Students’ deviant behavior is an important issue for school management and education authorities in a number of countries. In recent years, the governments of Japan and Taiwan have paid serious attention to deviant behaviors of students in both countries, which have been increasingly frequently reported (Kawanishi, 2004; Laser, Luster, & Oshio, 2007; Lin & Mieczkowski, 2009). For example, on campuses in Japan, a gradually increasing trend toward adolescent deviant behaviors (James, 2001), such as abusive language and alienation from school has been reported (O’Day, 2006, cited in Ambaras, 2006; Platt, 2006, cited in Ambaras, 2006). In Taiwan, there have been more and more junior high school students challenging teachers’ authority and school rules, and skipping school (Lin, Chang, & Yuo, 2009).

Considering that research into students’ deviant behaviors has, in the past, been restricted to regional studies, and there have been few cross-nation comparative studies (Crystal, 1994), and also because deviant behaviors among students in Japan and Taiwan have become a great concern, we decided to conduct a survey to examine the deviant behaviors of senior elementary school students in Taiwan and Japan, in terms of their similarities and differences as well as the factors affecting these behaviors.

In past relevant studies, attachment theory and parental attachment have been frequently used to investigate students’ deviant behaviors. A relatively complete discourse on attachment theory was first set out by John Bowlby, who suggested that the mother is the food source for children during infancy, thus forming the basis of a secure attachment relationship. In other words, infants believe that their mothers are trustworthy, and feel concern about their needs (Bowlby, 1980). This type of attachment through parental relationship has been discussed in many studies (Stevenson-Hinde, 2007), and the research on parent-child mental health has been further extended in those studies (Steele, Hodges, Kaniuk, & Steele, 2010).

Hirshi suggested that parent-child attachment relationships may be correlated with deviant behaviors because, in the process of psychological attachment, parents raise and educate their children to understand what actions are prohibited by society and law (i.e., the beliefs of social control) to reduce deviant behaviors (as cited in Sokol-Katz, Dunham, & Zimmerman, 1997). It has been suggested in some studies that social control could reduce deviant behaviors (Komiya, 1999) and that parental discipline is related to deviant behaviors (Sampson & Laub, 1994). It has also been suggested that teacher discipline is related to deviant behaviors (Shirasa & Azuma, 1961).

Based on the above suggestions and findings gained in previous research, in our study we included factors such as teacher-student relationship, parent-child relationship, parenting, teacher discipline, and social control in our questionnaire survey. In addition to investigating the influences of these factors on students’ deviant behaviors, in this study we also analyzed the differences in deviant behaviors between Taiwanese and Japanese students. Therefore, the research purposes were as follows: a) to investigate the similarities and differences in deviant behaviors between senior elementary school students in Taiwan and Japan; b) to investigate the influences of the teacher-student relationship, the parent-child relationship, parenting, teacher discipline, and social control, in these two countries, and c) to examine the factors affecting Taiwanese and Japanese students’ deviant behaviors.

Literature Review

Between seven and eight percent of elementary schools in Japan have reported that they have experienced the phenomenon of classroom collapse, in which teachers cannot deliver a lesson normally, and students use abusive language (Kawanishi, 2004). Behaviors in defying or ignoring school rules including using abusive language and alienation from school, cause concern for many parents in Japan (James, 2001). The undesirable behaviors of these students include skipping school, skipping class, and breaking classroom or school rules. These behaviors fail to reach the standards desired by society and are the so-called students’ deviant behaviors.

