THE HIDDEN CIVIC LESSONS OF PUBLIC AND PRIVATE SCHOOLS

Curriculum theory has long acknowledged the presence of a hidden curriculum in schools. Whereas the formal curriculum is explicit and documented, the hidden curriculum involves those attitudes, experiences, and learnings that are largely implicit and unintended. This article compares the hidden civic lessons found in public and private schools. Catholic and other private schools have measurable organizational strengths that socialize students into participation in public institutions more effectively than public schools.

H ave schools lost their capacity to provide an effective civic education for youth?Do public schools today instill democratic virtues, including the value of active participation in America's political institutions?Are schools able to solidify a long-lasting commitment to pursuing the common good in civic life?Some social commentators and public policy analysts have sounded the alarm about a rising tide of civic education mediocrity in America's schools (National Commission on Civic Renewal, 1998).Of particular concern is the apparent drastic decline in the younger generation's interest in politics, including sharp declines in voting among 18 to 25-year-old citizens (Keeter, Zukin, Andolina, & Jenkins, 2003;Putnam, 2000).Fingers have been pointed at civic education curriculum in public schools, and a fresh set of curricular standards and guidelines has been proposed for revitalizing civic education in America's classrooms (Center for Information and Research on Civic Learning and Engagement, 2003).
The problem with concentrating on civic education curriculum is that what is taught in class is only a small part of the learning process in schools (Dreeben, 1968;Jackson, 1968;Waller, 1965).The focus of this article is on civic education that is unintended and mostly invisible to participants, but nonetheless potentially powerful in shaping the civic attitudes and behavior of students.Nurturing democratic "habits of the heart" (Bellah, 1985) depends in part on how schools are organized and how schools as institutions are experienced by students.The civic education literature has not systematically addressed the role of characteristics of schooling organizations, such as routine practices in the classroom and school, which indirectly teach norms, values, and orientations to public life.This article investigates how school organization and culture per se may provide a civic education without intending to do so.These hidden civic lessons may affect how students conceive of engagement in public institutions later in life.
Even less apparent in the civic education literature is attention to whether and how school socialization varies by school sector in ways that may have long-term implications for civic life.This article argues that Catholic and other private schools have organizational strengths that effectively socialize students into active and constructive participation in public institutions.Though private schools of course do not represent "the public" in the same sense as public schools, the collective identity and normative climate in many private schools provide a socialization experience in which students practice sociability and civic skills, and learn the importance of placing the collective good above personal self-interest.These experiences, which may be particularly important since schools are one of the first and most comprehensive socialization settings outside the family, may affect whether students see their adult public involvement as a means to achieve individualistic and self-interested ends, rather than the pursuit of the common good in the public square, whether they expect to find (or work toward) solidarity or a struggle of all against all in institutions outside the family.In the following sections, I describe organizational differences across school sectors that may have significant long-term consequences on student orientations to public institutions.

SCHOOL SOCIALIZATION AND CIVIC EDUCATION
Civic education in schools is not only about teaching the historical development and workings of American democracy, but also about teaching democratic values and principles in the classroom.What do students learn about public life from their experience of school institutions?How do civic lessons vary across different types of schooling organizations?
Since the 1940s, if not before, public schools in American society played a major role in initiating youth into public life.In American culture, public schools are seen as one of the foundations of democracy; they are charged with teaching civic virtues and public purposes in a democracy (Meyer, Tyack, Nagel, & Gordon, 1979;Tyack & Hansot, 1982).Public school legitimacy depends on this role of linking family and public life through socialization of youth into democratic citizenship (Gutmann, 1987;Macedo, 1990;Nie, Junn, & Stehlik-Barry, 1996).
How are public schools to accomplish the task of democratic socialization?
One common notion is that public schools offer civics courses, which teach democratic virtues, the mechanics of democratic participation, and an appreciation of the historical struggle to build American democracy.Another way public schools provide democratic lessons is by uniting the diversity of a community in one institution, which builds on and reinforces the collective identity of the community, and teaches the importance of setting aside parochialism in the pursuit of the common good.The public school system, American culture tells us, is the institutional locus of the American melting pot.Or so it seemed.
One of the primary challenges to the idealized picture of public schools and civic socialization is the changing relationship between families, communities, and public schools (Epstein, 2001).Even earlier in the 20th century, some would argue, public schools were never quite this virtuous; in practice, the civic education of public schools was heavily slanted toward the interests of dominant fractions of the community, such as White Anglo-Saxon Protestants (Jorgenson, 1987;Ravitch, 1974).Whatever the historical reality, it is clear that social conditions have created new challenges for the civic role of public schools.Communities, or collective identities, are less bounded by geography, which makes it much more difficult for the public schools to function as an institutional expression of a geographically defined community (Coleman & Hoffer, 1987).Perhaps as a result of cultural pluralism at the community level, over time, public schools have come under political pressure from various pressure groups (Ravitch, 2000), which have made it necessary for schools to attempt to satisfy diverse special interests.This political context disrupts the expression of community through the public schools, and pushes public schools toward a bureaucratic organizational form in order to enforce neutrality toward and equality for each individual (Chubb & Moe, 1990).The combination of social change and restructuring of public schools on the local level raises questions about whether civic education socialization is possible in many of today's public schools.
The common conception of most private schools, and especially religious schools, sees them as having an inherent disadvantage in providing democratic socialization.First, their student body does not represent the entire population within a particular geographic boundary; they do not bring together the entire diversity of a community.They simply are not "public" in the literal sense of being funded by government money and of expressing a collective identity of a diverse geographic community.And religious schools are believed to be sectarian and overly authoritarian institutions, which further isolate students in religious "total" worlds (Peshkin, 1986;Rose, 1988).The basis of collective trust in these school organizations, according to some research, is best described as "organic," in which members of the school have unquestioning allegiance to school authority (Bryk & Schneider, 2002).If true, we would expect that religious school students are ill-prepared for democratic participation, or at least we would expect that they would be better prepared to participate in democracy if they did not attend religious schools.
There is reason to question this sharp dichotomy between the public purposes of public schools, and the private and sectarian interests of religious schools.Would democratic socialization of religious school students differ that substantially if they instead attended their neighborhood public school?Probably not.To begin with, the picture of religious schools as sectarian, counter-cultural institutions is overdrawn.Even conservative religious schools are a collage of Christian and American cultural strands; the most conservative Christian schools absorb much from the surrounding cultural milieu (Wagner, 1990).There is not an impenetrable boundary between religious schools and the surrounding community and culture.Outside of a small minority of fundamentalist schools, most Christian schools, even conservative Protestant schools, are not marked by "more or less unquestioning beliefs…in the moral authority of the particular social institution," where trust is given "unconditionally" (Bryk & Schneider, 2002, p. 16).This type of "organic" trust is made nearly impossible by the diversity of Christian traditions, the pervasiveness of teen culture in media and entertainment, the market forces that push religious schools to expand their potential customer base, and so on (Sikkink, 2001).Nor would we expect tight and harmonious connections between public schools and their geographic communities.Moreover, these ideal types-public schools, the source of democratic socialization and experience and religious schools, a "total" world focused on private ends-do not carefully consider the student experience of school as an institution, and the effect this may have on student orientations to public life.
What are the organizational characteristics of schools that foster effective democratic socialization?The answer obviously depends on how one defines what students need to learn to participate well in public institutions in their adult life.If we assume that students should experience a school as a place of collective mission and engagement, a place in which norms place individual self-interest subordinate to the collective good of the institution-analogous to ideal public institutions-then many public schools may not provide a social context conducive to democratic socialization.School climate may matter for what is learned about civic life, which may in turn affect how students think about public institutions and how to operate within them.

