Challenges in Fostering Democratic Participation in Japanese Education

I draw on Dewey’s concept of democratically constituted society to investigate whether Japanese schools are organized in a way that students can experience democratic living, by examining diversity and interaction within schools. I also rely on Reimer’s notion of citizenship education to explore whether schools in Japan foster the competencies necessary to understand, care about, and act upon global challenges, by examining the teaching of relevance and development of sense of purpose. Based on the analyses of PISA data, I find the following characteristics of Japanese education compared to OECD countries: (1) the high school system sorts students not only by their academic achievement but simultaneously by their family background, creating the least diversified schools; (2) interaction, measured by student participation and debate in class, is low; (3) teaching relevance and application of scientific concepts in class are limited; and (4) students have a low sense of purpose at the end of compulsory education. The lack of opportunity to practice and internalize democratic values in the school, to connect what is being taught to real-world issues, and Education Policy Analysis Archives, Vol. 31 No. 106 SPECIAL ISSUE 2 to develop one’s sense of purpose may partly explain the current youths’ political apathy and why there is little youth-led collective activism in Japan.

descriptive comparative data shows that Japanese youth may not have fostered these values and dispositions to care about and to take a role in addressing the global challenges that the society faces.
According to the 2018 survey by the Japanese Cabinet Office, only 10.8% of Japanese youth agreed that they would "want to get involved in resolving social issues for the betterment of society," compared to 43.9% of American youth1 (Cabinet Office, 2018).This is alarming when considering the current major issues affecting Japanese society, such as resource poverty, import dependence, population crisis, stagnant economy, rising inequality, Internet hate speech, and journalism's legitimacy, need to be dealt with public awareness and collective actions (Nguyen, 2022).Despite some recent youth street demonstrations such as Fridays for Future, inspired by Greta Thunberg, Japan is a country known for relatively few large scale protests and student-led activism (Rauner, 2020).Japanese youth are also characterized by political apathy, which is evident in the strikingly low voting rate of this population group.In the 2021 House of Representatives Election, the voting rate was 36% for people in their 20s, compared to 71% for people in their 60s (MIC, 2021).These data urge us to question whether the current Japanese schools are adequately serving the democratic purposes, by preparing students to understand the global challenges, to care about them, and to gain the skills to address them.

Conceptual Framework
I draw on two literatures as my conceptual framework to investigate the relationship between schools and democracy in Japan.First, the seminal work by John Dewey on democracy and education provides us with crucial insights on how to think about the relationship between schools and democracy.According to Dewey (1916Dewey ( /2018)), "Democracy is more than a form of government; […] it is primarily a mode of associated living, of conjoint communicated experience" (pp.93).Dewey argued that for nation-states to become more democratic, students need to internalize democratic ideas and values through experiencing democratic living, not through reading and hearing about democracy from teachers.In a school setting, democracy is not about what is taught, but how students interact with teachers and peers, how schools are organized to encourage their communication, and how students are sorted into schools.Dewey (1916Dewey ( /2018) ) highlighted two elements that characterize a democratically constituted society.One is diversity, having numerous and varied perspectives.Second is free interaction, where there is equal opportunity for each member to receive and take from others, and where members have the chance to readjust and change their ways of thinking through encountering new situations produced by the interaction.Dewey claimed that students experience democratic living and internalize democratic values through having varied perspectives and open interaction among different members of the school.Banks (2004) further elaborated on the latter point of interaction, explaining that democracy is best learned in a setting where participation is encouraged, where views can be expressed openly and discussed, where there is freedom of expression for both pupils and teachers.In this paper, I investigate: (1) to what extent Japanese school systems are organized to provide diversity to its members, and ( 2) to what extent interaction and participation are encouraged in Japanese classroom setting.
The second framework draws on Reimers ' (2006) argument that we need to redefine educational quality to explicitly include civic purposes of schools in order for schools to foster the competencies necessary to understand and act upon global challenges.For example, that schools educate people with skills to help them make sense of the global events and issues and become agents of change who contribute to global peace and stability efforts should be considered as an indicator of educational quality.By doing so, schools can serve the public and democratic purposes of schooling.Reimers (2006Reimers ( , 2020) also pointed out that what is taught in school needs to be relevant to global issues and should help students develop a sense of purpose.For example, if science teaching and learning is detached from the real world, students may not have the interest to learn about the causes of climate change or environment issues.If students do not develop a sense of purpose, they may not think of global issues as their own issues or view themselves as having the responsibility and the power to act upon global challenges.Building on these ideas, I also examine: (3) to what extent the current Japanese teaching and learning emphasizes relevance, and ( 4) to what extent Japanese students develop a sense of purpose as they go through the education system.These four questions allow us to look at the relationship between democracy and education at multiple levels, through the organization of schools (system level), interaction and teaching (classroom level), and student development (student level).

