The Consensus on Citizenship Education Purposes in Teacher Education

: Although education for democratic citizenship has long been a powerful rationale for social studies education, researchers still report a signiﬁcant gap between this purpose and what is really taught in classrooms. Explanations of this phenomenon vary, but literature on citizenship education (CE) research has largely interpreted this gap as a result of (preservice) teachers’ political worldviews or lack of civic experiences. Other evidence, however, suggests that teacher socialization processes generate conventions about what is necessary, possible, and reasonable in CE that go beyond teachers’ political views and behaviors. This mixed-method study, developed at a Spanish university, aims to explore the understandings of CE shared by preservice teachers with different political ideologies and levels of civic engagement. The ﬁndings of this study have deep implications for teacher education courses aimed at fostering CE and the curricular inclusion of current social issues.


Introduction
Although education for democratic citizenship has long been a powerful rationale for social studies education, researchers still report a significant gap between this purpose and what is really taught in classrooms (e.g., Bickmore 2014;Evans 2006;Sant 2013). To better understand the origin of teachers' reluctance toward including this purpose in the social studies curriculum, several scholars have explored teachers' perceptions on citizenship education (CE) (e.g., Brownlee et al. 2016). Studies have found that many teachers hold shallow understandings of CE and are not usually well-prepared to deal with the difficulties of engaging students in current social issues (e.g., Davies et al. 1999;Knowles and Castro 2019;Lee and Fouts 2005;Patterson et al. 2012). Similarly, studies with preservice teachers have shown that the majority of preservice teachers tend to confuse CE with moral and character education (Carr 2006;Marri et al. 2014;Martin 2010), favoring personally responsible views of citizenship (Westheimer and Kahne 2004). Some scholars have suggested that this is due to both in-service and preservice teachers having little experience in political participation (e.g., Sim et al. 2017) or holding certain political ideologies (e.g., Castro 2013; Knowles and Castro 2019). Other evidence, however, suggests that teacher socialization processes (Bullough 2000;Lortie 1975;Su 1992) generate conventions about what is necessary, possible, and reasonable in relation to CE that go beyond political ideologies and levels of civic engagement.
The present study delves into the CE beliefs shared by a group of Spanish preservice teachers with different political views and levels of civic engagement. The study is a continuation of a previous research focused on the teacher educators of the participants (Estelles and Romero 2019) where we found generalized, idealistic views of CE and difficulties in bringing the purposes of CE into teaching practices. The findings of this study have

Teachers' Perceptions about Citizenship Education
The gap between the critical purposes of CE and CE teaching practice has led many educational scholars to focus on teachers' perceptions of CE (e.g., Marri et al. 2014;Martin 2010). The premise is that teachers' views on citizenship, democracy, or civic life often condition their subsequent educational practices. Several studies have found quite superficial understandings of these concepts and have interpreted this result as a serious handicap for committed CE (e.g., Castro 2013;Fry and O'Brien 2015;Marri et al. 2014;Patterson et al. 2012;Ross and Yeager 1999;Sunal et al. 2009). These studies attribute this handicap, at least partially, to certain political ideologies (e.g., Castro 2013) and/or to a general lack of civic participation experiences (e.g., Carr 2008;Sim et al. 2017).
Some studies have also explored the above relationship and suggest that teachers' political ideology and experiences have some impact on CE practices. For example, Rogers and Westheimer's (2017) quantitative study concluded that, in North America, teachers' level of political engagement is positively correlated with the frequency with which they teach about social problems such as economic inequality. Yet, they did not find any relationship between this frequency and teachers' political ideology. The mixedmethod study developed by Hess and McAvoy (2015) in the United States revealed a connection between the pedagogical approach and teachers' political views. This result was also obtained by the variable-centered analysis conducted by Gainous and Martens (2016) of the CivEd data from civics teachers from the United States. Using qualitative methods, Schugurensky and Myers (2003) interviewed a group of North American civic education teachers about their most meaningful learning experiences for teaching CE. The authors concluded that political experiences provide affective and cognitive support for teaching CE. Additionally, Kenyon's (2017) narrative inquiry demonstrates that the manner in which social studies teachers conceive of their relationship to the government structures "impacts the way they teach their students both about and through citizenship" (pp. 94-95).
Other evidence, however, also suggests that to understand teachers' dispositions and practical knowledge about CE, the focus cannot be exclusively on their political socialization, whether it occurs "by default" or "by design" (Lopes et al. 2009). It is also necessary to take into account that both the "grammar of schooling" (Tyack and Tobin 1994) and the "subcultures" of each school subject (Cuesta 1997;Goodson et al. 1998) implicitly delimit what is necessary, possible, and reasonable in relation to CE. As soon as CE crosses the school doors, CE is trapped and shaped by this field of forces and, often, diminished in its scope and pretensions (García and De Alba 2012). Within this framework, it is not surprising that teachers feel more confident in teaching CE when it is a distinct school subject (Alviar- Martin et al. 2008), they rarely prioritize political participation among the main aims of CE (Reichert and Torney-Purta 2019), and they frequently avoid teaching about controversial issues in the classroom (Byford et al. 2009;Ho et al. 2017;McAvoy and Hess 2013).
When asked, teachers who prefer not to introduce current social issues in the classroom argue that it is due to its highly politicized nature and the age of their students (McAvoy and Hess 2013, p. 34). The beliefs underlying these reasonings include both that the teacher must be neutral-leaving out ideological interpretations-and that students are too naïve and manipulable. The illusion of neutrality in education (Apple 2008;Ross 2017) and protectionist views of childhood innocence (Wyness 2006) are often present in both liberal and conservative political views (Estelles and Romero Forthcoming). These types of assumptions usually go beyond the explicit ideology of teachers and are part of what Bourdieu (1999) calls the doxa. These assumptions underpin our own common sense (Wagner et al. 2011) and play a key role in our "practical knowledge" (Bourdieu 2007).
To reduce the gap between the critical purposes of CE and CE teaching practice, we need to understand the conventions of CE that go beyond the explicit will and intentions of teachers (Estelles and Romero 2019;Bougher 2014;Fischman and Haas 2012). For this reason, the present study aims to explore the CE understandings that remain beyond ideological discrepancies and civic engagement. This study was conducted with preservice teachers enrolled in a Spanish public university, and it addresses the following key questions:

