Association of bullying on happiness at school: evidence from Thailand’s national school-based survey

ABSTRACT This study aims to investigate the association of bullying victimization of primary school students with their happiness in urban and rural areas of Thailand. A nationally representative survey from Thailand Healthy School Data 2017 was employed. A total of 7,825 students (age 6 to 12 years) from 160 schools in 9 regions were included in the analysis. Multistage probability random sampling was applied to select students enrolled and collected data through face-to-face interviews. About 49.7% of students had been bullied. Bullying victimization was higher among males (52.0%) and younger students. Bullying was also associated with urban/rural (x2 = 5.3, p = 0.021). Students who had been bullied were 28.0% less likely to be happy (p 0.000; CI 0.647–0.810). Bullying victimization has a significant negative effect on students’ happiness. The findings suggest that specific action should be taken by entities which directly contribute to the implementation of the education system in Thailand.


Introduction
Studies documented bullying experience as one of the major components affecting children's psychological health. The prevalence of bullying varied across context, ranged from 9% in Korea (Hosozawa et al., 2021), 27% in Thailand (Pengpid & Peltzer, 2013), 30-40% in Australia (Biswas et al., 2020), and 68% in the Philippines (Hosozawa et al., 2021). Children and adolescents who experienced bullying were more likely to have suicidal ideation (Baiden & Tadeo, 2020), depressed, and involved in risky behaviours including smoking, drinking alcohol, and using marijuana (Lin & Lin, 2017). Bullying victims also often experienced frustration and low self-efficacy in undertaking their daily or academic tasks, and that were associated with their physical and mental health, as well as happiness and life satisfaction (Huang, 2021;Lin et al., 2018).
Happiness refers to a feeling of enjoyment, such as a warm internal sensation, a sense of achievement, feeling coordinated, internal calmness, a happy experience, joyful feelings, a sense of release, and being confident in oneself to make healthier decisions in order to increase one's intellect and imagination (Argyle, 2013;Furstenberg et al., 1983). Studies have found that the provision of green space, nutritious foods in school, and the allocation of hours for students' creativity and innovative learning are all determinants of effective factors influencing happiness at school (Talebzadeh & Samkan, 2011). Students who participate in social activities achieve greater happiness in their lives and achieve better academic performance, as well as a developing a stronger appreciation for shared values and new friendships (Nasoohi, 2009). School-related psychological atmosphere variables, e.g. a teacher's encouragement, learning flow, and friendships were found to have a positive association with student happiness in middle and high school (Lee, 2016). In the family level, children who were not abused had fewer psychological problems when compared to children who were abused (Afifi et al., 2014;Rathinam et al., 2021). Those who received more affection and social support from their parents and peers are happier and had a higher quality of life (Kim et al., 2006).
Bullying experience was also cited as one of the contributing factors to a student's happiness, school attachment, and academic performance . Bullying is a multifaceted phenomenon related to a range of indicators such as individual, relational, family, educational, and cultural dimensions (Bibou-Nakou & Markos, 2013). The critical factor in bullying is negative, deliberate, and consistent actions aimed at the most susceptible peers, including systemic, deliberate, and sustained acts. Bullying encompasses repetitive, aggressive behaviour towards the physically and emotionally abused (Carter, 2012;Salkind, 2008). Bullying is a group problem that exists in a social environment where such behaviour is encouraged, sustained, or suppressed by various factors. Bullying behaviour is an interaction between the bully and the victim, leading to certain detrimental effects on either the bully or the bullying victim, and includes the development of signs of depression, anxiety, low self-esteem, and loss of motivation (Bibou-Nakou & Markos, 2013;Swearer & Hymel, 2015;Teel, 2007). An aggressive youth diagnosed with a behavioural disorder may bully others. Young people who are 'rewarded' for bullying activities may continue bullying.
On the other hand, victimization has been found to predict poor physical health and poor transition in school (Swearer & Hymel, 2015). Victimization has also been linked to myriad internalizing issues, such as loneliness and withdrawal. Family factors such as poor parentchild communication were also found to be potential predictors of depression and violence (Kokkinos, 2013;Segrin et al., 2012). Children who have encountered several kinds of family violence are more likely than children who have experienced one form of family violence to internalize and externalize difficulties (Sternberg et al., 2006).
Bullying in school appears as the most common expression of violence occurs in educational settings (Menesini & Salmivalli, 2017). About 246 million children worldwide have been bullied at school by their peers (UNESCO, 2017;UNICEF, 2019), with a higher prevalence in rural rather than urban schools (Laeheem et al., 2009;Leadbeater et al., 2013;Patel et al., 2020). In Thai schools, bullying is more prevalent among younger, less physically active, and truant students. Bullying is also more frequent among students with psychosocial distress and lack of family affection (Pengpid & Peltzer, 2013). In 2019, of nearly five million primary school students in Thailand, 43% experienced bullying (BICT, 2019), and 0.06% (1,756) dropped out from school. This figure was a significant increase over the previous year 0. 03% (930;OBEC, 2018).
Student bullying is seen in schools as a problem that most youth are well aware of, as are their parents/guardians, instructors, and educational personnel. Bullying and school-related stress can have a negative effect on emotional well-being and an exacerbating effect on anxiety (Ringdal et al., 2020). Student dropout is not only caused by a low quality of life, a lack of life and social skills, and a proclivity to engage in illegal and risky activities, but also by the school system, poverty, low academic achievement, and risky behaviour (violent behaviour, drug problems, crime, etc.; Pellegrini & Long, 2002).
While previous studies on bullying in Thailand (Laeheem et al., 2009;Pengpid & Peltzer, 2013;Sherer & Sherer, 2011) predominantly focused on the prevalence and its risk factors, research on bullying and how it may affect happiness and school engagement of Thai primary school children is lacking. This study aims to investigate the association of bullying victimization of primary school students with their happiness and attachment in both urban and rural areas of Thailand. The results of this study may serve as evidence for educators, government programme managers, and policy makers in designing future programmes to improve student happiness and academic achievement, and reduce school dropout.

