor in three of the four predictions. To test whether political reputation may function as a booster of already existing political predispositions, we applied a mediated model in which political reputation mediates the relationships between political interest at T1 and at T2 and between political efficacy at T1 and at T2, respectively (controlling for gender and immigrant status). The indirect effects of political interest at T1 on political interest at T2 were positive and significant in both cohorts (standardized coefficients: .03, p < .001 at age 13, .02, p = .003 at age 16), as too were the indirect effects of political efficacy at T1 on political efficacy at T2 (standardized coefficients: .02, p = .02 at age 13, .02, p = .003 at age 16). These analyses show that political reputation in class links youths' political interest and political efficacy at T1 to their political interest and political efficacy at T2. The results indicate that political reputation partly functions as a booster of youths' initial political predispositions over time.
Finally, the educational resources in the family were controlled for. To do so, we made use of the parental reports of their level of education and replicated all the analyses for the subsamples of youths with parental reports (n = 516 for the age-13 cohort, and n = 477 for the age-16 cohort). The results of this replication are reported in the Appendix S1 in the online supporting information. They largely resemble the findings reported above, with a few exceptions of effects that lost significance most probably due to the lower power.

Discussion
The classroom should be an ideal social setting for understanding how a youth's political reputation develops and the consequences of having such a reputation. Most obviously, the classroom is a social environment in which students meet almost daily. Over numerous observations they can identify their classmates' social identities and compare these with the impression their friends have formed of the same classmates. This kind of regularly repeated social interaction over time is a hallmark of theoretical ideas about social identity and the development of a personal reputation (Emler, 1990(Emler, , 2013(Emler, , 2019. Studying intact classes over time was a main reason for utilizing a one-year longitudinal design. Here, the students in one class at one point in time were about the same as the students in the same class one year later. This study addressed two basic questions about adolescents' political reputations in class: the predictors of this social reputation and the role of political reputation in the further development of political interest and political self-efficacy. Two cohorts of youths, 13-and 16-year-olds, were examined. First, very high one-year stabilities for political reputation in class were found. The students who had a high or low political reputation at the ages of 13 or 16 were much the same as those found to have a high or low political reputation one year later. This suggests that students already in early adolescence identify the classmates who are politically informed and verbal about political issues in class, and that, once acquired, reputations are not likely to change much. Apparently, political reputations, as perceived by classmates, consolidate very early, at least by age 13. This agrees with findings from analyses of the stability of other social behaviors (Aleva et al., 2017;Bukowski & Newcomb, 1984;Liu et al., 2014), and we interpret the findings as an indicator of the robustness over time of political reputation as a shared social construction.
Second, we had expected that particularly those students with high political interest would be likely to have high political reputation in class. The existing literature largely supports the idea that political interest is a prerequisite for and motivational factor behind numerous manifestations of young people's political engagement and participation (Prior, 2019). We also found that political interest was a significant predictor of political reputation in both the younger and the older cohort. Social fear was a negative predictor in both cohorts, and political efficacy was a significant positive predictor in the older cohort. Youths' self-perception of being active in class was also a significant predictor in the older cohort. These findings were based on cross-sectional analyses to understand why some, rather than other students in a class, have a political reputation. Utilizing a longitudinal design to predict changes in political reputation over one year, we found significant but weak contribution of social fear in both cohorts and of political interest and self-perceived political activities in the older cohort. The very high stability over time in political reputation seem to preclude additional explanations. It appears as once the political reputation in class is stabilized, it is not affected by the individual characteristics assessed in this study.
Third, we showed that those who have a political reputation in class are more likely to increase in both political interest and political efficacy over a period of one year. This is noteworthy, especially given that political interest has traditionally been considered as a predictor of different aspects of young people's political engagement, rather than as an outcome (Prior, 2019). We suggest that the increased political interest can be explained by youth with a political reputation striving to maintain their reputation by strengthening their interest in political matters further. Also, the awareness of being recognized as having a political reputation in class might strengthen their sense of efficacy in the broader political realm. These speculations await future research to better understand the group dynamics of political discussions in class to provide the mechanisms that link political reputation to increased political interest and efficacy over time. It should be added that mediation analyses showed that youths' political predispositions, political interest, and political efficacy at T1 were significantly linked to their interest and efficacy at T2 via the political reputation in both cohorts. This is an indication that political reputation partly functions as a booster of youths' initial political predispositions over time.
Several additional research questions arise from this first attempt to understand these group processes in the classroom. If a political reputation is an accurate reflection of the political visibility of a student, it is tempting to think of young people with a high political reputation in their classes as either already being opinion leaders or as potentially influential future political voices. Of course, this will depend on these students' political views and acceptance among others. This would be in line with research on reputations in work organizations, where reputations for high performance and helpfulness have been found to predict others' perceptions of these persons' career success and approval (Zinko et al., 2011). Contemporaneously, though, if they are viewed as opinion leaders in class, the number of their committed followers is limited. Only one out of four students at age 16 said that they were "very" or "fairly interested" in politics. Indeed, about as many reported that they were "totally uninterested." If they are indeed opinion leaders, knowledge is needed about whether other students and friends (or adults in the family or community) are influenced by their views and arguments. This is unknown. Still, the increase in confidence in their ability to effect change in society, due to their political reputation, might be connected to a perceived positive reception from other people. There are many research questions about young students with a political reputation that require answers. The most central is the long-term consequences of having a political reputation.
Gender differences need to be considered. The strongest predictor of a political reputation in class was political interest. Political interest was not correlated with gender. Despite this, fewer girls than boys in both cohorts had a high political reputation. This might be due to gender differences in fear of speaking in public, which was more common among girls than among boys. Thus, one reason for girls having a lower political reputation in class than boys is that the social fear of speaking in class to a greater extent discourages girls from participating in classroom discussion about societal issues. However, removing social fear in the regression equation in Table 2 does not affect the impact of gender on political reputation in the two cohorts. This suggests that social fear is not an explanation for why girls have lower political reputations than boys. Other explanations are needed.
A person' perception of himself/herself as highly politically active in class represents an insider perspective, whereas for a person to have a reputation among classmates for being politically active in class is an outsider perspective. We included in the study self-perception of one's political activity in the class as a comparison with reputed political activity. The correlations indicated that in the younger cohort they were weakly associated (r = .07, n.s.) but moderately associated (r = .29, p < .001) in the older. Only in the older cohort was youth's self-perceived political activities in class a significant predictor of political reputation in the cross-sectional analyses. In contrast, whereas political reputation predicted significant increases in political interest and political efficacy over time, self-perceived political activity in class did not predict increases in political interest and political efficacy over time. That political activities in class affect students' political interest and political efficacy not primarily through the filter of each individual student's perception underlines the role for political reputation to be of interest for future research and for broader theories about the political development of youths.
Finally, it seems that students' political reputation in class is not a proxy for high SES. As reported in the Appendix S1 in the online supporting information, the correlation between the youths' political reputation and parents' educational level was only .11, p = .01 in the younger cohort and .06, n.s. in the older. Further, analyses not reported in this study show that separate cross-sectional predictions of political interest and political efficacy from political reputation and parents' educational level indicate significant effects of political reputation for both outcomes in the younger and the older cohorts (betas ranged from .20 to .34, all significant at the .001 level), whereas parents' education was nonsignificant in all cases. Tentatively, what happens in classrooms where some students gain a political reputation might compensate for an unfavorable socioeconomic background.
Why is it important to understand youths' political reputations in their classrooms? At the broadest level, human development is shaped by the interactions between individual characteristics and the everyday settings people encounter, at different levels of generality (Bronfenbrenner, 1979). More narrowly, the understanding of why some young people become interested in politics and society while others do not requires attending to experiences in everyday settings which may increase or thwart early political and societal interest. Even more narrowly, this study has asked what causes some students to gain a political reputation in their classes and whether this reputation has consequences for their subsequent political interest and efficacy. Apparently, having a classroom reputation for taking an interest in societal issues increases a young person's political interest and efficacy. This is the short-term scenario. What is lacking is evidence of the longer-term consequences. This is still much unknown, though. However, having documented that young students' political reputations in their classrooms affect their political interest and political efficacy provides a starting point for placing the person-by-setting interaction on the agenda for understanding how young people's political engagement is shaped in their everyday life settings.