•
What are preservice teachers' political ideologies and levels of civic engagement? • Which understandings of CE are shared by preservice teachers with different political ideologies and levels of civic engagement?

Research Design
This study used an explanatory sequential mixed-method design, which implies "collecting and analyzing first quantitative and then qualitative data in two consecutive phases within one study" (Ivankova et al. 2006, p. 4). The aim of the quantitative phase was to provide a general understanding of preservice teachers' political ideology and civic engagement (i.e., to answer Research Question 1) and to select a smaller sample for a second qualitative phase aimed at deeply exploring the understandings of CE that seem to elicit more consensus among a group of participants with different political ideologies and levels of civic engagement (Research Question 2).

Study Context and Participants
The study was conducted in a Spanish medium-sized public university that offers early childhood education and primary education teaching programs. An independent sampling approach for each phase of the study was used (Curry and Nunez-Smith 2015).
For the quantitative phase, participants were selected using a simple random sampling. Microsoft Excel software was used to conduct the random selection. The inclusion criterion used was part-time or full-time enrolment in the early childhood education or primary education teaching programs. The target population of the study was 1335 students enrolled in the previously mentioned teaching programs during the 2017/18 academic year. In order to have sufficient statistical power to detect the associations of interest, a sample size of n = 299 was obtained from considering a 95% confidence level, a precision of 5%, and a variance of 0.5. After determining the sample size, a 10% loss rate was applied, resulting in a final sample size of 334 participants. Finally, a total of 324 student teachers agreed to participate in the study.
For the qualitative phase, as the aim was to explore common understandings of CE among preservice teachers with different political ideologies and levels of civic engagement, the authors selected a purposeful sample from the quantitative strand. Participants from the different categories related to their political ideology (liberal, moderate, and conservative) and level of civic engagement (low, moderate, and high engagement) were chosen. Selection criteria were also established to include preservice teachers from both teaching programs (early childhood education and primary education), all levels (from first-year to fourthyear students), and a representative proportion of sexes. The total number of preservice teachers who participated in the qualitative strand (n = 27) was based on the achievement of saturation (O'Reilly and Parker 2012).

Instruments for Data Collection
The instrument used for the quantitative phase was an online questionnaire that included questions about preservice teachers' background information, self-reported political ideology, and civic engagement. This questionnaire was an adapted version of the questionnaire used by Rogers and Westheimer (2017). Following Rogers and Westheimer's (2017) work, participants' self-reported political ideology was assessed by the question: 'How would you characterize yourself?' (very liberal, somewhat liberal, moderate, somewhat conservative, or very conservative). As measured by Rogers and Westheimer (2017), civic engagement was a composite variable that considered participants responses to questions about the frequency with which they follow political news, talk about politics with family and friends, and participate in organizations that aim to make a difference in their community or broader society: How often (never, once or twice, monthly, weekly, or a few times a week/daily) have you: • (Civic Item 1) followed news by reading a newspaper or news magazine, watching national news on TV, listening to news on the radio, or reading news online? • (Civic Item 2) talked about politics or government with your family or friends? • (Civic Item 3) participated in an organization that tries to make a difference in your community or broader society?
For the second phase of the study, qualitative data were collected through visual interviews. We used photo-elicitation techniques to elicit intuitive beliefs about CE. The visual images used during the interviews were researcher-generated (Prosser 2007;Richard and Lahman 2015). Preservice teachers were required to view and comment on the images and sentences that corresponded to each dimension/purpose of CE considered by the Eurydice Reports (see Table 1). Cultural and historical heritage Children dressed up with traditional costumes in a Spanish regional festivity Cultural and linguistic diversity of society A school celebrating the Day of the Nations