Data
In this paper, I use PISA 2015 and PISA 2018 data to explore the four questions raised in the previous section.PISA data is useful to answer these questions for two reasons.First, it is a comparative data that allows us to look at Japan from a comparative and relative perspective.Second, PISA targets 15-year-old students, which is the age that marks the end of compulsory education in Japan.These data, then, allow us to examine whether the accumulation of the education that students have received up to this point has cultivated the respectful and ethical mind.
To look at diversity within schools, I use PISA 2018 data, as it is the most recent data available.I also use the 2018 data for looking at students' sense of purpose and meaning of life, because PISA 2018 added new questions in this concept that were not included in the prior cycles.I use 2015 data to examine the extent of interaction and relevance, because PISA 2015 had several questions on whether students are given opportunities to debate and argue and whether teachers explain relevance and application when learning science.2I limit the sample to OECD countries, and 37 countries are included in the analyses.Participating countries and number of students are shown in Table A1, and data and variables used in the analyses are summarized in Table A2.

Diversity within the School
First, with a focus on education as a system, I examine to what extent Japanese schools as an institution are organized to provide diversity to its members.Do students have the chance to interact with students from different background and mindset than themselves in the school they attend, and how large or small is that diversity compared to other OECD countries?In this paper, I focus on diversity in terms of students' SES and academic achievement.To look at the extent of diversity within schools, I employ a hierarchical linear model for each country, and use three metrics. 3The percentage of variance in SES within-schools will show whether students are homogeneous or heterogeneous with their peers in terms of their family background.The percentage of variance in achievement within-schools will show whether students are homogeneous or heterogeneous with their peers in terms of their academic achievement level.Finally, the school-level association between SES and achievement will show to what extent schools are stratified by both student achievement and SES.If schools are highly stratified, it will suggest that schools differ greatly in their composition.For the measure of achievement, I use reading score, as reading is the main domain in PISA 2018.For the measure of SES, I use the PISA economic, social and cultural status (ESCS) composite score, which is comprised of parental occupation, parental education, home possessions related to family wealth, and home educational resources, and has a mean of zero and standard deviation of one across OECD countries.
Figure 1 shows the amount of variance in SES, decomposed it into within-school and between-schools and sorted by the percentage of variance in SES that lies within-schools.For example, in countries like Norway, Iceland, and Finland, approximately 90% of the variance in students' socio-economic status lie within-schools.The diversity within schools is large and students are likely to encounter students from very different family background within a school, regardless of which school a student attends.On the other hand, in countries like Chile and Mexico, more than half of the variance in SES lie between-schools.The family background of peer students will vary greatly depending on which school a student attends, and the diversity within schools is limited.In Japan, 39% of the variance in SES lies within-schools.By this metric, Japan is in the middle, somewhat closer to the right, countries with relatively large within-school SES variance.It is also worth noting that that the overall variance in SES, the sum of within-school and between-school variance, is relatively small in Japan.

Variance in SES Within-Schools
Figure 2 shows the amount of variance in achievement within-school and between-schools and is sorted by the percentage of variance in achievement that lies within-schools.In countries like Iceland, Finland, and Norway and Canada, 90% of the variance in achievement is found withinschools, whereas in countries like Hungary, Turkey, the Netherlands, and the Czech Republic, less than half of the variance in achievement is within-schools.Students in the former countries are more likely to be in a diverse school with students of varying achievement level, whereas students in the latter countries will be in a non-diverse school, where the peers' achievement level will be homogeneous and differ greatly depending on which school the student attends.In Japan, 61% of the variance in achievement is found within-schools.By this metric, Japan is again in the middle, but somewhat closer to the left, countries with relatively little within-school achievement variance.Looking at these two metrics, high schools in Japan are not homogenous but also not very diverse in terms of their SES or achievement, compared to other OECD countries.