HIDDEN CIVIC CURRICULUM: LATENT CIVIC SOCIALIZATION IN SCHOOLS
In 1968, Dreeben took an innovative look at what is learned in schools.He focused on the "hidden" curriculum, the routine practices of classroom life that socialize students into adult roles-socialization that is "invisible" rather than explicitly taught (Brint, Contreras, & Matthews, 2001).What is learned in school, he argued, is not simply the three "Rs," but the fact that the norms and practices of life outside the family are very different than life in the home.
Children learn in school that the particularistic values of the home, in which personal relationships govern reward and punishment for example, would have to give way in public life to universalism, in which all children are treated interchangeably as members of a general category without regard for their particular characteristics and relationships.Relationships within a school organization are "specific" rather than diffuse: multidimensional relationships in the family are supplemented increasingly with relationships animated by a single purpose as children move through their school years.Children also learn that they will be treated as autonomous individuals rather than as persons embedded in families and communities, and they learn norms for individual achievement through recurrent evaluation in school (Dreeben, 1968).This type of school socialization has been dubbed the "hidden curriculum."Children learn through participation in a particular type of organization, which embeds principles of conduct and norms that govern relationships between various organizational roles.By learning these roles and principles, children are socialized into norms of behavior for public institutions outside the family, and, in many ways, Dreeben would argue, prepared to participate in work organizations and in the polity.These norms, expectations, values, and orientations are not explicitly taught by school personnel, but they are learned through the practice of schooling.
In a different approach to the importance of a hidden curriculum, Bowles and Gintis (1976) emphasized that what is learned in school are orientations that prepare students for hierarchical relations in the workplace.They argued that school organization serves to prepare working class kids for working class jobs, and middle class kids for management positions, which grant authority and demand creativity (Willis, 1981).Again, this socialization effect is not entirely intended, but is an outcome of the organization of schooling-and it may have implications for orientations to civic life.
What do the roles and principles of schooling organizations teach about participation in civic life?Dreeben would argue that youth gain expectations about what institutions outside the family are like, and how one is supposed to operate within them.In particular, students learn the values of universalism and individualism, and this prepares them for participating in a democratic polity (Dreeben, 1968).Dreeben argued that the success of the two-party democratic system required citizens who had learned and internalized the orientations of individualism, achievement, and universalism-just the orientations that Dreeben found in the hidden curriculum in schools.
We might ask whether school socialization has changed since Dreeben wrote about what is learned in school.Recent work by Steven Brint and colleagues (Brint et al., 2001) on the content of socialization within schools takes a comprehensive look at the value messages conveyed in teacher-student interaction, classroom materials, school rituals, and school organization.This work argues that the primary values learned in school are the importance of hard work and individual effort, orderliness, and, to a lesser extent, respect for others.Brint and colleagues agree with Dreeben that one of the key avenues of learning in schools is through routine practices embedded in classroom and school organization.As Dreeben found decades earlier, values of individualism and achievement remain a strong part of the hidden curriculum.But the hidden curriculum has expanded to include other forms of socialization implicit in the organization of the school and classroom.For example, token economies, in which approved behavior is rewarded with material rewards, bring the competitive individualism and commodification of the free market into the classroom.On the other hand, group projects in class offer opportunities to develop sociability and leadership skills.Activity centers and rotations increase the importance of variety and choice-which builds on and fosters cultural themes of expressive individualism (Bellah, 1985)-in the classroom experience.
The work of Brint and colleagues goes a long way toward confirming that routine school practices still constitute an implicit curriculum in public schools, and, though Brint and colleagues do not focus on civic education, they note that the hidden curriculum has implications for civic education.The organizational imperative for orderliness in public schools, they argue, leads to a narrowing of the definition of citizenship.Essentially, "good citizenship" is redefined according to organizational interest of public schools in order and conformity.Citizenship has little to do with civic participation and democratic virtues, but with not creating discipline problems in school.While "following the school rules" may have some relationship with respect for legitimate authority within other public institutions, the notion of citizenship is greatly impoverished when put in service of orderliness and conformity at school.While raising concern about whether public schools are providing an effective civic education, the broader point of interest is that organizational culture within schools shapes civic "lessons" in school.
The hidden curriculum remains powerful for civic education, but we must broaden our view of what specifically about schools affects civic orientations of students.Beyond learned values of universalism, the general experience of schooling institutions shapes how children view other public institutions, such as government, and shapes behavioral expectations within these public spheres.
Here we are diverging slightly from the specific claims of Dreeben's model regarding the fit between socialization in schools and participation in U.S. political life.First, it is not clear that Dreeben's focus on individualism and universalism are the only value orientations that are important for the success of U.S. democracy.These value orientations prepare Americans for a particular type of civic participation, one in which conduct in political institutions is centered in the struggle of competing (private) interests, with outcomes dependent on coalition formation and majority votes.This view tends to sideline other visions of a healthy democracy in which the common good and collective purposes have pride of place, and a struggle of pluralistic interest groups cannot achieve these higher collective concerns.In this view, persuasion and debate, social trust and personal sacrifice, are central "civic skills" for democratic participation (Verba, Schlozman, & Brady, 1995).Effective participation in civic life depends on developing civic skills, including social trust and sociability, and the willingness to set aside personal self-interest in public debate over the common good.This vision of a vital democratic life highlights the importance of a school experience that includes the practice of trust and sociability, as well as the experience of putting collective purposes above private interests.
Second, the increasing emphasis in public schools on the values of universalism and individualism may not have the positive effect on participating in the polity that Dreeben expected.A relatively low threshold, an organization that embodies these value orientations, foments student alienation from the schooling organization, which may shape student orientations to public life.Dreeben (1968) mentions this possibility, but sees it as either the experience of a minority or limited to normal adjustment as students learn how to behave outside the family.Social conditions have made the experience of alienation from school institutions much more problematic in public schools.
Third, achievement, individualism, and universalism are not the only orientations implicit in school organization that may have an impact on student orientations to public life.Dreeben misses the importance of collective identity formation at schools for the extent and type of civic participation in adult life.Much of Dreeben's focus is on teachers' activities in the classroom, which sanction particular norms for behavior outside the family.School-level processes come into play in his model, but are primarily limited to issues of class size, age grading and promotion, and evaluation.The discussion in the next section adds that the relationship of collective and individual identity formation must be considered to understand the relationship between school socialization and adult civic participation.