Critical Thinking and Analytical Skills
A student in a class asking her teacher to teach them to think, not to memorize

Attitudes and Values
Respect A poster with the rules of a classroom Solidarity Children helping each other in class Learning to listen and resolve conflicts Two children angry with each other

Active Participation
Practical experience of democracy at school Young children discussing as a group in a classroom Involving students in the community at local, national, and international levels Children taking part in a demonstration against racism and xenophobia We selected images and sentences that directly appealed to the common sense understanding of CE-frequently in the form of stereotypes-based on previous research on the topic. For example, we used a photo of a school celebrating the Day of the Nations to capture the participants' ideas about the dimension of "respect for cultural and linguistic diversity", as in Spanish schools this goal is often reduced to a mere stereotyped representation of the dominant cultures of other countries (Torres 2011). We also included a photo of children dressed in traditional and folk costumes of the Autonomous Community where they live to trigger the participants' beliefs about "the promotion of the recognition of cultural and historical heritage", because these folkloric symbols have become useful for the Autonomous Communities in Spain to promote territorial identities (Martín 2001). For the dimension related to the teaching of political issues in the classroom, we chose to use the sentence "politics is very important for a citizen, but it might be too complex and conflictive to address with children at schools", and a vignette of a teacher with a burned head whose doctor is asking if they had dared to discuss controversial issues in the classroom. These texts contain arguments that teachers often use to avoid teaching political issues in the classroom (Hess and McAvoy 2015;Ho et al. 2017).

Quantitative Data Analysis
Descriptive statistics were used to describe the main characteristics of the sample, the participants' self-reported political ideology and the prevalence of low, moderate, and high civic engagement. Verification of normality of quantitative variables was performed with the Shapiro-Wilk test. The chi-squared test was applied to assess whether there were significant differences between males and females in the following variables: year level, civic engagement, participation in organizations, following the news, and talking about politics. This test was also used to check whether there were differences in the distribution of civic engagement among different teaching programs and ideological categories. Fisher's exact test was employed to examine differences in civic engagement between males and females and among different year levels. The Student's t-test with Welch's correction was employed to assess whether age differences existed between sexes. Finally, a logistic regression model was developed to identify the contribution of some variables to civic engagement. Adjusted ORs and their confidence intervals were calculated for the included variables. Statistical procedures were conducted using the R 3.3.1 software along with R Studio.

Qualitative Data Analysis
The qualitative data analysis was iterative and included both inductive and deductive analysis (Bogdan and Biklen 2007). We used Atlas.ti software (ATLAS.ti Scientific Software Development GmbH, Berlin, Germany) for data codification. Analysis began by open coding the visual interview transcriptions. Then, we looked for categories, patterns, and themes related to CE across participants with different political ideologies and levels of civic engagement. In particular, we focused on the reasons and rationales that the participants provided to justify the importance of CE and its different dimensions. For this process, we drew upon the different dimensions of CE described by the Eurydice reports (Eurydice 2012(Eurydice , 2017 and the framework developed by Carr (2008Carr ( , 2011 that distinguishes between 'thin' and 'thick' approaches to education for democracy (see Table 2). This approach to qualitative data analysis based on the themes that emerged from the data and the relationship between the data and pre-existing categories provided us with a deep understanding of how preservice teachers articulate their CE beliefs, increasing this study's internal validity (Maxwell 2013).

Critical Thinking and Analytical Skills
The ability to distinguish between fact and opinion The capacity to detect power imbalances

Active Participation
Practical experience of democracy at school Being consulted and informed (Hart 1992, p. 12) Involving students in decision making (Hart 1992, p. 14) Involving students in the community at local, national, and international levels

Ethical Considerations
Ethical issues were carefully contemplated in this study due to the sensitivity of the topic. Participants were informed about the objectives and procedures of the study and how their rights were going to be protected. Participation in the research was voluntary and anonymous. The preservice teachers who freely agreed to collaborate in the study signed an informed consent form and they were asked for feedback on the transcriptions of the interviews. The study has the ethical approval of the Ethics Committee of the University of Cantabria [Project No. 07/2017].