Are schools’/teachers’ methods to correct students deviant behaviors effective? Findings in a study in Japan indicated that the deviant behaviors of adolescents sent to reform schools significantly improved (Shirasa & Azuma, 1961). It can be inferred that teacher discipline is beneficial to improvement in deviant behaviors. However, in the study mentioned above the focus was solely on the differences in the relative frequencies of deviant behaviors and affecting factors were not taken into consideration, therefore, the emphasis could simply be on the meaning and value of school education in students’ deviant behaviors. It has also been suggested in some studies that family factors are more important than school and are the main factors affecting adolescent deviant behaviors (Cheng, 2001). Loss of the parent-child attachment relationship may result in the development of deviant behaviors in students, and children from a broken home may exhibit more antisocial behaviors than other children do (Murray & Farrington, 2005). It has even been indicated in some studies that deviant behaviors are correlated with parental discipline (Laser et al., 2007). The findings of Sampson and Laub (1994) in their study conducted in Boston, USA also supported the foregoing results. These authors found that deviant behaviors were correlated with weaker parent-child attachment and low parental discipline. To summarize the above, it has been found that teacher discipline, parent-child relationship, and parenting are all correlated with deviant behaviors (Cauffman, Farruggia, & Goldweber, 2008).

In Taiwan, family factors such as poor parent-child relationships and inadequate parenting are often viewed as the main factors affecting personal crimes (Hsieh, 1996). Normal adolescents and juvenile criminals were selected by Hsieh as research participants to conduct a study and he found that positive parenting and parent-child relationship had a negative effect on deviant behaviors, and inadequate parenting was viewed as the most important factor affecting deviant behaviors (Hsieh, 1996). However, in some studies it also has been suggested that school factors are the main cause of deviant behaviors (Jou, 2001). In this study we investigated whether school factors or family factors are more important.

Moreover, some Japanese scholars suggest that social control can change the problem of adolescent deviant behaviors (Platt, 2006, cited in Ambaras, 2006). Ezinga, Weerman, Westenberg, and Bijleveld (2008) investigated 800 adolescents aged 12 and 13, and their results indicated that deviant behaviors are significantly correlated with adolescents who have low levels of self-control. As these students are less able to resist temptation, if their behaviors can be restrained by social control, they may be more able to resist temptation.

Past researchers have indicated that students’ gender significantly affects their deviant behaviors in many aspects. For example, Shek (2005) found that male students’ attachment relationship with their father and female students’ attachment relationship with their mother affected their deviant behaviors. Therefore, in this study we used gender as a control variable, along with the grades of the students, which may also affect the research results, in order to understand the net influence of each factor on deviant behaviors.

Based on the research purposes and literature review described above, in this study we sought to answer the following questions:

(a) What are the deviant behaviors exhibited by Taiwanese/Japanese students? Is there a significant difference between the students in these countries?
(b) Is there a significant difference in perception of parent-child relationships and teacher-student relationships between Taiwanese and Japanese students?
(c) Is there a significant difference in perceptions of teacher discipline and parenting between Taiwanese students and Japanese students?
(d) Is there a significant difference in perceptions of social control between Taiwanese students and Japanese students?
(e) What are the main factors affecting Taiwanese/Japanese students’ deviant behaviors? What are the differences in deviant behaviors between students in both countries?

Method

Participants

The participants were 600 senior elementary school students (grades 5 and 6) in Taiwan and Japan. Because it is difficult to conduct a nationwide sampling survey for cross-nation data collection, we selected a city in Taiwan and a city in Japan with similar geographical conditions and economic situations as the sampling target. In Taiwan, Kaohsiung County was chosen, and in Japan, the Miyagi Prefecture was chosen. The former is the surrounding area of Kaohsiung City, the largest city in southern Taiwan, while the latter is the surrounding area of Sendai City, the largest city in northeastern Japan. These two cities are similar in population but the Miyagi Prefecture is much smaller in area than Kaohsiung County. Miyagi Prefecture (including Sendai City) has an area of 6,862.11 km² (Geospatial Information Authority of Japan, 2011), and a population of 2,336,000 (Portal Site of Official Statistics of Japan, 2011); Kaohsiung County (including Kaohsiung City) has an area of 2,946.2671 km², and a population of 2,773,483 (Department of Statistics of Ministry of the Interior, 2011). In Taiwan, 518 questionnaires were received for a return rate of 86.3%; and 487 questionnaires were received in Japan, for a return rate of 81.2%. After invalid questionnaires with missing data and inconsistent answers were removed, a total of 516 and 476 valid questionnaires were obtained in Taiwan and Japan, respectively.

