Factors of bullying

Department ofMathematicsand Computer Sciences, Faculty ofScience and Technology, Prince of Songkla University. Abstract The purpose of this study was to study the technique for identifyingbullyingoutcomes, and to investigate the risk factors associated with bullying behaviour at Pattani primary schools, southern Thailand. A cross-sectional survey was conducted among 1, 440 students.

Descriptive statistics, Pearson’s chi-squared test, and logistic regression were used for data analysis. In this study, factor analysis and standardized score techniques were used to identify bullying outcome. It was found that 301 (20. 9%) students could be identified as a bully. Witnessing parental physical abuse and preference for action cartoons were major risk factor for bullying others. Students having parental physical abuse experiences were more likely to bully others than were those who had never witnessed parental physical abuse (odds ratio 7. 11, 95% confidence intervals 5. 6-9. 60). Students who preferred action cartoons tended to bully more than did students who preferred comedy cartoons (odds ratio 2. 96, 95% confidence intervals 1. 99-4. 43). Key words: bullying, cartoon, factor analysis, parental physical abuse, risk factors The 2nd International Conference on Humanities and Social Sciences April 10th, 2010 Faculty of Liberal Arts, Prince of Songkla University Beliefs – Literature – Lifestyle_002 1 1.

Background and significance of problem

Bullying behaviour in primary schools is well-known to students, parents, teachers and educational personnel.

School bullying is a serious problem which affects students’ quality of life, inflicting psychological, emotional, and physical damage and occurs throughout the world. School bullying can be defined as any negative actions repeatedly inflicted by a stronger student or student gang toward another student (Olweus, 1999). This negative action must be deliberate and carried out with the intent of causing harm to the victim (Farrington, 1993). Bullying might be classified in a variety of ways including physical assaults and psychological or emotional or verbal harassment.

Beale (2001) explained that physical bullying is action oriented and intended to intimidate or physically hurt the victim through pinching, pushing, kicking, and hitting, while verbal bullying is using words to humiliate or hurt someone’s feelings through teasing, insulting, or threatening behaviour. The major reasons that children bully others are to enjoy exercising power and status over their victims, boredom, jealousy, attention seeking, showing off, anger, revenge, and selfprotection (Besag, 2006). In this way, bullying eases the way for children to be drawn into a path of delinquency, vandalism and criminality (Junger, 1996).

The targets or victims of school bullying are at risk of a variety of negative outcomes. They are more likely than nonvictimized children to become anxious, insecure, lonely, depressed, to be rejected by their peers, drop out of school, feud, or decide to protect themselves by carrying guns/weapons to school (Craig, 1998). There are many causes of bullying, such asdomestic violence(Baldry, 2003), preferring cartoonviolence(Blumberg, et al, 2008), older students (Wolke et al, 2001), and boys (Mouttapa et al, 2004).

Studies have indicated that 38% of students in Netherlands (Veenstra, 2005), 30% in Nigeria (Egbochuku, 2007), 22% in Italy (Gini, 2008), 21% in Canada (Hawkins et al, 2001), 20% in Malaysia (Wan Salwina et al, 2009) and 42% in Thailand (Tapanya 2006). This study aims to study the technique for identifying bullying outcomes, to investigate the prevalence of bullying and the risk factors associated with bullying in Pattani primary schools, southern Thailand.

By identifying students who are likely to bully others, educational authorities can introduce better strategies for reducing and preventing this problem.

Objectives

1. To study the technique for identifying bullying outcomes
2. To estimate the prevalence of bullies at primary schools in Pattani, southern Thailand
3. To analyze the risk factors associated with bullying, in Pattani primary schools

Technical terms The 2nd International Conference on Humanities and Social Sciences April 10th, 2010 Faculty of Liberal Arts, Prince of Songkla University Beliefs – Literature – Lifestyle_002 “ Bullying” has traditionally referred to a person’s actions to cause physical or psychological harm to another person. Physical bullying is the most visible and easily identifiable form of bullying. It might take the form of a kick, hit, bite, push, throw of something at someone, bite, or pinch. Psychological bullying includes name-calling, insulting the victim’s name, appearance, economic status, academicachievement, or parent’s occupation, or making negative statements about a victim’s physical disability.