Data, population and sample
The Thailand Healthy School Survey was conducted in 2017 by the Institute for Population and Social Research, Mahidol University, and is the source of data for this study. The survey focused on student behaviour and school characteristics among a nationally representative sample. Data were collected by face-to-face interviews by trained interviewers. Multistage probability random sampling was applied to select Thai students aged 6-12 years enrolled in early and late primary school stratified by geographical area (region and urban/rural), sex, age and grade.
A total of 8,587 student sample from 160 schools in urban and rural areas from nine regions of Thailand were enrolled. However, about 8.8% (762 cases) were dropped from the analysis due to missing data (546 cases) or ineligible (216 cases). Finally, a total of 7,825 sample with inclusion criteria of students who are enrolled in selected schools received parental permission to participate in the study and were able to communicate and answer questions clearly were included in the analysis. Those students who 1) did not obtain permission to include details, 2) had disabilities or impairment, 3) did not attend school on the day the data was collected, and/or 4) were unable to communicate or answer questions, were excluded.

Measurement
Bully victims were identified by using a dichotomous variable (Yes/No) based on response to the question of whether the student had any experience of being bullied by their peers. Happiness at school was measured by using a dichotomous variable (Yes/No) based on response to three questions: (1) I really like going to school, (2) I think the classroom atmosphere is good, and (3) I'm happy at school. Students who responded 'yes' to all three items were defined as 'happy'.
Other covariates included in the analysis comprising student, school and family characteristics. At the individual level, students were differentiated by their sexes (male or female) and age (6-12). By its grade, participants were categorized whether they were enrolled in early (grade 1-3) or late (grade 4-6) primary. At the school level, participants were differentiated by school area (urban or rural), and school size following the Office of Basic Education Commission classification (small = 1-119 students, medium = 120-719 students, large = 720-1,679). At the family level, participants were classified whether they lived with 1) both parents, 2) one parent, 3) relatives only or 4) extended family.

Data analysis
Descriptive statistics were used to display the frequency distribution of the following: 1) student characteristics, e.g. gender, age, grade, school area, school size, and living arrangements; 2) experience of ever being bullies; and 3) happiness at school. The Chi-square test was used to analyse factors associated with happiness. Since individual data of students nested with the school level data, hierarchical linear logistic regression with a random-effect model was used in the multilevel analysis. The random effects model aims to investigate the impact of community level and simultaneously consider the variables associated with the happiness of students.

Ethical consideration
The Healthy School project received ethical approval from the Institute for Population and Social Research, Mahidol University, Thailand (COA#2019/04-151).