Instruments

The questionnaire was drawn up based on the Deviant Behavior Questionnaire, designed by Harada et al. (2004), the Social Relationship Questionnaire, designed by Lee, Huang, and Shen (2009), and the Teacher Discipline Attitude Scale, designed by Sun and Wang (2008). The questions were revised according to the research purposes. In terms of parenting attitude, the items in the questionnaire were developed based on the definition of Maccoby and Martin (1983). To validate the questionnaire we consulted with experts who were two educational scholars in Taiwan and Japan, respectively. The construct validity was established by conducting a factor analysis and reliability was established using Cronbach’s α value.

According to the statistical results, the reliability and validity of each subscale was good. The first subscale, the Deviant Behavior Scale, contained five items: “I frequently break school rules”, “I frequently talk back to teacher in defiance”, “I never comply with teacher’s requirements”, “I frequently express a view that is opposite to that of the teacher in class”, and “I frequently use abusive language”, and the reliability was .80 and the validity was .69-.79.

The second subscale, the Social Control Scale, contained six items: “I do not do anything to break the rules because I am afraid of legal sanctions”, “I do not do anything to break the rules because I am afraid of the punishment of the teacher”, “I do not do anything to break the rules because I am afraid that my parents will feel sad”, “I do not do anything to break the rules because I am afraid of the judgments of others”, “I am worried that if I do anything to break the rules that my classmates will not like me”, and “The school rules prevent me from doing anything to break the rules”. The reliability was .86 and the validity was .64-.85.

The third subscale, the Teacher Discipline Attitude Scale, contained three dimensions. Dimension 1, with six items, was liberal teacher discipline: “The mentor will establish rules after asking for our comments”, “The mentor will show respect for our ideas”, “When we make a mistake, the mentor will give us an opportunity to correct it”, “When we have difficulties in understanding the lecture, the mentor will make efforts to help us understand”, “The way the mentor treats the students is my role model”, and “When my performance is bad, the mentor will console me”. Dimension 2, with five items, was authoritarian teacher discipline: “When I fail a test, the mentor will scold me harshly”, “The mentor will ask us to follow rules without exceptions”, “The teaching method of the mentor is old-fashioned”, “The mentor tends to scold students loudly”, and “The mentor is unfair to us”. Dimension 3, with three items, was laissez-faire teacher discipline: “The mentor does not attach importance to our academic performance”, “The mentor seldom cares about what we are doing in our day-to-day lives”, and “The mentor does not much care about us”. The reliability was .52-.86 and the validity was .44-.81.

The fourth subscale, Parenting Attitude Scale, contained three dimensions. Dimension 1, with six items, was parents’ authoritarian parenting: “My parents ask me not to fight with classmates”, “My parents ask me to tell them everything and not to make decisions without their permission”, “My parents ask me to go home after class”, “My parents ask me not to play games until I complete my homework”, “My parents forbid me to be self-willed or get mad at people”, and “My parents ask me to keep my own items tidy”. Dimension 2, with six items, was parents’ liberal parenting: “My parents like to listen to me talking about interesting things at school”, “When I am in a bad mood, my parents will be concerned”, “My parents will chat with me no matter how busy they are”, “When I encounter a problem, my parents will discuss it with me to solve it”, “When my classmates visit my home, may parents will treat them kindly”, and “My parents encourage me to give my opinions”. Dimension 3, with four items, was parents’ laissez-faire parenting: “My parents are seldom interested in what I do day-to-day”, “My parents never care about my academic performance”, “My parents allow me to deal with many things arbitrarily”, and “My parents do not much care about me”. The reliability was .63-.84 and the validity was .46-.78.