Framework of the study Determinants School factors School type School location Demographic factors Gender Age groupEnvironmentfactors Parental physical abuse Preference of cartoon type Number of close friends Outcome Bullying behaviour

Research methodology

Study design and sampling technique This study used a cross-sectional study design involving interviews and surveys of primary school students attending school between November 1, 2005 and March 31, 2006. The participants were selected by using a multi-stage sampling method.

The first stage involved selecting school location by using purposive sampling, with the criterion being a cluster of four types of school (public school of BasicEducationOffice (B. E. O. ), public school of municipalities, Islamic private school, and Chinese private school). Pattani City was selected as the urban location and Saiburi district as the rural one, because these were the only two districts that met the school-type cluster criterion. In the second stage, public schools were selected by simple random sampling and private schools were selected by purposive ampling (there was only one of each such school in each district). Finally, participants in each school grade were selected by using a systematic sampling technique which was done proportionate to population size across each class; choosing every 4th seat number where there was a single class in a grade and every 6th seat number where there was more than one class in a grade. Sample size calculations followed an Italian study of bullying (Baldry, 2003) and were based on the main outcome and exposure to parental violence and non-exposure to parental violence.

The prevalence of bullying by the Italian primary school students in the ‘ nonThe 2nd International Conference on Humanities and Social Sciences April 10th, 2010 Faculty of Liberal Arts, Prince of Songkla University Beliefs – Literature – Lifestyle_002 3 exposure to parental violence’ group was 45. 7%. This information was then used to calculate the required sample size for this study, obtaining an estimate by substituting ? = 0. 05, 1-? = 0. 2, OR = 1. 344 so Z? /2 and Z? are 1. 96 and 0. 84 respectively, r = 1 (ratio of non bully to bully subjects), p2 = 0. 6 (prevalence of bullying in non exposure to parental violence group), p = 0. 50, p1 = 0. 53, into a formula for sample size given by the following (McNeil, 1996), namely ? ? Z? / 2 ? ? 1 ? 1? + Z? ? 1 + ? ? r ? p (1 ? p ) ? 1 1 ? + p1 (1 ? p1 ) rp 2 (1 ? p 2 ) ? ? 2 n1 = (InOR) 2 Where p1 = p2 p + rp2 , and p = 1 p2 + (1 ? p2 ) / OR 1+ r This gives n1 = n2 = 719. It was thereby concluded that a minimum sample size of 1, 438 was required for this study.

Data collection

Verbal consent to participate in the study was obtained from students after assurance of confidentiality was given to individuals and group administered.

The collection assistants were teachers in target schools, who volunteered to participate and were studying for a Graduate Diploma in Professional Teaching at Yala Islamic University. These teachers were trained in the interviewing techniques and the details of the questionnaire. They were asked to take care not to rush through the questionnaire and also to record responses accurately. The teachers interviewed students in the classroom after permission was granted by the school principal. Each individual was interviewed with grades 1 to 3 students. Interviewed lasted approximately 20 to 30 minutes.

Group administered (narrated) surveys of grades 4 to 6 students took approximately 40 to 60 minutes. With older students, the interviewer read the instructions to them and then allowed the student to write their own responses. Most of these responses were uncomplicated and involved just ticking a box.

Data management and statistical analysis

The data were analyzed using Webstat (a set of programs for graphical and statistical analysis of data stored in an SQL database, written in HTML and VBScript), and R program Factor analysis was first conducted to identify possible factors for future analysis.