Figure 2
Variance in Achievement Within-Schools Lastly, I turn to the school-level association between SES and achievement.Figure 3 shows the expected difference between the means of two schools, which differ by one unit in mean SES. 4  For example, in the Netherlands, a school with many affluent students (school mean SES being one standard deviation above average) will have 143 points higher mean score than an average SES school, whereas in Spain, the score difference between those two schools will be only 43 points.Japan has the second highest association between SES and achievement at the school level.This indicates that schools with many high achieving students tend to have the most affluent students, whereas schools with many low achieving students tend to have the most disadvantaged students.In the case of Japan, PISA is conducted in June of first year of high school, 10th grade.Considering that the school year starts in April, and that majority of students have to take the entrance exam to get into high school, the association should be interpreted not as school effects, but rather selection effects.In other words, high SES students tend to enter schools dominated by high achieving students.Although Japan ranks in the middle in terms of diversity within schools, when looking at SES and achievement independently, it is one of the least diversified within schools, or put differently, most stratified between-schools when we combine the two factors.

Figure 3
Expected Difference in the Mean Reading Score of Two Schools which Differ by One Unit in Mean SES 4 For Figure 3, I run the below model with group-mean centered ESCS, as my interest is in decomposing into within-and between-group components. 01 will represent the expected difference between the means of two schools, which differ by one unit in mean SES.
=  0 +  1 (  ) +    0 =  00 +  01 (  ) +  0  1 =  10 In Japan, there is a clear ranking of schools at the high school level, which is widely recognized through the so-called 'hensachi.'Hensachi is a standard deviation score which tells how far from the statistical mean a typical student admitted to a given school scores on a test and is commonly perceived as a measure representing how difficult it is to get into the school.Hensachi creates a school ranking that is solely based on test score, and finely places every school on a normal curve from the top to the bottom, which results in a rigid hierarchical ranking of schools (Takeuchi, 1995).In addition, because the entrance into high school is primarily based on test score, a majority of parents invest heavily in juku (shadow education) to prepare their children for the entrance exam to get into higher ranking schools (BERD, 2017).Such widely institutionalized testing, sorting and shadow education creates a highly standardized and stratified educational structure (Taki, 2020), and may explain why high-rankings schools are dominated by high achieving and high SES students and low-ranking schools are dominated by low achieving and low SES students.It has also been shown that high achieving schools are not only homogeneous in terms of their SES, but also in their study habit, motivation, and educational expectation (Matsuoka, 2019;Nonoyama-Tarumi, 2019).Based on this metric, one can conclude that high schools in Japan, compared to many other OCED countries, are highly stratified across schools and least diversified within schools.Depending on which school a student attends, the student is likely to interact with students from very different family background and academic ability.Across Japan, high achieving students are likely to be surrounded by students who place a high value on studying and expect to go to college.For these students, they may have little chance to grasp the issue of inequality or living in poverty as they rarely see or interact with disadvantaged students once they enter high school.With little diversity in achievement level and family background inside a school, it is hard to have varied perspectives.One crucial aspect of building and sustaining democracy is the ability to reach out to new and different people, and to find common ground with people one may disagree with by listening and having an open mind, but the chance to do so will be limited if there is little diversity in the school to start with.The Japanese high school selection mechanism may be an efficient way to sort students by their ability, but because it also sorts students by their family background, it may have detrimental consequence in terms of the chance for students to experience diversity and democracy.

Interaction within the Classroom
Next, I focus on classroom within the school, and investigate to what extent interaction and student participation is encouraged in Japanese classroom setting.As a measure of interaction, I use the responses to "Students are required to argue about science questions," "There is a class debate about investigations," and "Students are given opportunities to explain their ideas."5I construct a composite index, which summarizes the above three questions by taking the first principal component (Eigen value 1.88, 63% of variance explained). 6The construct indicates not only the horizontal interaction among classmates, but the vertical interaction between teacher and students.
Table 1 shows the descriptive statistics for each question included in the interaction index, comparing Japan and OECD average. 7The score ranges from 1 to 4, and higher value indicates more student participation and discussion within a class.In all three questions, the average of Japanese students is lower than the average of OECD countries.In addition, the percentage of students that responded "never or hardly ever" is about two times larger for Japan than OECD countries.Table 1 also shows the mean of interaction composite index.The value for Japanese students is 0.71 standard deviation below the OECD mean, and the difference is statistically significant (t (5,118) = -56.24p<.000).The value is the lowest among OECD countries (Results shown in Table A3).
One can conclude that on average, students are given less opportunities to explain their ideas, to argue, and to debate in class in Japanese schools, compared to other OECD countries.If students are not encouraged to vocalize their ideas and to exchange their thoughts, they may not internalize the value of democratic participation, receiving and taking from others and readjusting one's ways of thinking through encountering new situations produced by the interaction.