Quantitative Results: Political Ideology and Civic Engagement
The main characteristics of the sample and quantitative results are described in Table 3. As can be seen in this Table, 268 females (82.7%) and 56 males (17.3%), aged between 17 and 44 years old, participated in the study. Participants were evenly distributed among the four years of the primary education (57.1%) and early childhood education (42.9%) teaching programs. As can be seen in Table 3, most participants had a moderate ideology (45.5%; 95% confidence interval [CI]: 36.9-47.7), followed political news every day (39.3%; 95% CI: 29.9-40.2), talked very infrequently about politics (39.3%; 95% CI: 37.9-48.7), and had participated once or twice in a social organization (55.4%; 95% CI: 58.3-68.8). There were no significant differences between males and females in any of these cases.
In relation to the distribution of civic engagement, most participants showed low civic engagement (76.8%; 95% CI: 76.3-84.9). Forty-one participants had a moderate civic engagement (16.1%; 95% CI: 9.6-16.9), and 20 participants had a high civic engagement (7.1%; 95% CI: 4.1-9.5). No significant differences were found between male and female civic engagement (p = 0.578). Civic engagement was not age-related either (p = 0.588). There is, however, a correlation between year level and civic engagement (p = 0.032); there were significantly less participants with low civic engagement in the third year than in the other year levels (see Table 3). A relationship between teaching program and civic engagement was also found; high civic engagement was recorded at seven percentage points higher in the primary education program than in the early childhood education program (2.2% vs. 9.2%; χ2 = 5.46, df = 1, p = 0.020). Finally, self-reported political ideology and civic engagement were not related (p = 0.233). Yet, the greatest engagement was found in those participants who self-identified as very liberal (36.8%), and the highest prevalence of low and moderate civic engagement was found among those participants that described themselves as politically moderate (45.1% and 34.1%, respectively).
The results obtained with the logistic regression model are presented in Table 4. This Table shows the adjusted odds ratios (ORs) of the included variables and their confidence intervals. The analysis reveals that those participants who were enrolled at the primary education program are 3.6 times more likely to have a high civic engagement than participants from the early childhood education program. Compared to the participants enrolled in the first year, third-year participants are 3.1 times more likely to have high civic engagement. Additionally, participants identified as conservative are four times less likely to have high civic engagement compared to the participants who described themselves as liberal.

Qualitative Results: The Consensus about Citizenship Education
When reflecting on the photographs, the interviewed preservice teachers mentioned a broad range of reasons to justify the need for CE. All the preservice teachers agreed that CE should be an essential school mission. The reasons, which are summarized and exemplified in Table 5, included the need to be informed, preserve traditions, obey laws, respect diversity, and participate in decision-making, among others. As can be seen in Table 5, the reasons mentioned by the preservice teachers could be categorized into the four dimensions of CE described by the Eurydice reports (Eurydice 2012(Eurydice , 2017. However, the attention given to each dimension was variable. The most highlighted dimension was the dimension related to attitudes and values. Some aspects of the political literacy dimension such as the knowledge of local and regional traditions were also highly regarded. However, other aspects of this dimension such as the teaching of current social issues or civic institutions did not arouse much enthusiasm among the preservice teachers interviewed, nor did the critical thinking skills dimension that was only mentioned by one participant. The references made to the participative dimension were mainly focused on student participation in the classroom and the importance of using active teaching methods and democratic processes to encourage this participation.

CE Dimensions
Reasons for Advocating CE

Knowledge About Civic Institutions
Quotes mentioning the importance of teaching children about civic institutions to value their role in a democratic society.

Political and Social Issues
Quotes advocating for the inclusion of current political and social issues in the classroom to engage children in their potential solution.

Cultural Heritage
Quotes refering to the importance of children knowing about cultural traditions and symbols to preserve their heritage.

Obedience/Responsibility
Quotes pointing to the importance of children learning to respect social norms and taking responsibility for their own behavior.

Tolerance/Respect
Quotes indicating the importance of children being respectful to others.

Respect for Cultural and Linguistic Diversity
Quotes pointing to the importance of children being respectful to other cultures and languages.

Conflict Solving/Discussion
Quotes highlighting the importance of children learning how to discuss ideas and peacefully solve discrepancies through dialogue.

Provide Reasons for Their Ideas
Quotes refering to the importance of children reflecting on their own ideas and being able to defend them.

Avoid Manipulation
Quotes mentioning the importance of developing children's analytical thinking skills to avoid manipulation.

Participation in Classroom Management and Organization
Quotes highlighting the virtues and/or the importance of making children participants of the classroom management and organization.

Participation in Social and Political Activities
Quotes highlighting the virtues and/or the importance of engaging children in social and political activities that occur beyond the school walls.
From the analysis of the rationales provided by the participants to justify these dimensions, four common themes were identified in the interviews of preservice teachers with different political ideologies and levels of civic engagement: (a) values education at the core of CE, (b) cultural heritage as the main goal of CE, (c) political literacy education through teaching factual events, and (d) children unable to think critically and participate.

