Bullying Victimization Among Junior High School Students in Aceh, Indonesia: Prevalence and its Differences in Gender, Grade, and Friendship Quality Afriani1 & Denisa2

This study aimed to describe the prevalence of bullying victimization and its differences in gender, grade level, and friendship quality among junior high school students in Aceh, Indonesia. The study also investigated the moderation effect of gender on the association between friendship quality and bullying victimization. A total of 360 students from four schools in Banda Aceh was selected using Cluster and Disproportionate Stratified random sampling. The Revised Olweus Victim Questionnaire and Friendship Qualities Scale were used to obtain data on bullying victimization and friendship quality, respectively. The results indicated that 46.9 % of the students were reported being victimized by peers “2 or 3 times a month” or more often. Verbal was the most common form of victimization reported by students. Significant differences in bullying victimization were found concerning gender and friendship quality. The study revealed that males were more likely to report being victimized than females, while students who have a good relationship with their friends reported fewer experiences of victimization than others. Further, regression analysis indicated that gender significantly moderated the relationship between friendship quality and bullying victimization. Preventive interventions regarding peer relationships and gender may be useful in reducing bullying victimization in schools.


A. Introduction
Bullying victimization has been a serious school problem that is sparking public concern worldwide. It is defined as being the object of physical or psychological aggression carried out intentionally and repeatedly by one or several people who have more power or strength (Olweus, 2010). Victimization is often related to negative effects on the adolescents' academic performance and psychosocial adjustment such as depression and delinquency (Cook, Williams, Guerra, Kim, & Sadek, 2010;Gini & Pozolli, 2009;Kendrick, Jutengren, & Stattin, 2012;Rothon, Head, Klineberg, & Stansfeld, 2011).
Because of its destructive effect, prevalence rates of bullying victimization have been studied by many researchers across countries. For example, a study in the United States (Schneider, O'Donnell, Stueve, & Coulter, 2012) showed that 25.9% of the respondents reported as victims. In New Zealand, nearly half of the respondents (47%) had been reported being victimized regularly (Marsh, McGee, Nada-Raja, & Williams, 2010). In England and Wales, 33% of the respondents reported being occasionally bullied by others, and 11% were frequently bullied (BIG, 2015). The cases of victimization were also documented in African countries (Kubwalo, Muula, Siziya, Paupulati, & Rudatsikira, 2013;Owusu, Hart, Oliver, & Kang, 2011) with the prevalence rate of more than 40%. In Pakistan, the rate was lower (17.9%) compared to the study conducted by Sanapo (2017) in Philippine where the frequency of victimization reached 40.6%. The data was uniquely different in Hongkong, Taiwan, Macao (Mok, Wang, jutenfren, Leung, & Chen, 2014) and Thailand (Sittichai & Smith, 2015), in which less than 10% of the students reported having experienced bullying.
Although it has been extensively studied in many countries it has not been the subject of much previous research in Indonesia especially in rural areas like Aceh. Indonesian Committee of Child Protection (Setiawan, 2017) reported that the cases of school violence involving students from different groups and areas have been increased in recent years including in Aceh as the westernmost Province of Indonesia. In 2015, a sixth-grade female student in Aceh died after she was strangled with a scarf by her classmates in the classroom (Phagta, 2015). In the same year, a first-year student in one of the Senior High Schools in Banda Aceh was harassed, beaten, and kicked by his seniors when attending school orientation program (Nizwar, 2015). Cases such as these have raised the issue of bullying victimization in Aceh. To the best of our knowledge, few international published works discuss bullying in Indonesia. For example, a study by Wiguna, Ismail, Sekartini, and Limawan (2016) explored the behavior of bullying and its association with age and gender among adolescents in Jakarta, the capital city of Indonesia. However, this study only focused on adolescents living in an urban area. Therefore, research on bullying in Aceh is very much needed.