Descriptive statistics were calculated as measures of the prevalence of bulling. Pearson’s chi-squared test was used to assess the associations between the outcome and the various categorical determinants. Logistic regression was used to estimate the relative odds of having bullied others, for risk factors. 6. Conclusions The 2nd International Conference on Humanities and Social Sciences April 10th, 2010 Faculty of Liberal Arts, Prince of Songkla University Beliefs – Literature – Lifestyle_002

Identifying bullying outcome

The technique for identifying bullying outcome in this study is using statistical method; factor analysis and standardized score as follows. The first method involved an exploratory factor analyses using maximum likelihood method with varimax rotation, eigenvalue greater than one, and item loadings greater than 0. 30. Factor analysis resulted in the identification of four types of bullying: a serious physical bullying factor comprising ‘ kick’, ‘ hit’, and ‘ bite’, a minor hysical bullying factor comprising ‘ push’, ‘ throw something at’, ‘ beat’, ‘ pinch’ and ‘ scold’, a psychological bullying by maligning a parent factor comprising ‘ insult parent’s occupation’, and ‘ insult parent’s name’, and a psychological bullying by maligning the student factor comprising ‘ insult students appearance’ and ‘ insult students economic status’, as listed in Table 1. In the second method, the new scores for four types of bullying were calculated by using discrete scores to compare with the criteria that were adapted from a Likert rating scale: loadings 0. 00-0. 25 scored as 0, 0. 26-0. 50 scored as 1, 0. 51-0. 5 scored as 2, and 0. 76-1. 00 scored as 3. The resulting new scores were thus as follows: kick, hit, and bite: 3, 3 and 1, respectively; push, throw something at, beat, pinch and scold: 3, 2, 2, 1 and 1, respectively; insult parent’s occupation: 3 and insult parent’s name: 1; and insulting students appearance is 3 and insulting students economic status:

The resulting new scores are listed in Table 1. Factor loading Psychological Bullying behaviour categories Serious Minor bullying (Maligning physical Physical parent) bullying bullying Kick 0. 822 (3) Hit 0. 825 (3) Bite 0. 380 (1) Pinch 0. 783 (3) Beat 0. 587 (2) Throw something at 0. 07 (2) Push 0. 458 (1) Scold/ name-calling 0. 366 (1) Insult parent’s occupation 0. 878 (3) Insult parent’s name 0. 399 (1) Insult economic status Insult appearance Eigenvalue 1. 85 1. 77 1. 21 Variance explained 15. 4% 14. 7% 10. 1% Note: Number in the ( ) is resulting new scores for each type of bullying Psychological bullying (Maligning student) 0. 765 (1) 0. 448 (1) 1. 01 8. 4% Table 1: Factor loading scores and resulting new scores for each type of bullying The 2nd International Conference on Humanities and Social Sciences April 10th, 2010 Faculty of Liberal Arts, Prince of Songkla University Beliefs – Literature – Lifestyle_002 In the third method, the total scores for each type of bullying were calculated by using new scores for four types of bullying. The resulting total scores are thus as follows: serious physical bullying (scores 0-7): (3*hit) + (3*kick) + bite; minor physical bullying (scores 0-9): (3*pinch) + (2*beat) + (2*throw something at) + push + scold; psychological bullying by maligning a parent (scores 0-4): (3*insult parent’s occupations) + insult parent’s names; and psychological bullying by maligning the student (scores 0-4): (3*insult students economic status) + insult students appearance.

In the fourth method, the bullying scores were analyzed by combining the total scores for each type of bullying: serious physical bullying (scores 0-7) + minor physical bullying (scores 0-9) + psychological bullying by maligning a parent (scores 0-4) + psychological bullying by maligning the student. The fifth method, transform the bullying scores into Z-scores (standardized to a mean of 0 and a standard deviation of 1). Finally, the students were identified into two categories for bullying, ‘ bullied’ or ‘ not bullied’. The students who had a standardized score greater than 1 were identified as a bully.