Values Education at the Core of Citizenship Education
As mentioned before, the most highlighted dimension of CE was attitudes and values. This dimension was highly regarded by most of the preservice teachers interviewed (24 out of 27), especially in the following aspects: learning to behave responsibly (23 participants), strengthening a sense of respect and solidarity (19 participants), and learning to resolve conflict peacefully (16 participants). Accordingly, the words values, respect, tolerance, solidarity, and responsibility were frequently raised in the interviews. References to the issue of living together (consensus, dialogue, social norms, etc.) were also abundant, especially when they were exposed to the image of a school celebrating the Day of the Nations. Most of the interviewees (24 out of 27) defended the importance of values education while criticizing teacher-centered instruction based on memorization of factual information. They argued that CE involves something else other than the mere transmission of civic knowledge; that is, the promotion of certain values and attitudes. This idea was reflected through many expressions: "Of course, it's important to know what a democracy is and what the Congress does, but there is no point in teaching all that if they [students] aren't tolerant, don't respect the norms, and don't care about others". During the interviews, only one participant emphasized the importance of civic knowledge: Here I see the Congress and the Senate and . . . Well, I believe CE is necessary because it helps children to know about all these institutions where important decisions are made. It is necessary because, unfortunately, we have no idea of how they work.
Regardless of the participants' political ideology, they tended to idealize the power of values education through references to the future and social salvation. As Sandra (conservative and high engagement) and Maria (moderate and low engagement) stated, respectively: It seems that people do not realize that for solving the problems that our country has, it's necessary to educate children in values because if you don't educate them from childhood, then it is very difficult to change them, unfortunately.
Education is everything because the children you educate will be soon the future of the country . . . You must educate them from early childhood and teach them values. If everyone had a set of common values of respect, tolerance, etc., there would be no terrorism, hatred, none of that.
Even for the most liberal and highly engaged participants, when firstly described, CE was frequently associated to values: "for me, educating for citizenship means educating in democratic values such as respect, participation, and solidarity" (Carlos, liberal and high engagement). The specific values the participants supported differed (i.e., conservative participants tended to put more emphasis on honesty and law-obedience, while liberal participants usually focused more on respect and solidarity), but the redemptive narrative remained beyond ideologies. The idea of moral education as an essential mission of schools was not questioned by any preservice teacher, and only one participant acknowledged its limitations when they recognized that critical thinking skills were also necessary to exercise citizenship.
Additionally, the issue of values was generally related to the ideal of convivencia 1 -living together in harmony (Hernando-Lloréns 2019), understanding CE as a practice of peacekeeping. For this reason, terms such as consensus, dialogue, tolerance, respect for social norms, and harmony were abundant in both liberal and conservative participant interviews. For example, note the following quotes from participants with opposite ideological positions: I think CE is necessary because students are part of an educational community with teachers, families, etc., and it's important to value that coexistence and to comply with established norms. Values are necessary to live together in harmony. (Francisco, conservative and low engagement) I believe that CE also involves students and the teacher agreeing on rules of coexistence through dialogue and taking responsibility for compliance. (Ana, liberal and high engagement).
When discussing how to bring values education into teaching practices, the participants made some generic references to the use of active teaching methods to promote student participation in the classroom: To teach citizenship and values, students need to be active and to participate in the classroom. We cannot teach them through direct instruction, giving lessons with textbooks and doing exams. (Alberto, moderate and low engagement).
In CE children need to learn how to respect each other's turn to speak, to respect their partner, to respect their partners' opinions . . . Children can only learn that if they participate in class and have opportunities to speak up. (Ana, liberal and high engagement).
Despite this emphasis on participation, most preservice teachers did not interpret the image of a group of children arranged in a circle/assembly with some hands raised as a space of participation, but rather as evidence of good manners.
Beyond political discrepancies, the preservice teachers interviewed also generically mentioned the importance of the teacher acting as a good role model in values education and CE: To teach all those democratic values, the teacher must, first of all, teach through example. You cannot ask your students to be respectful if you are not respectful with them. (Alejandra, moderate and high engagement).

Cultural Heritage as the Main Goal of Citizenship Education
Another common theme among participants with different ideological positions and levels of civic engagement was the defence of teaching of the cultural and historical heritage of the local region. Most participants (25 out of 27) agree that CE should promote the knowledge of local and regional traditions. The photo of some children dressed up with traditional costumes in a regional festivity inspired feelings of tenderness and aroused enthusiastic reactions among all participants: "it's very nice to see children participate in these activities because it is a shame that these customs get lost and that they forget where they come from" (Pilar, conservative and low engagement). For some, this image evoked memories from their childhood and a bucolic image of traditional country life: I remember when I was little, and we prepared the town festivities at school, and everyone in town helped us with the costumes. Everyone participated in some way because we all knew each other. It is a shame that children in cities do not know this. So I think CE should try to rescue all this. (Carlos, liberal and high engagement).
Others explicitly recognized that the image accurately represents what CE means for them, with exclamations such as: "for me, this is really educating for citizenship", "that's what CE is about, right? Knowing where we come from, our roots, to know who we are" and "in the end, I think this must be the main purpose of CE, that we learn to appreciate where we come from and worry about taking care of it". As can be observed in these quotes, there is a generalized admiration for the culture and traditions of the region. This admiration not only caused that the stereotyped image of country life to not raise any criticism among the participants, but also the construction of historical knowledge and the goals of CE to be subject to the construction of territorial identities. "Knowing our roots to know who we are" becomes here a more important goal than prior invocations to respect, tolerance, and convivencia. Not surprisingly, diversity at this point of the interview was totally forgotten.
The image of the children wearing traditional costumes was shown a few photographs after the image of a school celebrating the Day of the Nations depicting a very stereotyped representation of the dominant cultures of other countries. As explained above, the image of a school celebrating the Day of the Nations also aroused admiration among the participants. All preservice teachers seemed to agree with activities such as celebrating the Day of Nations, as they promote "respect for other cultures" and "the relativization of our own cultural customs". However, no tensions were perceived between the preservation of the regional culture and respect for diversity and the issue of living together. The generalized reaction to both photographs shows that the preservice teachers interviewed hold stereotyped and essentialist views of human cultures.