BEYOND UNIVERSALISM AND INDIVIDUALISM: HOW SCHOOL ORGANIZATION SHAPES CIVIC SOCIALIZATION
What are the particular characteristics of school organizations that constitute the hidden civic curriculum in schools?Several organizational characteristics of schools may affect student socialization in ways that influence views of civic life.The following attempts to identify characteristics of the social experience in school that are likely to shape a student's orientation to civic life.What most likely would not contribute to effective civic socialization is an experience of school as a collection of isolated and marginalized individuals or small cliques.A school organization that is experienced by most students as an alienating environment, in which students have no sense of belonging to the school, may provide a lesson about public life: that one should not expect to find solidarity and community when entering public institutions.This experience may increase cynicism about and mistrust of public institutions, and may dimin-ish the expectation that public institutions are about the pursuit of the common good.Of course, it may be the case that a sense of alienation from one's school does not have a profound and long-lasting effect on youth's orientation to civic institutions.But this alienating organizational environment would be a missed opportunity for youth to experience social solidarity and a "functional" community, that is, a community with dense overlapping social networks marked by normative consensus (Coleman, 1988)-key ingredients of a healthy civic life.
The hidden civic curriculum may be detrimental in a similar way if students experience school as a competitive and individual struggle for personal advantage.The notion that competitive individualism is the organizing principle of public institutions, such as the market and politics, may be reinforced in several ways within schools.A limited example is the extent to which schools are experienced as a physical struggle for power-between students as well as between students and teachers.When students feel unsafe in school, this experience reinforces views of public institutions as a necessary evil, in which participation is not about responsible citizenship, but individual survival.
A more direct example is the type of school that is organized around an individualistic and competitive struggle to "get ahead."What students may learn in school is that educational institutions are essentially a staging ground for the private pursuit of elite colleges, and high status and well paying jobs (Cookson & Persell, 1985).Or the school may be experienced by students as an individual pursuit of prestige among peers.While these aspects of school organization are inevitable to some degree, if this organizational culture dominates, and is not balanced with collective purpose and mission at the school, the hidden curriculum in schools may reinforce the view that participation in public life is about the competitive and individualistic struggle for private advantage.
Take for example the argument of Bowles and Gintis (1976) on school socialization.They argue that the primary way in which students are prepared for relations in the workplace is through socialization into a school organization that is marked by an individualistic and competitive struggle, which parallels social relations in the free market economy.We would expect that this type of school socialization would not prepare students well for participation in democratic institutions designed to arrive at a collective good, rather than the imposition of the special interests of the powerful.
Implicit in these examples of negative organizational effects on student civic socialization is that schools that foster a sense of collective identity among students, school personnel, and parents will provide a better civic education for students.Schools that are more successful in fostering meaningful school identities-a sense of "we-ness" within the school-that become an important part of individual identity formation for students will teach "civic lessons" about expectations and norms for behavior outside the family.The organizational culture of these schools, which Anthony Bryk and colleagues (Bryk, Lee, & Holland, 1993) call "communal" organizations, fosters civic skills that improve participation in public life in the adult years.For example, schools with stronger collective identities are likely to increase social trust of students, and, as Putnam (2000) argues, social trust is a key dimension of active civic participation.School organizational cultures that temper competitive individualism, professionalism, and bureaucracy with "relational trust," which is predicated on mutual respect, perceived competence, personal regard for others, and integrity (Bryk & Schneider, 2002), create the kind of a school environment that models the civic life that we hope for in a democracy.Dreeben (1968) limited his focus to public schools, but it is important to consider variation in school socialization by sector.How are different types of schools equipped for imparting a "hidden" civic education?Are there systematic differences across school sectors?In general, what differences would we expect between public and private schools that would affect civic education for students?