The last subscale, the Social Relationship Scale, contained two dimensions. Dimension 1, with nine items, was teacher-student relationship: “I will spontaneously ask the mentor if he/she needs any help”, “I feel that the mentor likes me”, “I like the mentor”, “I will give the mentor a surprise on special days”, “I will do something to meet the expectations of teachers”, “I will keep in touch with the mentor after graduation”, “When the mentor is angry, I will console him/her after class”, “I will intentionally perform well to please my mentor”, and “I will listen seriously to the lectures of the mentor”. Dimension 2, with seven items, was parent-child relationship: “My parents make me feel that they care for and love me”, “Most of the time I get along with my parents”, “I like to spend my holidays with my parents”, “When I need help, my parents will provide me with the greatest assistance”, “My parents and I will engage in the same activities”, “When I have any setback, my parents will tolerate it”, and “I am willing to let my parents know who I am friends with”. The reliability was .87-.89 and the validity was .54-.86.

Data Analysis

SPSS version 12.0 was used to analyze data in this study. Frequency distribution analysis and F testing for mean difference were used to understand both Taiwanese and Japanese students’ responses to questions concerning deviant behaviors and the differences. In addition, t testing for paired sample mean was used to test the differences in various questions concerning deviant behaviors. F testing was used to test the differences in other research variables of the students of the two countries, and eta value was used to represent the level of correlation between two variables. OLS regression analysis was used to analyze the factors affecting the deviant behaviors of Taiwanese and Japanese students.

Results

In Table 1 Taiwanese and Japanese students’ responses to various questions concerning deviant behaviors are shown. The highest score for answers to each question was 5 points, the lowest score was 1 point, and 3 points was the median. The mean of answers to each question from both Taiwanese and Japanese students was less than 3 points, which meant that deviant behaviors exhibited by students in these countries were not as frequent as compliant behaviors. In terms of the international comparison on the mean of each question, the F testing of all the answers achieved a significant level. Except for Question 4, the scores of the remaining answers of Japanese students were significantly higher than those of Taiwanese students, which indicated that Japanese students exhibited more deviant behaviors, such as not complying with the requirements of teachers, talking back to teachers in defiance, not complying with school rules, and using abusive language. However, a higher proportion of Taiwanese students exhibited the deviant behavior of “expressing a view opposite to that of the teacher”. In terms of the cross-nation comparison, five deviant behaviors were exhibited by a relatively higher proportion of students, including Taiwanese students’ “not complying with the requirements of teachers” and Japanese students’ “not complying with school rules”. As for the deviant behaviors exhibited by a relatively lower proportion of students, the deviant behavior “expressing a view opposite to that of the teacher in class” was shared by the students in both countries.

In terms of the total score and mean of the answers to the five questions concerning deviant behaviors, the results of F testing indicated that there were more deviant behaviors exhibited by Japanese students than by Taiwanese students (see Table 2). Regarding other research variables, as shown in Table 2, there were significant differences between the students in these countries. The quality of parent-child relationships and teacher-student relationships of Taiwanese students was significantly better than that of Japanese students, as well as the perception of social control. In terms of the perception of teachers’ attitude towards teaching, as opposed to Japanese students, a significantly higher proportion of Taiwanese students suggested that discipline methods used by teachers were liberal. As opposed to Taiwanese students, a significantly higher proportion of Japanese students suggested that the discipline methods used by teachers were authoritarian. In terms of perceptions of parents’ attitude towards parenting, as opposed to Japanese students, a significantly higher proportion of Taiwanese students suggested that the parenting attitude adopted by parents was either authoritarian or liberal. As opposed to Taiwanese students, a significantly higher proportion of Japanese students suggested that the parenting attitude adopted by parents was laissez-faire. In terms of the mean of answers to each question, the parent-child relationships of students in both countries were better than their teacher-student relationships, especially the median of the parent-child relationship, which was less than 3 points. The teachers in both countries should reflect on this result. In terms of teacher discipline, liberal discipline was adopted by more teachers in both Taiwan and Japan. However, in terms of parenting, authoritative parenting was the most frequently used method by parents in Taiwan, whereas more parents in Japan used liberal discipline methods.