Political Literacy Education through Teaching Factual Events
The debate around the political literacy dimension was mostly focused on whether to include political and social issues in the classroom. In this debate, only four (out of 27) preservice teachers advocated for including political and social issues in early childhood and primary schools. This opinion does not seem to be associated with any particular ideology: two of these preservice teachers were liberal, one moderate, and one conservative. However, all of them were highly engaged in politics (although not all the participants who were highly engaged shared this belief). The reasons these participants provided to support their opinion ranged from the need to be informed about current issues through to the importance of increasing youth interest in politics to keep democracy alive. As one of the participants stated: I don't know why CE is not used to talk more about what is happening in Catalonia, for example, and less about the Constitution, the laws, and this kind of thing. We have no idea about that (Catalonia's independence issue), even though it is every day in the news, in the TV . . . and I don't know what to think about it! (Lorea, liberal and high engagement).
Most of the participants, however, did not advocate for the inclusion of this type of topic in the classrooms. As one of these preservice teachers said: I completely agree with this sentence (politics is very important for a citizen, but it may be too complex and conflictive to address with children in early childhood and primary education]. Politics is a very important issue nowadays, especially at older ages, as citizens. But recognizing that it is very important does not mean that it should be included in CE classes with such young children in early childhood and primary education. I think it's okay to teach it from secondary education. There were three main reasons provided by the participants who were reluctant to include political issues in the classroom: (a) that these issues are too controversial and, therefore, it is very difficult for a teacher to teach them from a neutral perspective, (b) that these issues are not suitable for young children, and (c) that these issues are too complex. We will further explain the second and third arguments in the next section. The first argument reveals a notion of knowledge as an "object" divorced from politics and, therefore, free from bias and suspicion. From this perspective, teachers should be neutral, leaving politics outside the classroom and teaching facts. For example, note the following claims of both liberal and conservative preservice teachers: I think we should know more about politics when we leave school, but, of course, it is very complicated because political issues are very contradictory, some say one thing and others another . . . So, yes, I think it's too conflicting. (Daniel, liberal and low engagement) It's true that there are issues in politics that are a bit tricky and that maybe it's better not to address . . . I think we have to teach them (students) what democracy is, how it works, what types of democracies there are, what are the differences between a dictatorship and a democracy . . . and all that, which is more objective. (Carmen, conservative and low engagement) Some preservice teachers with different political ideologies and levels of civic engagement also believed that political issues should be left out of the classroom because their highly emotional component impedes treating these issues as an "object to be taken in with minimal subjective interference" (Ross 2017, p. 11). As one of the participants claimed: "The problem is that these issues cannot be studied with a minimum of rigor because emotions are running very high. So it's better to treat them indirectly, making connections with other contents of the curriculum" (Alberto, moderate and low engagement). As another preservice teacher suggested: "You have to be very apolitical. If not, it is impossible to treat them from a neutral perspective" (Francisco, conservative and low engagement).
Two students also argued that these topics can cause conflicts and controversies that disturb the peace and stability of the classroom, setting students against each other: The problem is that these topics can accentuate the differences between some students and others. I believe that what we have to do is to generate mutual understanding, convivencia, and not conflict and disagreement. (Alejandra, moderate and high engagement)