SCHOOL SECTOR AND HIDDEN CIVIC CURRICULUM
The work of Brint and colleagues (Brint et al., 2001) offers one tantalizing suggestion that school socialization differs markedly across schools sectors.Brint includes 2 private schools in his sample of 64 schools, and includes a footnote on a major difference between the public schools and one of the private schools.The private school created an organizational culture and institutional mission focused on character formation.Data limitations precluded a careful analysis of public and private school differences in the work of Brint and colleagues.What remains to be done is to follow this lead on school sector differences in school socialization, and uncover the implications for civic education of differences in the school socialization environment.In this section, I account for organizational differences across school sector, which shape the "hidden" civic curriculum.

PUBLIC SCHOOLS
What makes it difficult for most public schools to provide an effective civic education is that they tend to be organized along bureaucratic lines and operated according to a rational or legal basis of authority.Research shows that public schools are increasingly organized along rationalistic, hierarchical, and bureaucratic lines (Meyer, 1992;Morrow & Torres, 1995;Salganik & Karweit, 1982).The post-World War II historical process in which schooling governance and funding shifted from local to state and national levels is one part of the development of a rational, bureaucratic public schooling system (Carnoy & Levin, 1985;Kraushaar, 1972;Meyer & Rowan, 1983;Meyer, Scott, & Strang, 1986, 1994;Ravitch, 1983;Tyack, 1974).
Recent research has shown that public schools are performance-oriented bureaucracies, which have organizational imperatives of individual achievement, conformity, and order (Brint et al., 2001;Lortie, 1975).In this organizational setting, the norms that govern behavior tend to be limited to rational, tech-nical, and "objective" or "neutral" criteria (Arons, 1983;Lawrence-Lightfoot, 1978).Public schooling interests, such as educators, professional organizations, and administrators, unable to agree on the meaning of education, focus on technical, bureaucratic process, rather than establishing agreement on a normative educational mission (Nord, 1995).This tends to strip specific content to the normative environment in public schools.The basis of organizational legitimacy is grounded for the most part in "contractual trust."The dominant culture in this type of organization places the individual and the institution in a more adversarial relation to one another.In an organizational culture of contractual trust: The basis for social exchange is primarily material and instrumental.Although personal friendships may arise over time through repeated interactions, social-psychological motivations remain modest, and the moralethical dimension is weak or nonexistent.A contract defines basic actions to be taken by the parties involved.(Bryk & Schneider, 2002, p. 17) Note that Bryk and Schneider do not claim that the contractual basis of trust defines many school organizations.The contractual basis of trust provides the dominant form of legitimacy for public school organizations, and is invoked when conflict arises within the school community (Arum, Beattie, Pitt, Thompson, & Way, 2003).
Instead of active socialization in the specific aspects of character and virtue, Brint et al. (2001) find that teachers tend to avoid normative discussions with explicit content except in the sporadic and isolated cases in which classroom disruption or student conflict demands it (in the name of the organizational imperative of orderliness).
Bureaucratic organizational characteristics do not exhaust the sometimes conflicting forces that shape public school organizational culture.Public schools are to some extent organized by professional norms (Lawrence-Lightfoot, 1978).But the extent that public schools are run in ways that make room for professionalism of teachers and administrators, a separate basis of organization, does not make the school friendlier to collective identity and mission formation than it is under bureaucratic norms.Professional norms may provide a basis for teacher unity, but professional norms disrupt bonds with other members of the school community, especially parents.Rather than contributing to the collective identity of the school community, professionalism orients teachers and administrators to norms and communities well outside whatever local community exists (Lopez, Scribner, & Mahitivanichcha, 2001;Sikkink, 1999).
There are at least two implications of the bureaucratic and professional orientation of most public school organizations that may affect the hidden civic curriculum in public schools.The first regards the relation of the student to the school, the limits and content of the student "role."In the performance-oriented bureaucratic schools, students are constructed as individual clients and consumers of services, who necessarily look out for their own interests in a formal transaction between school and student.This does not mean that public schools do not attempt to foster collective identity, in which students take on a "we" orientation rather than simply remaining an "I." Brint et al. (2001) have shown that school socialization includes efforts to get students to identify with the school.But the forms of collective identity fostered in a performance-oriented bureaucracy are thin-operating with little content or emotional energy.As Brint himself points out, the efforts toward building a collective identity in the school tend to serve the mission of the school in fostering individual performance and effort, and maintaining school order.Students likely see that connection as well, which makes an authentic collective identity more difficult to build.
In short, this bureaucratic organizational form impedes the formation of collective identity within the school-at least a collective identity that is meaningful to students and contributes to individual identity formation of students.In this context, it is less likely that schools can generate the kind of common mission in which they can legitimately set norms for community service, which is important for civic education (Center for Information and Research on Civic Learning and Engagement, 2003).Nor is it likely that students will gain a sense of personal obligation to participate in the civic life of the school and community.Any civic participation that is generated through the school is likely to be considered as a formal rule, another hoop to jump through on the way to graduation and college, rather than a meaningful activity that emerges out of and is legitimate because of a student's participation in the collective identity of the school.Moreover, value messages that may relate to civic participation are refracted through this organizational prism, which reduces the effectiveness of school socialization in preparing students for public life.Respect for others is reduced to a minimal sense of not disrupting class or fighting with others."Good citizenship" is redefined in terms of the organizational need for order and conformity (Brint et al., 2001).Rather than relating to behavior in political institutions, it is defined as not creating discipline problems for school authorities.