As shown in Table 3, the variables significantly affecting Taiwanese students’ deviant behaviors were teacher-student relationships, social control, authoritarian teacher discipline, laissez-faire teacher discipline, liberal parenting, laissez-faire parenting, and gender. The positive and negative β values could be used to interpret the influences of variables; therefore, it could be inferred that teacher- student relationships, social control, and liberal parenting are beneficial in reducing the frequency of Taiwanese students’ deviant behaviors. The better the teacher-student relationship is, the fewer the number of deviant behaviors will be. The greater is the students’ perception of social control, the fewer will be the number of deviant behaviors. The more liberal the parenting attitude, the fewer will be the number of deviant behaviors. When teacher discipline is authoritarian or laissez-faire or when parenting is laissez-faire, Taiwanese students’ deviant behaviors increase. The more authoritarian the teachers’ methods of discipline, the larger will be the number of students’ deviant behaviors. The more laissez-faire the parents’ attitude towards parenting, the larger will be the number of deviant behaviors of their children. Regarding gender difference, female students exhibited significantly fewer deviant behaviors than did males.

As for Japanese students, variables significantly affecting students’ deviant behaviors were social control, authoritarian teacher discipline, authoritarian parenting, laissez-faire parenting, grades, and gender (see Table 3). As shown by the β value (positive or negative), social control was beneficial in reducing Japanese students’ deviant behaviors. The greater the students’ perception of social control, the fewer will be the number of deviant behaviors. The more authoritarian the teacher discipline, the larger the number of students’ deviant behaviors will be. The more authoritarian the parenting, the larger will be the number of students’ deviant behaviors. Similar to Taiwan, female students exhibited significantly fewer deviant behaviors than did the males. However, there was also a difference according to the grade of Japanese students. Grade 6 students exhibited significantly more deviant behaviors than did grade 5 students, suggesting that Japanese students’ deviant behaviors may increase in number as they get older.

In this study, 11 variables were used to explore students’ deviant behaviors. In Taiwan, the explanatory power of the model was .472, which was approximately 50%; while that in Japan was .290, which was approximately 30%. According to the comparison on the statistical results of Taiwan and Japan, as shown by the b value, the factor affecting Taiwanese students’ deviant behaviors most was authoritarian teacher discipline, followed by social control and laissez-faire teacher discipline. The factor affecting Japanese students’ deviant behaviors most was social control, followed by authoritarian teacher discipline and laissez-faire parenting. The results indicated that, among the 11 variables investigated in this study, social control was the factor affecting Japanese students’ deviant behaviors most, and authoritarian teacher discipline was the factor affecting Taiwanese students’ deviant behaviors most.

Discussion

Few deviant behaviors were exhibited by either Taiwanese or Japanese students, and there were significantly fewer deviant behaviors exhibited by Taiwanese students than by Japanese students. These findings may relieve the anxiety of teachers and parents who are worried about the deviant behaviors of students. Although there is a growing trend of deviant behaviors, the overall number of deviant behaviors of elementary school students in this study were still few. Among various deviant behaviors, “expressing a view opposite to that of a teacher in class” was exhibited by a minority of students in both Taiwan and Japan. The behavior of “not complying with the requirements of a teacher” was exhibited by more Taiwanese than Japanese students, while the behavior of “not complying with school rules” was exhibited by more Japanese than Taiwanese students. The findings in our study indicate that elementary school students of both countries have high respect for teachers, and avoid confrontational conflicts with teachers. However, whether Taiwanese teachers’ requests for students are reasonable, or whether Japanese elementary school rules are reasonable, are two questions that deserve further investigation or follow-up studies by education authorities/schools in the two countries.

The quality of parent-child relationships of both Taiwanese and Japanese students were better than were teacher-student relationships, and in this study we found that the teacher-student relationship in both countries was quite poor. Because of the content of the questions used in our survey we cannot provide any reasons for this poor relationship. However, this finding is important for future research about the quality of teacher-student relationship. Qualitative studies should be conducted to probe into this issue.