Bullying Victimization, Gender, and Grade Level
In studying school bullying, most of the studies examined the frequency of bullying and victimization based on categories (Mok et al., 2014). A study by Bond, Wolfe, Tollit, Butler, and Patton (2007) revealed that students in Australia had more experiences of being hurt and threatened physically, teased, called names, and spread rumors. In Asia-Pacific and South America countries, more students had been called by unpleasant nicknames, being laughed and cursed (Jolliffe & Farrington 2011;Lai, Ye, & Chang, 2008;Sanapo 2017;Silva, Pereira, Mendonca, Nunes, & de Oliveira, 2013;). Results from the studies indicated that verbal bullying was found to be the most common category reported by students (Jolliffe & Farrington 2011;Marsh et al., 2010). Further, the experiences of school victimization that male and female students engage in vary across genders (Brighi et al., 2012;Felix & McMahon 2007). Some studies consistently found that males were more likely to be victimized, and experienced direct physical and verbal victimization than females (Carlyle & Steinman 2007;Crick & Bigbee 1998;Finkelhor, Ormrod, Turner, & Hamby, 2005;Fleming & Jacobsen, 2010;Sanapo 2017). Meanwhile, in other studies, females were reported having higher rates of victimization (Craig et al., 2009) and were more involved in the relational forms of bullying (Craig et al., 2009;Marsh et al., 2010;Wang, Iannotti, & Nansel, 2009;Wei, Jonson-Reid, & Tsao, 2007). The inconsistency findings from multiple studies raise the need for more research to examine the existence of sex differences in bullying victimization.
Grade level was also examined in studying school bullying because the rates of bullying were found to be lower as the age increases. Similarly, other studies have reported the prevalence of bullying decreased as grade level increased (He, 2002;Qiao, Xing, Ji, & Zhang, 2009;Schneider et al., 2012). High school students reported fewer experiences of bullying compared to elementary students who were more prone to be bullied by their classmates (Hong & Espelage, 2012). As indicated by several international studies, it appears that the prevalence rate of bullying is higher among younger students compared to the older ones. Concerning the forms of bullying, older students were more likely to perceive indirect and relational bullying than younger students. Further, the number of experiences of direct bullying decreased as the ages increase (Rivers & Smith, 1994). However, a study by Hartung, Little, Allen, and Page (2011) found that there were no significant differences in bullying regardless of grades. Thus, more research should address grade level in studying school bullying in order to clarify the gap in the literature.

Friendship Quality and Bullying Victimization
Beside personal characteristics, the quality of peer relationship is considered as the salient predictor in determining whether students are likely to bully or being victimized by others (Huang, Hong, & Espelage, 2013;Rodkin & Hudges, 2003). Friends play significant roles for adolescents to socialize and independent from others. Peer validation provides social and emotional support for adolescents, and contribute to healthy development (Ladd, Kochenderfer, & Coleman, 1996). At the same time, adolescents also have to deal with the pressures from the norms of their groups that may lead to negative behavior such as bullying or victimization (Patton et al., 2013).
Research examining the involvement of bullying or victimization was high among adolescents when they were under negative peer influence such as being forced by peers to drink alcohol (Farrar, 2006;Fitzpatrick, Dulin, & Piko, 2007). Other studies suggested that individuals with a lack of support from friends are vulnerable to being victimized (Goldbaum, Craig, Pepler, & Connolly, 2003). Then, having more friends were negatively associated with victimization (Hodges, Malone, & Perry, 1997;Veenstra, Lindenberg, Munniksma, & Dijkstra, 2010). Friends may act as protectors against victimization, but it must be in good quality characterized by giving support to each other, providing help, and companionship (Bollmer, Milich, Harris, & Maras, 2005;Hodges, Boivin, Vitaro, & Bukowski, 1999).
The relationship between friendship quality and affection differentiated by gender has been reported in studies. Some researchers examined how gender would moderate the effects of friendship qualities on group interaction, specifically victimization (Schimdt & Bagwell, 2007;Underwood, 2003). They found that certain aspects of friendship such as help and security were the protective factors for females. For males, help and security had the opposite effect associated with victimization and adverse outcomes. Schimdt and Bagwell (2007) explained help and security play a different role in the friendship between males and females. Females place help and security as the fundamental properties of friendships that protect them from negative effects, whereas males look for these aspects from their friends when they are already in trouble. It seems that friendship as the protective factor of victimization is stronger in females than in males. These differential links between friendship qualities, victimization, and its negative effect on males and females deserve further attention and investigation across different populations and cultures.