Children Unable to Think Critically and Participate
One of the most commonly used arguments by both liberal and conservative participants for leaving political issues out of the classroom is that children are too young (innocent, manipulable, immature, etc.) to understand the complexity of political affairs. See, for instance, the following quotes: I believe that politics at this age is not important because it does not really influence them (children) and they are not aware of what is happening, they are too naïve. (Sandra, conservative and high engagement).
I think it (politics) is a very conflictive and complex issue for such young children, maybe not in secondary school . . . I don't know . . . Maybe it's better to let them enjoy childhood and not disturb them with this mud of adults. (Cristina, liberal and low engagement).
The metaphor used by this participant to describe "politics" is representative of the perceived opposition between the world of children (joyful, pure, and innocent) and the world of adults (political, stained, and muddy). This opposition also appears in other interviews made by participants with a different self-reported ideology: "I believe that politics is so corrupt, so complex that it is better to keep children away from it" (Maria, moderate and low engagement). Additionally, a simple/complex dichotomy related to the previous opposition of the child/adult world appears in these interviews. As one of the preservice teachers said: "I don't think we have to discuss politics with children because . . . what is politics really? what does it do? Children don't understand that" (Pilar, conservative and low engagement). Interestingly, many preservice teachers (15 out of 27) recognized that they do not feel comfortable dealing with political affairs because they do not have enough knowledge about it: "To be honest with you, I don't understand much about politics and I wouldn't know how to teach it in a primary classroom" (Alberto, moderate and low engagement). Whether liberal, moderate, or conservative, the majority of these preservice teachers had low levels of civic engagement.
The protectionist view of childhood can also be indirectly observed in the omission of children's social and political participation by most of the preservice teachers interviewed (25 out of 27). Only two participants reacted positively to the photograph of some children taking part in a demonstration against racism and xenophobia, and they were both liberal and high engagement. However, for two other interviewees with this same profile, the image aroused surprise and rejection. See, for instance, the following quotes: "This is a sad thing, it's sad that children have to go out and demonstrate because adults don't know how to solve these problems, it's a shame, it's very cruel" (Ana, liberal and high engagement); "I am angry to see that children have to go to the streets to demonstrate because we don't do it. They shouldn't have to do this, you know?" (Carlos, liberal and high engagement).
These participants' disagreement with children's involvement in public demonstrations contrasts with their claims for more student participation in the classroom and school life. For example, when they were shown the image of a group of children discussing in a circle in a classroom, these participants mentioned the importance of using active teaching methods and democratic processes to encourage student participation: "In this image, I can see an assembly where children are deciding, probably, on the rules of the class. I like this because it allows them to debate, participate, and commit more. Schools should offer more opportunities for democracy" (Carlos, liberal and high engagement).

Discussion
Although there are a few studies that delve into the relationship between teachers' political ideologies and their dispositions toward different dimensions of CE (e.g., Knowles 2018;Rogers and Westheimer 2017), a general picture of both preservice and in-service teachers' political ideologies and levels of civic engagement is still lacking. The ideology of the participants of this study is similar to that found by the Spanish Youth Institute (INJUVE 2017) for the population of young Spanish people. The percentage of participants with a moderate ideology (42%) is only three percentage points lower than their age group average, the proportion of liberal (38.2%) is five percentage points higher, and the presence of conservative (16.4%) is three percentage points lower. The frequency with which the participants follow political news and talk about politics does not differ greatly from those of their generation either (INJUVE 2017, pp. 13-16). The low levels of civic engagement that the preservice teachers showed were not unexpected considering those of their age group (INJUVE 2017) and the results of previous qualitative studies (Castro 2013;Sim et al. 2017). As the data provided by the studies conducted by the Spanish Youth Institute do not distinguish between different ages, it is difficult to determine if the increase of civic engagement in the third year is due to the teacher education program or another age-related cause.
This article also suggests that preservice teachers with different political ideologies and levels of civic engagement share a set of beliefs about CE that are rarely questioned. This consensus about the meaning of CE consists of four main assumptions.
The first assumption is related to the importance given to developing values in CE. Aligned with other studies (e.g., De la Montaña 2012; Marri et al. 2014;Martin 2010), this study found that the majority of trainee teachers tend to associate CE with moral education. We agree with Westheimer (2016) that the moral discourse and its inherent focus on the individual may have detrimental implications for critical CE such as the depoliticization of social problems. The best example of this depoliticization is the preservice teachers' reluctance to both deal with political affairs in the classroom and promote children's political participation. The focus on generic individual virtues and values overlooks the importance of real-life political experiences in the learning of democracy (Biesta 2007;Knight-Abowitz 2008;Schugurensky 2010). Not surprisingly, the preservice teachers interviewed showed difficulties in giving examples of how to bring values education into teaching practices. This result was also found by Thornberg (2008), who warned that for the teachers interviewed in his study, it seems that "values education happens without the teacher thinking a lot about it" (1795).
Regardless of the preservice teachers' political ideology and engagement, the values education discourses were frequently related to the ideal of convivencia (living together in harmony). As Hernando-Lloréns (2019) explains, this ideal "has permeated educational policy and practices in Spain since the beginning of the democratic period in 1978" (2). What is worrying about the salvationist overtones that surround the discourses related to convivencia (Itçaina 2006) is that these discourses often present this idea as the solution to the "problems" and "conflicts" derived from diversity (Hernando-Lloréns 2019). For the sake of convivencia, conflict and controversy are avoided in the classrooms. Although this discourse is shared by both liberal and conservative preservice teachers, its benign and conflict-avoiding tones seem to correspond to the political trends that the quantitative results show as well; that is, a large presence of moderate students with low engagement. Therefore, any correlation between ideology and this assumption cannot be ruled out, not in individual terms but rather in the form of a dominant doxa (Bourdieu 1999).
The second most common denominator from the majority of the interviews was the generalized defence of teaching of the cultural and historical heritage of the region in CE. This advocacy made the participants overlook the increasingly multicultural realities of the region and even subordinate the goals of CE to the construction of territorial identities. As Martín (2001) explains, invocations to tolerance, solidarity, and peaceful coexistence permeate the intentions and objectives, but it is the logic of the territorial identity that permeates the decisions likely to determine the teaching practice (180). The idealized admiration for the culture and traditions of the region was held by both liberal and conservative preservice teachers. In this regard, it would have been interesting to have collected data about the nationalist feelings and preferences of the participants. In any case, their responses indicate a tolerant attitude toward cultural diversity, as other studies with Spanish teachers have pointed out (Cardona et al. 2010;Chiner et al. 2015;Wassell et al. 2018). Yet, they seem to hold what Smith (2009, p. 47) describes as a "tourist" view of cultures, as they do not go beyond the folkloric surface.
The third assumption refers to the neutrality of political literacy teaching, which implies not including current social and political issues in the classroom. Our findings confirm other studies' results conducted with teachers (Byford et al. 2009;Oulton et al. 2004) and preservice teachers (Misco and Patterson 2007) that indicate that both groups usually refuse to teach about current public affairs in order to preserve concord and neutrality, despite the importance of getting children involved in current political debates (Andersson 2016;McAvoy and Hess 2013). Beyond political ideologies and levels of civic engagement, school knowledge is deemed objective and neutral and, therefore, not political.
Finally, the fourth assumption refers to the view of children as not-yet citizens and unable to think critically and participate. For most preservice teachers interviewed, children should be freed from the uncertainty and anxiety that the complexity of political issues and participation generates. Beneath their responses, we observed what Wyness (2006, p. 120) described as an opposition between the innocent child world and the fully sophisticated adult world. This view of children as "not-yet citizens" can be observed through the little attention given to youth involvement in real political affairs. As other studies have previously highlighted (e.g., García and De Alba 2012; Ortega-Sánchez and Pagès 2020; Reichert and Torney-Purta 2019; Roberts et al. 2019), (preservice) teachers usually perceive the promotion of citizen participation as something very disconnected from their teaching goals.