PRIVATE SCHOOLS
In terms of school socialization, private schools may have an advantage in civic education over today's public schools.One advantage is rooted in the contrast between the bureaucratic form of organization in public schools and the communal form of organization in religious schools.A communal organization is organized less by hierarchical, contractual, and single-purpose relations ("specificity," in Dreeben's [1968] terms) than by multidimensional and personal relations marked by mutual obligation.In Catholic schools, for instance, professionalism is tempered in favor of a personalism that focuses on meaningful relationships within the school (Bryk et al., 1993).They are marked by a relatively strong sense of mission, which draws normative sanction and legitimacy from religious moral communities.And religious schools provide a setting in which value consensus can thrive (Coleman, Hoffer, & Kilgore, 1982).The communal organiza-tion provides the basis for developing relational trust (Bryk & Schneider, 2002), in which the school culture embeds respect and personal regard for others.Moreover, the social organization of religious schools is marked by high levels of social capital (i.e., social ties among parents, administrators and students marked by trust and mutual obligation), the experience of which may be important since social capital tends to generate greater civic participation (Paxton, 1999;Putnam, 2000).
The result is that the communal form of organization, with its strong sense of a normative and binding school mission, combines with higher levels of social capital to provide a strong basis for collective identity formation within the school.A "communal" school organization is not only more likely to generate a collective identity at the school level; school collective identity is more likely to play an important and meaningful role in the individual identity formation of students.
This private school organizational culture tends to offer a setting in which students experience and practice civic-mindedness.Community service activity, for example, is made meaningful for students-an expression of a personal identity molded by the collective claims of the school community, rather than being driven by interest in private gain.The enhanced role of school collective identities and the heightened level of social capital among students increase the likelihood that students find at school opportunities for practicing sociability and learning norms of cooperation and collective service.In a communal form of school organization, students are more likely to experience school as a place in which solidarity is experienced, and participating in the school "commons" trumps self-interest.The participation in this school community outside of the family provides an important model for participation in public life.
What we are arguing, in sum, is that the organizational form of schools is relevant to the civic education of students.Organizational forms that build collective identity and solidarity, rather than interest group conflict and individualism, provide an important socialization experience that may contribute to positive orientations to public institutions that affect behavior though the life course.That differing organizational forms matter for student experiences of school should be evident in surveys of student attitudes toward and attachment to their school.

EVIDENCE ON CIVIC SOCIALIZATION DIFFERENCES ACROSS SCHOOL SECTORS
If the above school sector arguments are correct, we should find that, on average, private schools, and especially religious schools, have students who are more committed to and less alienated from their school community.We should also find evidence that the school is a functional community, in Coleman's sense (1988), which is marked by mutual respect and a legitimate normative order.
I look to the National Household Education Survey (NHES), conducted by the National Center of Education Statistics (NCES), for evidence for these claims.The youth file of the NHES is a nationally representative telephone survey of American teenagers from the sixth through 12th grade.I used the 1996 version of the NHES, which includes questions on school climate along with important controls for demographic variables and church attendance.The total sample size is 7,940.Missing values are imputed by the NCES according to a hot-deck routine.
The 1996 NHES dataset includes five questions that serve as indicators of student attachment in school, and the extent that the school is a "functional" community, marked by mutual respect and a normative communal order, and relational trust.Students are asked to agree or disagree with the following statements: "I enjoy school.""In my school, most students and teachers respect each other.""In my school, the opinions of the students are listened to." "My teacher maintains good discipline in the classroom.""The principal…maintains good discipline in my school." I use the first three statements as indicators of the quality of the community, including the level of trust among members of the community.Relational trust within schools depends on respect for community members (Bryk & Schneider, 2002), and this should especially be reflected in the respect among students and teachers and the respect for student voices at school.The last two measures are indicative of effective authority within the school community, which is likely to depend on the success of the community in imparting communal norms for individual behavior.Again, in terms of relational trust in schools, perceived "competence" is a key factor in maintaining institutional trust, and, for school personnel, this is indicated by their ability to maintain an effective order in school (Bryk & Schneider, 2002).