In terms of teacher discipline, liberal teacher discipline methods were more frequently used than any others by teachers in both countries. However, as for parenting, authoritarian parenting was the most common in Taiwan, while liberal parenting was the most common in Japan, revealing the difference in parenting between Taiwan and Japan. In terms of teacher discipline, compared to Japanese students, a significantly higher proportion of Taiwan students suggested that the teacher discipline methods were liberal compared to Taiwanese students, while a significantly higher proportion of Japanese students suggested that the teacher discipline methods were authoritarian. Whether the poorer teacher-student relationships of Japanese students are caused by teachers’ excessively strict discipline or indifference deserves further investigation. In terms of parenting, compared to Japanese students, a significantly higher proportion of Taiwanese students suggested that parenting was authoritarian or liberal rather than laissez-faire. Compared to Taiwanese students, a significantly higher proportion of Japanese students suggested that parenting was laissez-faire rather than authoritarian or liberal. We found that the excessively laissez-faire attitude of parents may result in poorer quality parent-child relationships, and in designing parenting education programs in Japan importance should be given to this issue.

Regarding the factors affecting the deviant behaviors of Taiwanese students and Japanese students, we found the following differences:

The factor affecting Japanese students’ deviant behaviors most was social control, whereas the factor affecting Taiwanese students’ deviant behaviors most was authoritarian teacher discipline. Social control is beneficial in reducing the number of students’ deviant behaviors, and authoritarian teacher discipline methods may increase the number of student deviant behaviors. Therefore, to reduce the incidence of student deviant behaviors, we suggest that education authorities in Japan should encourage the teaching of various legal and social systems and should achieve the goal of social control through school rules, class rules, and peer influence. In Taiwan teachers should attach more importance to the adverse effects of authoritarian teacher discipline and parenting to avoid increasing students’ deviant behaviors.

We found that the more authoritarian the parenting was perceived to be by Japanese students, the greater the number of their deviant behaviors. However, authoritarian parenting did not significantly affect Taiwanese students’ deviant behaviors.

Neither parent-child relationships nor teacher-student relationships have a significant influence on the frequency of Japanese students’ deviant behaviors. However, in Taiwan, the better the teacher-student relationship is, the fewer the number of deviant behaviors will be. This finding is also worthy of concern by those involved in parental education research in Taiwan. In this study we found that the teacher-student relationship in Taiwan was quite poor, thus, in order for teachers and school administrators to reduce the number of deviant behaviors of students, improving the teacher-student relationship may be a productive strategy. As there has been very little research conducted on cross-country comparison of deviant behaviors, we attempted to identify the similarities and differences of deviant behaviors of elementary school students in Japan and Taiwan. Our findings, thus, have significant academic value. However, because of the limitation of funding and manpower, we compared only one city with similar conditions in each of the two countries. Although the results could help in the understanding of the differences of students in those two areas, these are not comprehensive results. In future studies stratified random sampling could be conducted in the main cities of Taiwan and Japan for data collection in order to obtain results with better inference.

Ambaras, D. R. (2006). Bad youth: Juvenile delinquency and the politics of everyday life in modern Japan. Berkeley, CA: University of California Press.

Bowlby, J. (1980). Attachment and loss: Loss (Vol. 3). New York: Basic Books.

Cauffman, E., Farruggia, S. P., & Goldweber, A. (2008). Bad boys or poor parents: Relations to female juvenile delinquency. Journal of Research on Adolescence, 18, 699-712. http://doi.org/c8fw2j

Cheng, J. (2001). Family violence victimization as an influential cause to juvenile deviance. Journal of Criminology, 8, 215-246.

Crystal, K. C. (1994). Book reviews. Journal of Popular Film & Television, 22, 138.

Department of Statistics of Ministry of the Interior. (2011). Number of villages, neighborhoods, households and resident population in Taiwan. Retrieved from http://sowf.moi.gov.tw/stat/month/m1-01.xls

Ezinga, M. A. J., Weerman, F. M., Westenberg, P. M., & Bijleveld, C. C. J. H. (2008). Early adolescence and delinquency: Levels of psychosocial development and self-control as an explanation of misbehaviour and delinquency. Psychology, Crime, & Law, 14, 339-356. http://doi.org/hnb

Geospatial Information Authority of Japan. (2011). The measure of area. Retrieved from http://www.gsi.go.jp/KOKUJYOHO/MENCHO/201010/opening.htm#03

Harada, Y., Saitoh, K., Iida, J., Sakuma, A., Iwasaka, H., Imai, J., Hirabayashi, M., ... Amamo, N. (2004). The reliability and validity of the Oppositional Defiant Behavior Inventory. European Child & Adolescent Psychiatry, 13, 185-190. http://doi.org/bdn7vg

Hsieh, W.-Y. (1996). A research of the relationship between family factors and juvenile delinquency. Journal of Police Science, 28, 139-156.