Descriptive statistics of bullying and risk factors The distributions of seven determinants in this study involved half of students being from an urban school, 72. 2% were from public school, and 55. 4% were female. Slightly more than one third (34. 5%) were 8 years or less of age, 34. 0% were aged 9-10 years, and 31. 5% were 11 years old or more. Most of students (79. 7%) reported that they had not witnessed physical abuse between their parents. Nearly half (48. 2%) of the students preferred ‘ mystery’ cartoons type, 26. 0% preferred ‘ action’ and 25. 8% ‘ comedy’. Regarding number of close friends, 41. % had 3-5 close friends, 32. 2% had two close friends or fewer and 26. 4% had six close friends or more. In this study, bullying outcome was identified as a dichotomous variables; ‘ not bullied others’ (1, 139 students) and ‘ bullied others’ (301 students). The percentage of students reporting that they had bullied others in school was 20. 9%. 6. 3 Associations between bullying and risk factors The associations between the outcome and the seven study determinants are shown in Table 2. Bullied behaviour Determinants School type Private Public School location Urban Rural Not bullied (1, 139) 78. 2 80. 0 82. 5 77. Had bullied (301) 21. 8 20. 0 17. 5 22. 2 Total (1, 440) 50. 0 50. 0 0. 7 27. 8 72. 2 0. 399 Chisquared 3. 9 p-value 0. 049* The 2nd International Conference on Humanities and Social Sciences April 10th, 2010 Faculty of Liberal Arts, Prince of Songkla University Beliefs – Literature – Lifestyle_002 6 Gender Female Male Age group 8 yrs or less 9-10 yrs 11 yrs or more Parental physical abuse Not witnessed Witnessed Cartoon type preference Comedy Action Mystery Number of close friends 2 persons or less 3-5 persons 6 persons or more.

Table 2: Associations between bullying and study determinants Table 2 shows that school type, gender, age group, parental physical abuse, and cartoon type were strongly associated with bullying. Odds ratio plots of bullying categorized by five different risk factors are shown in Figures 1. Bullying behaviour (Bullying/Not bullying) (Public/ Private) (Male/ Female) School type Witness/ Not witness) Gender Parental physical abuse Years Favorites cartoon type Age (group) Cartoon type Figure 1: Odds ratio plots of bullying categorized by four different risk factors The 2nd International Conference on Humanities and Social Sciences April 10th, 2010 Faculty of Liberal Arts, Prince of Songkla University Beliefs – Literature – Lifestyle_002 7 Figures 1 shows that the students from public schools reported bullying others significantly more often than did students from private schools (OR 1. 35, 95% CI 1. 01-1. 82). More males than females reported that they bullying others (OR 2. 07, 95% CI 1. 59-2. 69).

The students who had witnessed physical abuse between their parents were more likely to be a bully than did those who had never witnessed physical abuse between their parents (OR 7. 22, 95% CI 5. 39-9. 67). The students aged 11 years or more were more likely to bully others than students aged 9-10 years and 8 years or less, (OR 1. 49, 95% CI 1. 13-1. 95). The students who preferred action cartoons tended to bully more than students who preferred mystery or comedy cartoons (OR 2. 93, 95% CI 5. 39-9. 67). 6. 4 Logistic regression analysis of bullying Logistic regression was used to examine the association between bullying and risk factors.

The results are shown in Table 3 and Figure 2. Table 3 shows the results of the logistic regression analysis after omitting determinants with p-values more than 0. 05 using backward elimination. In this reduced model the four factors least significantly associated with the bullying are omitted. The smallest p-values indicate the factors most strongly associated with the bullying were gender, age group, parental physical abuse and cartoon type.

Table 3: Reduced model of association between bullying and risk factors Figure 2 shows the odds ratio plot of the results from fitting the final logistic regression model. It was found that witnessing parental physical abuse was clearly the most strongly associated determinant for bullying others.