256}
The findings discussed above suggest that it is crucial to extend the study on bullying to the rural middle school population in Aceh to gain a better understanding of the prevalence rate of bullying victimization and its associated factors around the different contexts of the world. Thus, the current study focused on (a) how is the prevalence of bullying victimization as perceived by students in Aceh?. (b) Are there significant differences in bullying victimization based on gender, grade level, and friendship quality, (c) Is there a moderation effect of gender on the correlation between friendship quality and bullying victimization? The findings from the present study would fill some literature gap about school victimization in South East Asian countries especially in Aceh, Indonesia. This study would give valuable information for school administrators, counselors, and government alike as they plan to design the prevention and intervention strategies about the issue.

Participants and Procedure
A multistage and disproportionate stratified random sampling was deployed to select the subjects. A total of 360 students from four junior high schools in two selected districts in Banda Aceh was recruited to participate in the present study. The percentages of male and female participants were 51.7 % and 48.6 % respectively. The ages of the students ranged from 12 to 15 years with the mean of 13.3 years (SD=0.92). Regarding the grade level, the percentages of participants from each grade were made representative with the distribution of about 32.8 % for seventh graders, 34.2 % for eighthgraders, and 33.1 % for ninth graders.
Permission to collect data from schools was sought from the superintendents of each school before distributing the questionnaires.
Student consent forms were also distributed as approval for involvement in the study. The questionnaires were anonymous and self-administered by students during school hours. At first, researchers explained clearly to the students about the definition of bullying as described in the questionnaires.
Researchers remained in the classroom for supervising students who were in need while filling in the questionnaires.

Measures
Two self-reported questionnaires were used to collect the data: The Revised Olweus Victim Questionnaire (OBVQ) and Friendship Quality Scale. The OBVQ was developed by Olweus (1996) to assess direct physical and verbal aggression experienced by students during the past months. It consists of nine items; one of which is the global measure of victimization: "How often have you been bullied at school in the past couple of months", and followed by eight specific questions that measure the frequency of particular forms of bullying victimization (physical, verbal, racist, sexual bullying, etc). Examples of these items are "I was hit, kicked, pushed, shoved around, or locked indoors," and "I was bullied with mean names or comments about my race or color." In the questionnaire, bullying victimization is defined specifically, and is rated with 5 response alternatives, "it has not happened to me in the past couple of months", "only once or twice", "two or three times a month", "about once a week", and "several times a week". The response "two or three times a month" was used as the cut-off point for classifying victims as proposed by Olweus and then adopted by the World Health Organization (WHO) to identify the prevalence of school bullying. The internal consistency of the questionnaire in this study was satisfactory, at 0.79.
Meanwhile, the Friendship Quality Scale was designed by Bukowski, Hoza, and Boivin (1994) comprising five aspects: play/companionship, conflict, help, security, and closeness. It is a 23-item scale and rated on five Likert scales ranging from 1 "strongly disagree" to 5 "strongly agree." The total score was derived by summing all of the item scores after reversely coding some negative items (

Data Analysis
The collected data were analyzed using descriptive statistics to estimate the prevalence rate of bullying victimization. Meanwhile, a non-parametric test such as the chi-square test was utilized to explain the differences in victimization based on gender and grade level. Independent sample t-test was also used to evaluate the difference in friendship quality between victims and nonvictims, followed by Cohen d to measure the size effect. The moderation effect of gender on the relationship between friendship quality and victimization was tested using PROCESS by Hayes with 5000 bootstrapped samples and bias-corrected at 95 % of the confidence interval.