"I ENJOY SCHOOL"
Figure 1 shows the percent of students in each sector that report enjoying school.Most of the differences across school sector are not large, though the local public school students are the lowest on this measure of attachment to school.About 70% of local public school students report that they agree or strongly agree that they enjoy school.The students at non-Catholic religious schools are highest on this measure; 89% agree or strongly agree with the statement.This finding is interesting, since the importance of education is generally thought to be lower within conservative Protestantism (Darnell & Sherkat, 1997), which comprise the majority of schools in this category (Sikkink, 2001).
We would expect, though, that some part of the higher level of school enjoyment in private schools is due to the higher level of education of parents of private school students.Perhaps students in private school would report equally high levels of enjoyment of school if they were in the local public school.The next step, then, is to account for variation in socioeconomic status and other demographic characteristics of students and their families, which may influence reports of enjoyment of school, and to test whether the differences in attachment to school remain.A regression analysis will allow us to control for socioeconomic and other differences across families in public and private schools that may account for the differences in Figure 1.Since we are concerned that the factors that select families into private schooling may also be related to our measures of school climate, we include an extensive set of controls to level the playing field between the school sectors (see Table 1 for distribution of selected independent variables).
The controls included in this model and the models below include several family and parent-level controls.Included are standard demographic characteristics of parents, such as education, income, marital status, and race, along with educational expectations of the parents for the child and the extent that parents talk to their child about future plans for college or work.These educational expectations variables should capture some of the differences in school success and investment in education that may affect enjoyment of school.They also should help to control for the factors that lead parents to choose private schooling in the first place.As further controls for family differences, I include parent church attendance, parent civic involvement, home ownership, receipt of food stamps, and so forth.The models include controls for contextual variables, the poverty and racial distribution of the family's residential neighborhood.
To estimate the school sector effect, I first net out the effect of several other child-level factors that may be related to school sector and enjoyment of school, including the age, sex, and race of the child, work hours of the child during the school year, child's grades, the size of the school, ethnic distribution of the school, child participation in out-of-school activities, and so on (see Table 2 for complete list).In the model for enjoyment of school, I also included indicators for whether the child thinks the parents are too much involved in their school, and one for those who think that parents are too little involved in school.Parent involvement in school is likely to differ by sector, and this could shape student enjoyment of school.
Table 2 provides the results from the regressions.I used an ordered logistic procedure to account for the four category dependent variables (answers for most questions ranged from strongly disagree, disagree, agree, and strongly agree).The models in Table 2 compare students in each of the private sectors to the local public school population.It also compares students in public schools of choice to the local public school population.Public schools of choice include families whose children attend a magnet or charter school, or who chose their public schools under a school district or state-wide open enrollment plan.The private school students are divided into three groups: private nonreligious school, Catholic school, and non-Catholic religious school.
After accounting for demographic and contextual factors, the analysis in Model 1 (Table 2) shows that non-Catholic religious schools compared to local public schools remain positively related to enjoying school, though this is just short of the usual standards of statistical significance (the .05alpha cutoff).With the stringent set of controls in the models, and the relatively small sample size for the private school categories, it seems reasonable to discuss coefficients that are significant at the .1 alpha level.The private nonreligious students show a similar positive relationship to enjoying school compared to the local public school population.What is more interesting is that the students whose family chose their public school are significantly more likely to report high levels of enjoyment of school.The results seem to show that, after accounting for lower socioeconomic status of children, the public school chooser families are experiencing school much more positively than the local public school families.Some of this enjoyment may be rubbing off from parents who are more committed to a school that they actively chose.But it seems unlikely that this explains the entire effect.A reasonable interpretation of this finding is that the public schools of choice are creating a school environment that may have advantages for civic education socialization.
The coefficients for Model 1 reveal that the strongest effect on enjoying school is for the nonreligious private schoolers (.34), with the non-Catholic religious schoolers close behind (.28).The public school of choice students are quite strongly related to enjoying school; the estimate shows that we would expect on average a .19increase in enjoyment of school for public school choosers, net of the other variables in the model.While this effect is substantially smaller than the effects for religious schools, it is a fairly strong effect since the dependent variable is only a 4-point scale.For the private schooling categories, the effect is similar in size to the effect on enjoyment of a parent expecting the child to attend a 4-year college after high school.The Catholic school effect is similar in size to the public school of choice sector, but it does not reach the usual standards of statistical significance.Note, however, that the estimate of the effect of Catholic school attendance on school enjoyment is affected more than the other school sectors by a student's satisfaction about the level of their parent's involvement at their school.If those variables are dropped from the model (not shown), the Catholic school estimate rises to .24 and is significant at the .1 alpha level.

"IN MY SCHOOL, MOST STUDENTS AND TEACHERS RESPECT EACH OTHER"
The percentage of students in each school sector that report mutual respect between teachers and students differ rather markedly (Figure 2).About 31% of religious school students (Catholic and non-Catholic religious) strongly agree that students and teachers respect one another.The public school percentage is about half that of the religious schools.From these raw percentages, there is reason to be concerned about the kind of civic socialization received by most local public school students.Even if some of these school sector differences are due to demographic variation across school sector, these low numbers for local public schools are troubling; only 15% of public school students experience a schooling environment in which it is very clear that students and teachers respect each other.This important component of relational trust is not strongly felt by most public school students.
Despite that, we also consider whether these school sector differences in mutual respect are significant even after accounting for demographic and other forms of variation across school sector (see Model 2).After controlling for relevant factors, students in all school sectors are significantly more likely to agree that teachers and students respect each other than are students in local public schools.The strength of the effect for private school students is striking.Catholic and nonreligious private school students report almost one point higher on average on this 4-point scale, even after controlling for a large set of other factors that might affect student reports.Though public school of choice students are significantly more likely to report mutual respect than local public school students, the estimate (.29) is about a third of the size of the private school effect.Non-Catholic religious schools are not quite as strong as the other private school sectors (.75), but the effect for these religious schools is more than double the size of the effect for public school students of choice.
When it comes to experiencing a school community of mutual trust, local public school students are at a decided disadvantage.This does not bode well for the socialization experiences that would contribute to an effective civic education in public schools.By not experiencing an institution marked by mutual respect, public school students may be learning the wrong things about civic institutions and normative behavior within them.