Strengths and Limitations
The main strength of this study is that we have introduced political reputation in the classroom into research on political socialization and have outlined individual characteristics which affect this reputation and the consequences that this reputation has in the short-term. Most notable is the confirmation that students' political reputation increase their political interest and political efficacy over time. The findings reported in the study were remarkably similar for the two age cohorts and were found to be robust even after control for parents' level of education based on a subsample of our participants.
A limitation is that this study was conducted in one city in Sweden. Although the sample was broadly representative of the country as a whole on major social indicators (proportion of immigrants, income, level of education, unemployment rate, and voting turnout), this does not preclude the possibility that findings might be different in samples of young people in other countries. This calls for cross-national replication and validation.
Another limitation is that we have assumed that students are quite aware of their political reputations and that this awareness spurs their future political interest and efficacy. We have no direct empirical evidence to verify this awareness in the study, but earlier empirical studies suggest that people tend to be aware of their reputations especially if this concerns the persons they affiliate with most (Emler, 1990), and experimental studies suggest that in group activities people who come to know that their reputations differ from others tend to change their behaviors in line with these others (Feinberg et al., 2014;Piazza & Bering, 2008).

Concluding Remarks
This study focused on the results of psychological group processes in classrooms and how they affect students' political development, and it shows that a sorting process appears to go on in classrooms whereby some students gain a reputation among classmates as politically knowledgeable and verbal. These political reputations in class seems to form very early, at least by age 13. The main message is that having a reputation among classmates for being politically knowledgeable and verbal in class seems to have future consequences for students' political interest and political efficacy at least in the short term. Reputations positively contribute to increases in both. We believe that these findings broaden the way of looking at the influence of schools on adolescents' political development. Studies are needed to identify the microprocesses in classrooms that lead to an establishment of a political reputation in this setting and understand the mechanisms linking a political reputation to increased political engagement over time. Knowledge is also needed of the longer-term implications of political reputation among classmates for political development during the adolescent into adult years.

Supporting Information
Additional supporting information may be found in the online version of this article at the publisher's web site: Appendix S1. Control of Educational Resources in the Family