Result
Overall, one hundred sixty-nine students (46.9 %) were reported to have been bullied, at least 2 to 3 times a month during the past few months. As shown in table 1, the prevalence rate of behaviors involved in victimization varied in the present study. It was found that seventy-five students (20.8%) perceived themselves as being bullied for the last few months (question item number 1). One hundred and four of the students (28.9%) reported being called bad names and teased, sixty-six students (18.3%) had been bullied with their race or color, fifty-eight students (16.1%) were told or spread bad rumors about them, thirty-three students (9.2%) had things that taken away or destroyed by other students, another thirty-three students (9.2%) reported being bullied with comments or gestures with sexual meaning, twenty-nine students (8.1) had direct physical aggression such as kicked and hit, and lastly, twenty-three students (6.4) reported being excluded from group. Data in the study indicated that verbal abuses were the most common types of victimization.  *p  0.05 Table 3 presents the frequencies of victimization for each type by gender reported as occurring "2 or 3 times a month," "once a week," and "several times in a week." Data from the table showed that the higher prevalence was found among males in overt and relational victimization.
Specifically, more students both male and females experienced verbal victimization such as being called unpleasant names, teased, insulted for more than once in a week. Concerning physical victimization (hit, kicked, pushed, locked in a room), female students had fewer cases than did males.
Males more likely reported the experience of bullying victimization. Table 4 presents the frequencies of victimization for each type by grade level reported as occurring "2 or 3 times a month," "once a week," and "several times in a week." Overall, the data indicated that students in the eighth grade had been bullied more often in a week compared to other students.
Called bad names, teased in a hurtful way, and told a lie or false rumors about themselves were more likely reported by the eighth-grade students. Younger students tended to report themselves as being more frequently bullied than older students. Younger students were also had more cases of direct physical aggression like being hit and kicked, while no cases were found among ninth-grade students for being attacked or threatened physically more than once in a week (the items number 4 and 6).
To assess the moderation effect of gender on the relationship between friendship quality and bullying victimization, a regression was performed from gender (dummy coded 0 = male, 1 = female), friendship quality, and a product term to represent a gender-by-friendship quality interaction. Before conducting the analysis, the variables were tested for the assumptions of multivariate analysis. Results showed that assumptions of normality, linearity, multicollinearity, and homoscedasticity of residuals were met.
Meanwhile, multivariate outliers were checked by inspecting Mahalanobis distances, Cook's distance, and Leverage value. In this data, some cases were identified as multivariate outliers and were removed for further analysis. By using PROCESS model 1, the two predictors and the interaction were entered into a simultaneous regression model. As shown in Table 5 below, the overall regression was statistically significant and explained of  Each of the simple slope tests revealed a significant negative association between friendship quality and bullying victimization, but males (b = -0.19, t(324) = -8.12, p < 0.05) had more higher experiences of victimization than female adolescents (b = -0.11, t(324) = -4.12, p < 0.05), see table 6.   The findings also suggest that males who are less attached to friends were more likely to be involved in victimization than females. We can conclude that gender does moderate the relationship between friendship quality and bullying victimization.