Teaching Relevance
Next, with a focus on teaching style within the classroom, I examine to what extent the current Japanese teaching and learning emphasizes relevance.For this construct, I use the two questions, "The teacher explains how a <school science> idea can be applied to a number of different phenomena (e.g. the movement of objects, substances with similar properties)," and "The teacher clearly explains the relevance of <broad science> concepts to our lives." 8I compute the average of the two measures and standardize so that the composite index has a mean of zero and standard deviation of one across OECD countries.
Table 2 shows the descriptive statistics for each question.For both questions, the average for Japan is lower than the average for OECD countries, and the percentage of students that responded teachers "never or hardly ever" (explain how ideas can be applied or how it is relevant to the students' lives) is approximately two times larger than that of students in OECD countries.Table 2 also shows the mean of teaching relevance index.The value for Japanese students is 0.48 standard deviation below the OECD mean, and the difference is statistically significant (t (5,096) = -32.82,p<.001).The value is again the lowest among OECD countries (Results shown in Table A4).
The result should be interpreted with caution as the index is based on students' perceptions.Nevertheless, the emphasis on teaching relevance of what is learnt and applying science concepts to real life issues appears to be relatively low in Japan.If students do not see their learning to be relevant, they are less likely to use their scientific learning to improve the society.The question headings and the response items are identical to the questions related to interaction.

Students' Sense of Purpose
Lastly, with a focus on students, I explore to what extent Japanese students develop a sense of purpose as they go through the education system.For this construct, I use the questions "My life has clear meaning or purpose," "I have discovered a satisfactory meaning in life," and "I have a clear sense of what gives meaning to my life."The response ranges from 1 "strongly disagree" to 4 strongly agree."I construct a composite index, which summarizes the above three questions across OECD countries by taking the first principal component (Eigen value 2.35, 78% of variance explained) Table 3 shows the descriptive statistics for each question.For all questions, the average of Japan is lower than that of the OECD countries, indicating that Japanese students, at the end of the compulsory education, on average have a lower sense of purpose.When I compare the mean of sense of purpose index, , the value for Japanese students is 0.41 standard deviation below the OECD mean, and the difference is statistically significant (t (5,235) = -28.75,p<.001).The value is again the lowest among OECD countries (Results shown in Table A5).
On average, Japanese students tend to have less clear purpose to life or sense of what gives meaning to their life.If students don't have a sense of purpose, they may not see themselves as agents of change and be less concerned with issues in the larger society.