Sample characteristics
The sample consists of 52.1% males and 47.9% females, age between 6 and 12 years, and enrolled in early primary (50.5%) or late primary (49.5%) grades during the data collection period. More than half (56.0%) were enrolled in a rural school, whereas the rest (44.0%) were in an urban school. The proportion who were studying in a medium-size school was slightly higher (45.6%) than those in small-or large-size schools (31.0% and 23.4%, respectively). The vast majority of students resided with their extended family (49.9%), whereas those who lived with both parents accounted for only 16.3% (Table 1). Despite the fact that nearly half (49.7%)

Prevalence of bullying in Thai primary schools
Both male (52.0%) and female (48.0%) primary school students had been a victim of bullying. The prevalence of bullying victims was statistically significantly associated with sex, where the proportion of male students who had been bullied was slightly higher than females (x 2 = 15.5, p < 0.001). Having ever been bullied was also strongly associated with age (x 2 = 2271.2, p < 0.001) and grade (x 2 = 2782.1, p < 0.001). The proportion of younger children (age 6-8) with a history of being bullied was significantly higher than older children, and those enrolled in an early primary grade experienced bullying more than their higher-grade counterparts (Table 2). This study also found that bullying experience differed by school characteristics, e.g. urban/rural (x 2 = 5.3, p = 0.021) and school size (x 2 = 7.1, p = 0.020). The proportion of students in rural schools who had been bullied was slightly higher (57.3%) than their counterparts in urban schools (42.7%). Bullying was also more prominent in medium-size schools than small-or large-size schools. There was no statistically significant association between living arrangements and bullying experience among Thai primary school students (x 2 = 4.5, p = 0.207).

Happiness at school: prevalence and correlates
Over half (53.5%) of this sample of students expressed their happiness at school. The findings of the bivariate analysis with the chi-square test suggest that happiness at school was not associated with sex/gender (x 2 = 3.4, p = 0.063), but associated with age (x 2 = 121.3, p < 0.001) and grade (x 2 = 116.1, p < 0.001). Happiness at school was also significantly associated with school area, e.g. urban/rural (x 2 = 48.1, p < 0.001), school size (x 2 = 58.8, p < 0.001 and bullying experience (x 2 = 113.9, p < 0.001). There was no significant association between living arrangements and happiness at school of primary school students (x 2 = 4.5, p = 0.209; Table 3). The purpose of the hierarchical logistic regression analysis with random effect model is to investigate the correlates of bully victimization on students' happiness at school, by considering group structure (Wong & Mason, 1985). The covariates in the statistical model were as follows: sex, age, school area, school size, living arrangements, and whether or not the student was a victim of bullying. The insignificant of rho suggested there was no effect of community variables to all covariates in the model. The comparison between random (model 1) and fixed effect (model 2) showed there was no difference in the coefficients of both models, implying that the individual-specific effect is correlated with the independent variable (Table 4).
The analysis of the relationship between bullying experience and happiness (after controlling for covariates) found that bullying experience was indeed significantly associated with happiness at school. Students who had been bullied were 28.0% less likely to be happy (p < 0.001; CI 0.809-0.041). Sex had a significant, positive correlation with happiness at  school, whereby male students were 1.11 times more likely to be happy compared to their female counterparts (p 0.021; CI 1.106-0.051). Age, on the contrary, had a weak negative association with happiness at school, and was only significant among those age 7 years (OR 0.960) compared to those age 12 years as reference (p 0.039; CI 0.998-0.018). Grade showed no significant association with happiness at school (Table 4).
Happiness at school was significantly associated with a school's characteristics. In urban schools, students were 18.0% less likely to be happy compared to their rural counterparts (p < 0.001; CI 0.916-0.047). Students enrolled in large schools also had lower probability of being happy than those studying in smaller schools (p < 0.001; CI 0.836-0.055). The bivariate analysis did not find a significant association between happiness and student living arrangements. However, the multivariate analysis found a weak negative association between a student's living arrangements and happiness at school. Students who lived with only one parent (either mother or father) were less likely to be happy compared to those who lived with the extended family (p 0.015; CI 0.985-0.028).