Students having witnessed parental physical abuse were more likely 7. 11 times to bully others than were those who had never witnessed parental physical The 2nd International Conference on Humanities and Social Sciences April 10th, 2010 Faculty of Liberal Arts, Prince of Songkla University Beliefs – Literature – Lifestyle_002 8 abuse (95% CI 5. 26-9. 60). Preference for action cartoons was also a major risk factor for bullying others; students who preferred action cartoons tended to bully more than did students who preferred comedy cartoons (OR 2. 6, 95% CI 1. 99-4. 43). Among the age groups studied, older students (11+ years) were more likely to be a bully than did younger students (8 years or less); 1. 89 times (95% CI 1. 33-2. 69). Males were 1. 87 times (95% CI 1. 40-2. 50) more likely to have reported having bullied others than were females. Bullying others mystery Cartoon type action comedy Parental physical abuse yes no 11+ Age group 9-10 -8 boy Gender girl 1/8 1/4 1 2 4 8 Odds Ratio Figure 2: Risk factors of bullying in logistic regression; final model.

Discussion In this study, factor analysis and standardized score techniques were used to identify bullying outcome; a student with a standardized score more than 1 was identified as a bully. Using an exploratory factor analysis for divided type of bullying is in accordance with a study by Parada et al (2005) in which six factors were found of 36 items. Beran (2005) found four factors of 21 items. Carlyle and Steinman (2007) found two factors. Using a standardized score for identifying bullying in which a student with a standardized score more than 1 was identified as a bully (Scholte et al, 2007; Gini, 2008).

In this study, the identifying techniques led to findings that witnessing parental physical abuse and cartoon type were major risk factors for bullying. Witnessing parental physical abuse was clearly the most strongly associated determinant linked to bullying than those who had never witnessed parental physical abuse (7. 11 times more likely). Exposure to parentalfamilyviolence has been found to be related to negative behaviours of students; the students who had witnessed parental physical abuse were more likely to bully others, when compared to those who had not itnessed parental physical abuse. This is in accordance with the studies of Herrera et al (2001) and Baldry (2003) who all reported that parental modeling of aggression The 2nd International Conference on Humanities and Social Sciences April 10th, 2010 Faculty of Liberal Arts, Prince of Songkla University Beliefs – Literature – Lifestyle_002 9 and violence promotes the development of a child’s negative behaviour; the child might copy the parent’s physical actions and might then become a bully to gain success in their own social interactions.

Preference for action cartoons was also highly associated with bullying (2. 96 times more likely). This finding shows that the children who watch action cartoons or cartoon with superhero images were more likely to display aggressive or bullying behaviour. This was consistent with the studies of Kirsch (2006) and Blumberg et al (2008), who argued that cartoon violence may also influence young viewers to transfer violent acts from programs to realworld situations and has a significant additional effect in predicting bullying others.

Students often copy the physical actions of parents or of action cartoon characters and so through their life experiences learn to be aggressive. By personally observing others acting aggressively to achieve some goal the children might learn to act aggressively. With this modeling, the child might then become a bully to gain success in his or her own social interactions. This explanation is in line with finding by Larson (2003) and Williams (2007) who found that children use the same aggressive tactics that they observe; they learn to act aggressively when they model the behaviour of violent acts.

The children are more likely to copy someone they are looking at, and children have a greater tendency to imitate the modeling of those with whom they have the most contact (Cooke, 1993; Kirsch, 2006).

Recommendations

Implications of the study

This study showed that bullying is a serious problem in Pattani primary schools. Witnessing parental physical abuse and preferring action cartoons were the highest risk factors associated with bullying. Parents are the most important persons in providingleadershipand direction for the successful prevention and intervention of bullying.

They should provide close attention and talk regularly with their children about their feelings and relationships with friends at school. They should work in partnership with the school to encouragepositive behaviour. Moreover, they must have patience, try to avoid using violence, and closely advise and control the cartoon program viewing of their children. Findings from this study should help in the development of prevention and intervention policies in the primary schools and assist educational authorities to introduce better strategies for reducing the problem.