"IN MY SCHOOL, THE OPINIONS OF THE STUDENTS ARE LISTENED TO."
Another indicator of the type of hidden civic education curriculum within schools is whether students feel a sense of respect from teachers and administra-Source: 1996 National Householf Education Survey tors at the school.Again, this is likely to reflect the extent that the school functions as an effective community.And one might also expect that non-Catholic religious schools, which are predominately conservative Protestant schools, would not excel on this score.These religious schools have been characterized as authoritarian (Peshkin, 1986) rather than participatory, which may also provide a hidden civic curriculum that would be detrimental to orientations to civic participation.
What is surprising in the bivariate results (Figure 3) is that the non-Catholic religious schools have the highest percentage of students strongly agreeing the student opinions are listened to (25%).Catholic and nonreligious private schools are not far behind (22%), though local public schools are again about half as likely as the religious schools to strongly agree that students have a voice at school (12.9%).Whatever the reason for these differences, the findings raise concern that students experience local public schools as alienating rather than participatory environments.
The regression analysis, which accounts for other relevant factors that may influence student responses, confirms that all the school sectors are significantly stronger on student input than local public schools (Model 3).The coefficients for the models show that the non-Catholic religious schools remain the highest on this measure, net of the other variables in the model.On this 4-point scale, the private nonreligious school students have on average a .58boost in their sense of being listened to, compared to local public school students and net of the other variables in the model.The Catholic schools (.41) and the non-Catholic religious school (.46) are not far behind.The coefficient for each of the religious school sectors is more than twice the size of the coefficient for the public school choice sector.(A test of the statistical significance of these differences reveals that private nonreligious school students are significantly different from public school students of choice at the .1 level, and Catholic school students are nearly so.)At least in creating a participatory environment, which provides an important civic education in itself, all private schools are doing much better than local public schools.

"MY TEACHER MAINTAINS GOOD DISCIPLINE IN THE CLASSROOM"
The raw percentages on discipline in the classroom (Figure 4) show that the religious schools (about 35% strongly agree) are well above the public school sectors (20% strongly agree) on this score, though the private non-religious sector is doing nearly as well in perceptions of teacher discipline (32% strongly agree).Contrary to the above findings, the public schools of choice are not significantly different from the local public school (see Model 4).In the regression models, with the extensive set of controls, all of the private schools are significantly more positive on teacher discipline than local public schools.The public schools of choice are not significantly different from the local public schools.
In this case, we do not find substantial differences by type of private school, which seems to indicate that there is no special boost to functional community provided by adding a religious dimension to private schools.The functional community seems just as strong in private nonreligious schools.After the controls are accounted for, being a private nonreligious school student increases the likelihood of reporting good teacher discipline by about .6 (on a 4-point scale).Religious school students are roughly similar in the size of the effect.

"THE PRINCIPAL…MAINTAINS GOOD DISCIPLINE IN MY SCHOOL."
The percentages in Figure 5 show that almost half of non-Catholic religious schools strongly agree that their principal maintains good discipline.The Catholic schools are nearly as high on this measure, but the local public schools are much lower: only about 30% of local public school students strongly agree that the principal maintains good discipline.Again, public schools of choice are not distinctive on this measure, and the private nonreligious schools (36% strongly agree) fall in between the local public schools and the religious schools.The regression analysis (Model 5) reveals that only the Catholic and other religious schools are significantly different from the local public schools.Even with the controls, the religious schools show higher levels of principal discipline.Being in a religious school increases the average level of principal discipline by about .8 on a 4-point scale.Again, this is a large effect.The private nonreligious school effect is not significant at the .05level, but it does fall below the .1 alpha level.Still, the private religious school effect is about half of the size of the religious school effect.
On this measure, the religious schools are distinctive, which indicates that religious school students experience school to a greater extent as a functional community.Since religious schools are embedded in a larger religious community, they are likely to be more effective in uniting family and school, which provides legitimacy for authorities in the school.The normative environment of the school and religious community grounds the authority of school leaders in religious communities.Building on this functional community, the hidden civic curriculum at religious schools appears more positive than at neighborhood public schools.