Discussion
The results of the study showed that nearly half of the respondents (46.9%) experienced direct physical/verbal, relational, or sexual victimization from their peers at least 2 to 3 times in a month during the past few months. The adolescents in this study also perceived a high frequency of victimization. These rates of victimization are comparable to the studies done in Philippine (Sanapo, 2017), African countries (Kubwalo et al., 2013;Owusu et al., 2011) and some Western countries such as in England and Wales (BIG 2015), United States (Wang et al., 2009), and New Zealand (Marsh et al., 2010), but slightly higher than the average victimization rates reported in some other studies in Asian countries by Karmaliani et al. (2017) in Pakistan, Sittichai and Smith (2015) in Thailand, and Mok et al. (2014) in Hongkong, Taiwan, and Macao. These findings confirm that bullying exists in schools across geographical areas and cultures.
Concerning the form of bullying victimization, verbal abuse emerged as the most pervasive form of victimization experienced by students in the study. Verbal victimization is defined as a condition when bullies attack the person who has limited power to defend him-or herself in the form of derogatory remarks like calling bad names, mean comments, insulting or teasing in a hurtful way (Wang et al., 2009).
Current results were similar to those found by previous works especially done in Asia-Pacific and South American countries. As reported in the studies (Jolliffe & Farrington 2011;Lai et al., 2008;Sanapo 2017;Silva et al., 2013), "being called mean names and being made fun of" were the most common types of victimization reported by adolescents.
The results of the study based on gender showed that males have a higher percentage of all-types victims than females. Verbal abuse was also the most frequent type of victimization that had been reported by males. These findings are in line with the findings from the existing literature (Carlyle & Steinman, 2007;Crick & Bigbee, 1998;Finkelhor et al., 2005;Fleming & Jacobsen, 2010) in which males were more likely to be involved in bullying and victimization especially in verbal forms (Sanapo 2017;Wang et al., 2009), but differ from the findings in other studies which reported that verbal and relational victimization as the most common type of victimization among females (Owens, Shute, & Slee, 2000;Peskin, Tortolero, & Markham, 2006;Remillard & Lamb, 2005). Interestingly, verbal abuse was the highest proportion of victimization experienced by both male and female students with males experiencing more cases than females. It implies that secondary school students were frequently involved in verbal victimization. Findings from the study support the evidence that gender differences in victimization were found in both Western and Eastern contexts. In terms of the grade level, eighth and ninth graders had higher rates of verbal victimization than those did in seventh graders. Compared to other grades, eighth graders also had more experiences in relational and sexual victimization. Meanwhile, seventh graders had a higher number of direct/physical victimization compared to eighth and ninth graders. These data show that the rate of direct/physical victimization decreases with increasing grade level, but verbal and relational came out as the most prevalent type of victimization for older adolescents (8-9 grades). It seems that the increasing age changes the form of bullying victimization from direct to indirect forms. It is in line with the findings from prior research (He, 2002;Hong & Espelage, 2012;Qiao et al., 2009;Rivers & Smith, 1994;Schneider et al., 2012) that younger students were more likely involved in physical victimization than older students. The current result further supports other studies indicating that verbal victimization is more frequent than other types of victimization among secondary students (Mok et al., 2014;Sanapo, 2017). The findings in this study are also in line with the previous studies (Cook et al., 2010;Fleming & Jacobsen, 2010), which suggest that the prevalence and form of bullying victimization vary across age in both Asian and Western countries.
Consistent with the findings from previous works (Huang et al., 2013;Patton et al., 2013;Rodkin & Hudges, 2003), the study found a significant difference in victimization regarding friendship quality, with moderate effect size. Adolescents who were identified as non-victims expressed the greater quality of companionship, help, security, less conflict, and closeness in peer relationships than those who did not. It means that having a strong connection with friends inhibits the probability of being victimized by other peers. Crick and Grotpeter (1996), stated that adolescents who less attached and feel loneliness in peer relationships are prone to be involved as victims.
The presence of emotional and social support from friends may protect adolescents from being victimized (Bollmer et al., 2005;Goldbaum et al., 2003;Hodges et al., 1999).
As expected, the study also found that the effect of peer relationship on adolescents' victimization experiences differed by gender. Male adolescents got involved in victimization experiences more often than females. It seems that females received benefits more from their friendship than males. A work by Schimidt and Bagwell (2007) posited a plausible explanation for this different interaction. For females, the component of friendship such as help, support, companionship, and security are basic properties in peer relationship and the utmost importance to females' wellbeing, which in turn would protect them from the adverse effect such as victimization. Males, on the other hand, would search for these properties from their friends after negative experience has occurred. The results highlight the importance of positive peer relationship as an effective buffer against victimization on adolescence.
The findings of the study have contributed to the gap in the literature on bullying studies in Indonesia and Asia in general. The study confirmed that bullying victimization is present all over places around the world. The rate of students' victimization experiences revealed in this study will be useful for educators and policymakers in designing prevention and intervention programs by giving more priority to the cases with higher frequency. As found in the study and other prior studies, the role of gender and grade should be taken into consideration when dealing with bullying issues in which male and younger adolescents are more vulnerable to being involved in bullying victimization.
Some limitations were addressed in the present study. First, the data were gathered using a self-reported questionnaire and relied solely on adolescents' reports. Further studies could use multiple sources of information such as involving parents, peers, and teachers to gain a greater perspective on the experiences of victimization. Second, this study only focused on the prevalence rates of victimization while the behavioral ranks of bullies, victims, and bystanders in this area were unexplored. Thus, future research should examine factors associated with bullying and victimization among adolescents in Aceh referring to the high prevalence of victimization reported in this study. Third, the findings are limited to adolescents residing in Banda Aceh, the capital city of Aceh province. Future studies should include large samples representing adolescents from other areas in Aceh and provinces in Indonesia.

E. Conclusion
The present study is one of the few studies that contribute to the bullying research in Aceh-Indonesia by providing the prevalence rates of victimization and its differences in terms of gender, grade level, and friendship quality. Specifically, the results of the study confirm that students who lived and grew in suburban areas like Banda Aceh were not immune to being victimized by their friends as documented in urban areas and other countries. Moreover, evidence from this study suggests that gender and friendship quality differences in victimization were also documented in Aceh in much the same way as in other cultures and populations. Overall, this study gives valuable information on victimization especially in Asian countries and for international comparison studies, which in turn will enhance our understanding of bullying phenomena in adolescence and how contextual factors contribute to the occurrence of this act. BIG. (2015). Bullying intervention group: Current statistics August 2015.