School administrators and teachers are the next most important persons for preventing the prevalence of bullying in schools. The school environment should be safe, orderly, and bullyfree. Teachers have to provide positive and mature role-modeling in techniques to students and teach them how to interact with one another. They should develop a program that teaches The 2nd International Conference on Humanities and Social Sciences April 10th, 2010 Faculty of Liberal Arts, Prince of Songkla University Beliefs – Literature – Lifestyle_002 0 students about the dangers of bullying, develop school personnel in teaching bully-free practices and teach students about all forms of bullying: verbal, emotional, psychological, and physical, as well as teach students about disadvantages of cartoon violence viewing and physical abuse between parents.

Recommendations for future studies

Results of our study reveal the extent of bullying in primary schools within Pattani province to be a serious problem.

Future research could focus on comparing prevalence rates among different types of schools or in one specific type of school across districts within Pattani province or nearby provinces, such as Yala, Naratiwat, and Songkhla provinces. Results may give the necessary information to teachers and administrators responsible for improving existing awareness programs, as well as continuing to educate our children in the safest and best learning environment possible.

In order for bullying to be prevented or decreased in our schools, school personnel need to commit to the ideal that bullying is unacceptable, is serious, and should not be tolerated. Bullying is not a rite of passage that students must work out for themselves. When one incident of bullying occurs it is serious. Silence from students does not imply acceptance. Teachers need training that will help them to identify students who suffer in silence. The training needs to include strategies for victims as well as bullies. School personnel need to intervene appropriately in order to gain the confidence of the students.

To gain that confidence, teachers need to be aware that victims of bullying who remain silent will rarely take the initiative to tell teachers they are being bullied and would benefit from having someone pro-active and notice their circumstance and offer them help. There is also a large, silent majority of bystanders in our schools. These students are usually well-developed socially but they do not know how to reclaim the power from the bullies. Some of these students may be afraid to confront the issue and thus ignore or avoid bullying situations.

If we can tap into this silent majority and teach these students the skills they need, we can create a positive school climate with this silent majority holding the power and helping to make the school safe and secure for all. Further research should examine specificteacherreferrals after bullying incidents and whether there is a consistent method of reporting these incidents throughout the grade levels and among all teachers. The administration of disciplinary actions, when dealing with the types of bullying incidents should consistent throughout the school.

An increase in student learning is an overall goal of this study. If students feel safe at school, if they are not worried about the atmosphere in which they learn, then greater student achievements will be likely to The 2nd International Conference on Humanities and Social Sciences April 10th, 2010 Faculty of Liberal Arts, Prince of Songkla University Beliefs – Literature – Lifestyle_002 11 exist. In order to create and maintain a safe and orderly school environment, all stakeholders must take an active role in combating the bullying problem in Pattani primary schools today. .