DISCUSSION OF FINDINGS
Taking the findings together, all private school sectors tend to be distinctive on these measures of school climate.One measure shows a distinctive religious school effect: that of the discipline maintained by the principal.This effect is consistent with the claim that religious schools operate within a stronger functional community, and this effect of communal organization may be beneficial for the hidden civic curriculum.Beyond that, the private nonreligious school students tend to show as much or more of the organizational characteristics that would contribute to an effective civic education.Perhaps this confirms the view that many private nonreligious schools, such as the private school focused on character-building in Brint's (Brint et al., 2001) analysis, have a similarly strong sense of normative mission and collective identity compared to religious schools.Interestingly, public schools of choice appear to have the organizational strengths that would contribute to a good civic education.But the strength of these effects are substantially less than the private schools on nearly every measure, and the difference between public schools of choice and local public schools disappears altogether in regard to principal discipline.This analysis is not complete, however, since we do not know from these results whether civic socialization in private schools affects orientations toward and actions within civic institutions later in life.The results here only show that private school students experience their schools in a way that offers the opportunity for building collective identities and practicing sociability within their schools.Finding evidence that connects school experience and adult orientations to public life is left for future research.In addition, it is still possible that the move to private education per se orients students to private means toward individual goals, while participation in public schools sets an example of coming together from all walks of life without regard to religious or economic differences.Private schooling has no connection to public purposes, in this view.The stronger hidden civic curriculum in private schools may be overwhelmed by a privatized conception of education (Coleman & Hoffer, 1987) that is reinforced within private schools.This possibility, too, deserves further research.The results so far, however, which show some fairly low levels of functional community and relational trust within local public schools, would indicate that the more likely scenario is that the "public" nature of the public school choice is likely to be overwhelmed by the weak hidden civic curriculum in local public schools.

CONCLUSION
Civic education in schools is not only about what is learned in civics class, but also what students learn through their school experience about what to expect and how to behave in institutions beyond the family.Dreeben (1968) has shown us that forming citizens is not simply about what is taught in the classroom, but what is learned in the hidden curriculum, Extending that model, this article has argued that the hidden civic curriculum is not limited to the organizational char-acteristics that instill the value orientations of individualism, universalism, and achievement.While these are important aspects of a civic education in school, and likely influence the capacity of citizens to participate well in the polity later in life, school experiences for children include the extent and nature of collective identity formation at the school.An alienating school environment, one without a strong functional community, teaches civic lessons as well.In an alienating school environment, all the training in civics class about participating in democratic life for the common good may be contradicted by a student's school experience.Perhaps cynicism and disengagement from civic participation of the younger generation is not due to the lack of civics training in class, but to the experience of alienation and isolation in school organizations.Whether students participate out of self-interest later in life, or whether they drop out of public life altogether, may depend in part on how they experienced schooling in their formative years.
Schools are of course not the sole or perhaps even the most important site for this form of socialization.Certainly religious institutions and perhaps even bowling leagues could play a role here.But I would expect that schools are not insignificant factors either.For most students, schools are their induction into public institutions, and for that reason may have a disproportionate influence on youth expectations for life outside the family.And, simply in terms of time commitment, it is clearly the most significant institution in a child's life beyond the family.One could argue that religious institutions in the United States provide a strong competitor on this score to public schools, but this would not apply to all children, nor would most consider churches as "public."What a student learns in school is likely to have significant long-term effects on the nature and extent of their participation in public life.
To know whether we are getting the civic education we hope for would therefore require asking questions such as: Are students participating in the building of a collective identity within the school?Are they learning that at times it is necessary, important, and personally satisfying to contribute to the common good?We would want to ask not only whether students are in a school that represents the diversity of a geographically-bounded community, but whether the school is organized in a way that contributes to developing sociability and to learning to move beyond self-interest.In general, we would want to know whether the organizational culture of schools is teaching the importance of participating in the discipline (social control) of a group that transcends the self, through which the self is given purpose and direction.
The experience of this type of organization is an important way in which schools can provide an education for civic life.On this score, religious schools, and private nonreligious schools, seem to have an advantage in civic education.The communal form of organization at religious schools provides a setting in which collective identity and relational trust allows sociability and other civic orientations to grow and flourish.
It is troubling that local public schools are not able to provide as strong a hidden civic curriculum as private schools.Even if we agree that schools must be "public" in the traditional sense for a positive civic education to take place, our evidence raises the question of whether civic education in public schools will be contradicted by the hidden civic curriculum that students are currently experiencing in the local public schools.Though structurally we haven't taken the "public" out of public schools, students seem to be learning the wrong things about public life.Given the organization of public schools that is apparently necessary to incorporate diversity and special interests in an orderly, performanceoriented bureaucracy, it seems difficult for the hidden civic curriculum to change in public schools.The structural advantage of the public schools is overcome by the hidden civic curriculum.
There is no question that public schools could improve on this score.But that improvement may be limited by social changes that have altered the sources of functional communities and disrupted the relation of geographic communities, families, and schools.One possible way to improve civic socialization in this social context is to expand public school choice.Of course there are myriad issues to consider in decisions to expand public school choice (Cookson, 1994;Schneider, Teske, & Marschall, 2000;Wells, 1993;Wolfe, 2003).In terms of civic socialization, however, our findings suggest that public schools of choice do better than neighborhood public schools.I would speculate that public schools of choice are better able to foster collective identity and a functional community among parents, students, and administrators.While we do not have direct evidence on this, we do know that parents who chose a public school for their child are more satisfied with the teachers, administrators and order and discipline at their schools than are parents in the assigned public school (Bielick & Chapman, 2003).Perhaps this is indicative of a relatively higher level of functional community and relational trust at public schools of choice.We must also note that parents of public schools of choice are not more likely to be involved in their child's school (Bielick & Chapman, 2003).Future research is called for to definitively explain the school climate at public schools of choice.Further, the findings in this article seem to imply that improving civic education for public school students will require incorporating the best practices in the hidden civic education of private and religious schools.To the extent that public schools find ways to foster collective identity and a strong normative climate in schools, the hidden civic curriculum will complement renewed efforts to include more and better explicit civic education in public schools.