Conclusion and Discussion
In this paper, I examined the relationship between democratic participation and education in Japan.I relied on Dewey's (1916Dewey's ( /2018) ) framework to investigate to what extent students are experiencing democratic living and democratically constituted society by looking at diversity and interaction within schools.I then drew on Reimer's (2005) framework to examine whether schools are serving the democratic purposes and developing global citizenship, by focusing on teaching of relevance and students' development of sense of purpose.Based on the analyses of PISA data, I found that at the system level, high schools in Japan sort students not only by their academic achievement but simultaneously by their family background, creating homogenous schools.When examining what happens inside the classroom, the interaction among students or student participation, and teaching relevance and application of scientific concepts tend to be limited, compared to other OECD countries.Finally, when focusing on student development, Japanese high school students have a low sense of purpose, compared to students in other OECD countries.
The analyses are limited in a few ways.In looking at diversity, the analyses focused on family SES, but other dimensions such as immigration status should also be analyzed, given the growing population of non-Japanese.As this study is based on secondary analyses of PISA data, the available items do not make up a strong measure of the constructs of interest in this study-interaction, teaching relevance, and sense of purpose.For example, the three questions I used for the construct interaction were part of the eight questions that OECD uses to measure the construct inquiry-based science teaching and learning.Further, the questions related to interaction and relevance are asked in the context of science subjects and may not capture general classroom interaction and teaching relevance.In looking at students' development of sense of purpose, I assume that it is a result of the schooling system, but it is also the result of socialization by families, media, and larger cultural norms, which I am unable to disentangle in the analyses.Lastly, PISA is a cross-sectional data, and although useful in understanding the relative situation of Japan by comparing with other countries, the data do not show how these aspects change or grow across grade level.
In considering policy implications for how Japanese schools can better serve the democratic purposes and prepare students to become global citizens who can understand global challenges, care about them, and act upon them, I highlight two points.The stratified high school system that creates academically and socially homogeneous schools in Japan is a crucial aspect to scrutinize in thinking about democracy and education.Murphy-Shigematsu (2004) points out that although Japanese schools emphasize the building an empathetic community through encouraging students to cooperate and care for one another, this does not easily extend to caring for and empathizing with those of other cultures.He argues that this is because homogeneity is valued and children do not experience the impact of diversity in their daily lives.Empathy toward those from different backgrounds is much more challenging than empathy toward those with whom one is familiar.It does not come naturally and needs to be practiced through daily school life.Students who are members of homogenous schools are denied the chance to reach out to, learn from, and care for those who are different from them.Democracy is precisely about listening to those who are different, giving each other the benefit of the doubt, and trying to find common ground.Students cannot experience democratic living and a democratically constituted society without diversity and interaction.
Secondly, I speculate on why Japanese students develop a low sense of purpose and rarely take collective action to improve the society.Chen (2021) compared the official rhetoric of social justice in citizenship education in Japan and China and highlighted how justice is framed in Japanese education.He concluded that social justice in the Japanese discourse focuses on respect for other as an issue of interpersonal relation rather than a structural condition requiring redistribution.Second, injustice is perceived as eliminable through identical treatment, rather than through affirming the difference and treating differentially.Third, a majority of narrative concentrates on individual actions rather than a collective action.If students are used to being treated identically as all other students in the class, and if they perceive justice only as a result of individual action and interpersonal relation, they are less likely to collectively act and demand for change in structural conditions.In trying to foster a sense of purpose and self-efficacy as an agent of change, it may be necessary to reflect on the myth of identical treatment of students as well as the over-emphasis on individual action and relationship in Japanese schooling.

Special
to copy, display, distribute, and adapt this article, as long as the work is attributed to the author(s) and Education Policy Analysis Archives, the changes are identified, and the same license applies to the derivative work.More details of this Creative Commons license are available at https://creativecommons.org/licenses/by-sa/4.0/.EPAA is published by the Mary Lou Fulton Institute and Graduate School of Education at Arizona State University Articles are indexed in CIRC (Clasificación Integrada de Revistas Científicas, Spain), DIALNET (Spain), Directory of Open Access Journals, EBSCO Education Research Complete, ERIC, Education Full Text (H.W. Wilson), QUALIS A1 (Brazil), SCImago Journal Rank, SCOPUS, SOCOLAR (China).About the Editorial Team: https://epaa.asu.edu/ojs/index.php/epaa/about/editorialTeam

Table 1
Descriptive Statistics for Interaction in Classroom Note: Mean and SD for each item are computed, treating the Likert scale (1 to 4) as a continuous variable.

Table 2
Descriptive Statistics for Teaching Relevance

Table 3
Descriptive Statistics for Student's Sense of Purpose Taki, H. (2020).Gakkō Kyōiku to Fubyōdō no Hikaku Shakaigaku [Comparative sociology of schooling education and inequality].Minerva Sousho.Tarumi is professor in the Sociology Department at Musashi University in Tokyo.Her research examines patterns of educational inequality.In particular, she studies disparities in academic achievement by student's socioeconomic status, family structure, and gender in Japan.Her research also draws on international large-scale assessments and household surveys to compare educational inequality across countries.Fernando Reimers is the Ford Foundation Professor of the Practice of International Education and Director of the Global Education Innovation Initiative at Harvard University.He is an elected member of the U.S. National Academy of Education and the International Academy of Education.

Table A2
Data and Variables Used in the Analyses Note: SE is computed using replicate weights.