References

1. Baldry, A. C. 2003. Bullying in schools and exposure to domestic violence. Child Abuse& Neglect, 27, 713–732. Beale, A. 2001.
2. “ Bully busters”: Using drama to empower students to take a stand against bullying behaviour. Professional School Counseling, 4, 300-306. Beran, T. 2005.
3. A new perspective on managing school bullying: Pre-service teachers’ attitudes. Journal of Social Science, 8: 43-49. Besag, V. E. 2006. Understanding girls’ friendships, fights and feuds: A practical approach to girls’ bullying. Milton Keynes: Open University Press.
4. Blumberg, F. C. , Bierwirth, K. P. 2008. Schwartz AJ. Does Cartoon Violence Beget Aggressive Behavior in Real Life? An Opposing View. EarlyChildhoodEduc J. 2008; 36: 101–104. Carlyle, K. E. , and Steinman, K. J. 2007.
5. Demographic differences in the prevalence, cooccurrence, and correlates of adolescent bullying at school. Journal of SchoolHealth, 77: 623-629. Cooke, P. 1993. TV causes violence? Says Who?. The New York Times. Craig, W. 1998.
6. The relationship among bullying, victimization, depression, anxietyand aggression in elementary children.
7. Personalityand Individual Differences, 24, 123-130. Egbochuku, E. O. 2007.
8. Bullying in Nigerian schools: Prevalence study and implications for counselling, J. Soc. Sci. , 14(1): 65-71. Farrington, D. P. 1993.
9. Understanding and preventing bullying. In M. Tonry & N. Morris (Eds. ), Crime and Justice (Vol. 17). Chicago: University of Chicago Press. Gini, G. 2008.
10. Associations between bullying behaviour, psychosomatic complaints, emotional and behavioural problems. Journal of Paediatrics and Child Health. 44: 492– 497. Hawkins, D. L. , Pepler D. J. , and Craig, W. M. 001. Naturalistic observations of peer interventions in bullying. Social Development, 10 (4): 512-527. Herrera, V. M. , and McCloskey, L. A. 2001.
11. Gender differences in the risk for delinquency among youth exposed to family violence. Child Abuse & Neglect, 25: 1037–1051. Junger, T. J. 1996.
12. Youth and violence in Europe. Studies on Crime and Crime Prevention, 5(1): 31-58. The 2nd International Conference on Humanities and Social Sciences April 10th, 2010
13. Faculty of Liberal Arts, Prince of Songkla University Beliefs – Literature – Lifestyle_002 12 Kirsch, S. 2006.
14. Cartoon violence and aggression in youth. Aggression and Violent Behavior. 11: 547–557. Larson, M. S. 2003.
15. Gender, Race, and Aggression in Television Commercials That Feature Children. Sex Roles. 48: 67-75. McNeil, D. 1996.
16. Epidemiological research methods. New York: John Wiley & Sons. Mouttapa, M. , Valente, T. , Gallaher, P. , Rohrbach, L. A. , and Unger, J. B. 2004. Social network predictors of bullying and victimization. Adolescence, 39: 315-336. Olweus, D. 1999.
17. The nature of school bullying: A cross-national perspective. London: Routledge. Parada, R. H. , Marsh, H. W. and Craven, R. G. 2005.
18. There and back again from bully to victim and victim to bully: A reciprocal effects model of bullying behaviours in schools. Sydney, Australia: SELF Research Centre, University of Western Sydney. Scholte, R. J. , Engels, R. E. , Overbeek, G. , Kemp, R. T. , and Haselager, G. T. 2007.
19. Stability in Bullying and Victimization and its Association. Journal of Abnormal ChildPsychology, 35: 217-228. Tapanya, S. 2006.
20. A survey of bullying problem of students in Thailand. Chiang-Mai: Faculty of Medicine, Chiang-Mai University. Veenstra, R. , Lindenberg, S. Oldehinkel, A. J. , De Winter, A. F. , Verhulst, F. C. , and Ormel, J. 2005.
21. Bullying and victimization in elementary schools: A comparison of bullies, victims, bully/victims, and uninvolved preadolescents. Developmental Psychology, 41: 672-682. Wan Salwina, W. I. , Susan, M. K. , Nik Ruzyanei, N. J. , Tuti Iryani, M. D. , Syamsul, S. , Aniza, A. , and Zasmani, S. 2009.
22. School bullying amongst standard students attending primary national schools in the federal territory of Kuala Lumpur: The prevalence and associated socio demographic factors, Malaysian Journal of Psychiatry, 18(1): 5-12. Williams, G. 2007.
23. Gabriel Tarde and the Imitation of Deviance. Available at: http://www. criminology. fsu. edu/crimtheory/ tarde. htm. [Accessed date: September 21, 2007]. Wolke, D. , Woods, S. , Stanford, K. , and Schulz, H. 2001.
24. Bullying and victimization of primary school children in England and Germany: Prevalence and school factors. British Journal of Psychology. 92: 673–696. The 2nd International Conference on Humanities and Social Sciences April 10th, 2010
25. Faculty of Liberal Arts, Prince of Songkla University Beliefs – Literature – Lifestyle_002 13