Discussion
Almost half (49.7%) of Thai primary students had been bullied. This prevalence is higher than in China (30.4%; Xu et al., 2020) but lower than in India (77.5%; Hennig et al., 2017). Although the data across countries are not fully comparable due to different study designs and methodologies, it cannot be denied that there is an increasing trend of bullying among Thai students. A 2014 study found that 15.9% of Thai youth had engaged in physical fights, and 8.1% had been cyberbullied (Sirirassamee & Sirirassamee, 2015). At the global level, a meta-analysis of 80 international studies reported that the prevalence of bullying perpetration was 34.5%, and victimization was 36% (Modecki et al., 2014;UNESCO, 2017). Students experienced various types of bullying, ranging from verbal to physical abuse, and bullying has become more prevalent in online platforms (i.e. cyber-bullying; Bork-Hüffer et al., 2020;Goian, 2020;Kurniasih et al., 2020).
This study found that prevalence of ever being bullied was higher in males and students in lower grades (i.e. younger). Those results are consistent with previous studies from Thailand and various settings which found that male students were more likely to be involved in bullying than their female counterparts (Sherer & Sherer, 2011). Incidence of being bullied was lower among older students (Pouwels et al., 2018), perhaps because they have acquired an ability to deter bullying situations given their higher social skills than the younger, more naïve students (Pellegrini & Long, 2002). However, it is also true that older male youth are more likely to be perpetrators of bullying, and tend to target people who are younger than they are (Pengpid & Peltzer, 2013). Although our study did not have data on the time/location of the bullying behaviour, some studies have found that the majority of bullying occurs in the classroom and on the playground, and the majority of it occurs during lunch and afternoon recess, or just before the end of the school day (Bhatrajaras & Unisa, 2019). Students who were a victim of bullying were less likely to be happy at school. While such a finding is expected, it is a warning to Thai school administrators and policy makers, as well as mental health care providers. That is because having a history of repeatedly being bullied can lead to adverse mental conditions, such as paranoia, anxiety, depression, and generally feeling unsafe at school. (Chen et al., 2018;Trangkasombat, 2012). The level of happiness among students who had been bullied at school was significantly associated with sex (gender), school grade, school characteristics (urban/rural or school size) and living arrangements.
The findings show that younger students (early primary) were less likely to be happy at school than older students (late primary). This could be due to the process of adjusting to school life and adapting to changing class size, teachers, and classrooms (Rice et al., 2011). The stress may also increase vulnerability to being a bully victim (Plexousakis et al., 2019). With regard to bullying experience, we also found that early primary grade students had a higher prevalence of being bullied than those in higher grades (Flannery et al., 2004).
This study also found that students in urban schools were less happy than students in rural schools. A previous study showed that the classroom environment has an impact on psychological, emotional, and cognitive factors, particularly creativity (Jung et al., 2020). Although the infrastructure of rural schools may be less developed than their urban counterparts, the (spatial) size of rural schools is, on average, bigger than urban schools, and that encourages the students to engage in physical activity during recess or break, and that freedom of movement can contribute to happiness (Cote-Lussier & Fitzpatrick, 2016). In addition, Thai students in urban schools typically spend their after-school hours with private tutoring to boost their test-taking skills in such core subjects as maths and English (Nattida & Duangkamol, 2015). Urban youth are also more likely to live in an apartment with less space to play indoors, whereas outdoor play is considered unsafe by many parents who live in the city (Truong & Mahon, 2012).
This study also suggests that students in larger schools were less happy than their counterparts in small schools. It should be noted that the size classification of schools in Thailand is defined based on the number of students (small = 1-119, medium = 120-719, and large = 720-1,679; OBEC, 2021). Students in a larger school have the opportunity to make more friends and find peers who share their interests and values. Unfortunately, the higher enrolment does not always translate into more and better facilities. Thus, students in a bigger school may have less chance for free play and unstructured activities (Gray, 2011).
The strength of this study lies in the national representativeness of the sample, which means that the findings should be useful for central programme managers and policy makers. The sample also includes schools in rural and urban areas, as well as different sizes of school. However, several limitations of this study should be acknowledged. First, the happiness level of the sample of students was only measured by a yes/no variable. That is a rather crude measure to accurately classify happiness. Secondly, the bully-victim experience was not defined further by its types, e.g. verbal teasing, physical bullying, or social harassment. Hence, this study provides valuable evidence that should raise the awareness of the relevant Thai government officials and policy makers, as well as the schools and the families with young students to better address children's well-being in the future.

Conclusions
Bullying victimization has a significant, negative correlation on a primary school student's personal happiness and sense of happiness at school. The prevalence of happiness is higher in rural schools than in urban schools. The findings suggest that specific action should be taken by entities which directly contribute to the implementation of the education system in Thailand. That includes the policymakers who want Thai students to feel happy at primary school by encouraging students to be fully involved in group activities to boost/maintain their mental health and well-being. Increasing a student's sense of safety at school (for example, by reducing victimization and improving the general school and community safety atmosphere) is likely to be an effective method for increasing classroom engagement, especially for the quality of managing large classrooms. The findings from this study also imply that the sense of connection with the school can improve learning performance and could, in the future, decrease the problem of bullying among primary school children in Thailand, and that should contribute to reduced school dropout.