Implications for Teacher Education and Further Research
The findings of this study suggest that the weight of political socialization in the formation of CE beliefs and dispositions toward including political issues in the classroom is less than what has often been considered by educational research. CE practical knowledge seems to be deeply influenced by other pedagogical conventions such as the neutrality of the teacher or the naturalized conception of school knowledge. This finding is particularly important for two reasons.
Firstly, the above result provides additional evidence to rethink the relationship between political ideology and educational beliefs, as this relationship might not be operating in the conventional linear sense. Educational beliefs might also be shaping political ideologies and behaviors. Unfortunately, we are still far from speculating why high political engagement was seven percentage points higher in the primary education teaching program than in the early childhood education. Further research is needed in this regard.
Secondly, this finding directly challenges teacher educators, as those pedagogical conventions are intimately related to our field of work: school socialization processes. In other words, the scope of teacher education increases if the critical review of the conventions acquired during school socialization processes play a greater role in the formation of preservice teachers' understandings of CE. To date, however, several studies have pointed out that teachers do not receive enough training to teach CE (Akar 2012;Barr et al. 2015;Castellví et al. 2020;Chin and Barber 2010;Oulton et al. 2004;Thornberg 2008). In our previous work conducted with the teacher educators of the participants of this study (Estelles and Romero 2019), we identified several difficulties in bringing the purposes of CE into teaching practices and dismantling the entrenched assumptions that preservice teachers hold about the curriculum. Therefore, much work remains to be done in teacher education for CE.
In summary, this study suggests that teacher education and research should more seriously consider other variables, different from political views and behaviors, that might be affecting teachers' resistances regarding CE and the curricular inclusion of social and political issues.

Conclusions
The study of preservice teachers' political ideology and levels of civic engagement has shown a large presence of moderate students with low engagement that reveals a need to repolitize teacher education programs. This repolitization, however, cannot only consist of promoting preservice teachers' social and political participation. As this study has also found, there are several understandings of CE that go beyond political ideologies and levels of civic engagement. These shared understandings critically affect preservice teachers' attitudes toward including social and political issues in the classroom. Teacher education courses aimed at fostering critical interpretations of CE should help preservice teachers dismantle these common beliefs that might be part of what Bourdieu (1999) calls the doxa of the field.  Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.