James, V. (2001). Japan’s youth runs out of control. New Statesman, 130, 23-25.

Jou, S. (2001). Can a youth love to go to school and yet behave delinquently? A study on schooling and juvenile delinquency. Research in Applied Psychology, 11, 93-115.

Kawanishi, Y. (2004). Japanese youth: The other half of the crisis? Asian Affairs, 35, 22-32. http://doi.org/hnc

Komiya, N. (1999). A cultural study of the low crime rate in Japan. British Journal of Criminology, 39, 369-390. http://doi.org/hnd

Laser, J. A., Luster, T., & Oshio, T. (2007). Promotive and risk factors related to deviant behavior in Japanese youth. Criminal Justice and Behavior, 34, 1463-1480. http://doi.org/hnf

Lee, Y.-C., Huang W.-S., & Shen, S.-P. (2009). A study on the relationship between filial role perception and parent-child relationship among elementary school students. Journal of Research on Elementary and Secondary Education, 23, 161-188.

Lin, S.-C., Chang, H.-T., & Yuo, J.-Y. (2009). The effects of friendly campus on bullying and students’ deviant behavior. Journal of Elementary Education, 32, 75-104.

Lin, W.-H., & Mieczkowski, T. (2009). Subjective strains, conditioning factors, and juvenile delinquency: General strain theory in Taiwan. Asian Journal of Criminology, 6, 69-87. http://doi.org/hng

Maccoby, E. E., & Martin, J. A. (1983). Socialization in the context of the family: Parent-child interaction. In P. H. Mussen & E. M. Hetherington, Handbook of child psychology: Vol. 4. Socialization, personality, and social development (4th ed., pp. 1-101). New York: Wiley.

Murray, J., & Farrington, D. P. (2005). Parental imprisonment: Effects on boys’ antisocial behaviour and delinquency through the life-course. Journal of Child Psychology and Psychiatry, 46, 1269-1278. http://doi.org/cphrmv

Portal Site of Official Statistics of Japan. (2011). Regional statistics database. Retrieved from http://www.e-stat.go.jp/SG1/chiiki/SelectItemDispatchAction.do

Sampson, R. J., & Laub, J. H. (1994). Urban poverty and the family context of delinquency: A new look at structure and process in a classic study. Child Development, 65, 523-540. http://doi.org/cd2t7j

Shek, D. T. L. (2005). Paternal and maternal influences on the psychological well-being, substance abuse, and delinquency of Chinese adolescents experiencing economic disadvantage. Journal of Clinical Psychology, 61, 219-234. http://doi.org/hnh

Sokol-Katz, J., Dunham, R., & Zimmerman, R. (1997). Family structure versus parental attachment in controlling adolescent deviant behavior: A social control model. Adolescence, 32, 199-215.

Steele, M., Hodges, J., Kaniuk, J., & Steele, H. (2010). Mental representation and change: Developing attachment relationships in an adoption context. Psychoanalytic Inquiry, 30, 25-40. http://doi.org/bxz5fv

Stevenson-Hinde, J. (2007). Attachment theory and John Bowlby: Some reflections. Attachment and Human Development, 9, 337-342. http://doi.org/c6r4gc

Sun, M.-Y., & Wang, C.-H. (2008). The development and confirmatory factor analysis of teacher discipline scale for junior high school students. Psychological Testing, 55, 611-633.

Yi-Gean Chen, Department of Early Childhood Caring and Education, Chung Hwai University of Medical Technology, 89, Wenhwa 1st St., Rende District, Tainan City 717, Taiwan, ROC. Email: [email protected]