Development of Peer Bullying Scale (PeBuS)

The aim of this study is to develop a scale that can be used to detect bullying behaviors that can be seen in interpersonal relationships. The sample of the study consists of 381 randomly selected teachers working in primary and secondary schools in the central district of Edirne, Turkey, in the 2021-2022 academic year. In the research, the “Peer Bullying Scale”, which was prepared by the researcher and consisted of 50 Likert type items with 4 options, was used as a data collection tool. Varimax rotated exploratory factor analysis was used to determine the sub-dimensions. To determine reliability, item-total correlation, item residual correlation, Cronbach and Rulon coefficients were calculated. The construct validity of the scale was tested with Confirmatory Factor Analysis. As a result of the statistical processes, it was seen that the scale was valid, reliable and usable.

than social cohesion. Bullying can cause teens to gain popularity, but it can also cause them to lose their likes (Hensums et al., 2023).
It is known that 34% of children have been exposed to bullying in the last month (World Health Organization, 2019). Bullying with serious lasting consequences is a common and important problem among children and adolescents.

Bullying concept
Peer bullying, which is defined as the continuous negative actions of one or more students against another student, is an example of undesirable behavior. Bullying is an important behavioral problem involving persistence, power imbalance, and willfulness among peers. Bullying is a type of violence that causes one or more students to constantly and intentionally disturb weak students and where the victim cannot protect himself.
The frequent and systematic abuse and assault of some children by other children has been described in literary works, and many adults have personal experience of this from their school days. Bullying behavior usually occurs without explicit provocation and can be considered as peer abuse (Olweus, 1994). Bullying is the physical and psychological pressure applied by stronger individuals or groups to harm less powerful individuals and groups (Olweus, 1993;Gürhan, 2017). For a behavior to be considered as bullying, three criteria must be met: (1) Aggressive behavior with the intent to harm, (2) Continuity, (3) Power imbalance between the bully and the victim (Olweus, 1993). In addition to the power imbalance in bullying, there is continuity and systematic abuse of power (Rigby, 1999). Bullying is a form of interpersonal violence, but bullying is different from aggression. The mechanisms leading to bullying are related to establishing and maintaining social dominance (Olafsen & Viemerö, 2000).
Peer bullying is the physical, verbal, psychological attack, intimidation, and violence behaviors that the stronger child (bully) who has power inequality among them without provocation, aims to inflict fear, anxiety or harm on the other (victim) and repeatedly.
Bullying can be classified as physical, verbal-non-verbal, direct-indirect. Physical bullying is the direct use of force. Verbal bullying is hurting through words. Indirect bullying includes behaviors such as social isolation (Elliott, 2021). Bullying refers to direct or indirect aggressive acts that can be classified as physical, psychological, verbal, sexual, cyber (Özbek & Taneri, 2022).
Many researchers have presented different views on the types of peer bullying. While Haskaya (2016) diversifies peer bullying directly and indirectly; Olweus (1991) discussed peer bullying in three categories: physical bullying, verbal bullying and indirect bullying through social aggression. Social aggression is seen as ostracism, spreading rumors, or using "relationships as a weapon" (Simmons, 2002).
Various types of bullying have been classified as a result of a series of studies (Sulak & Altuntaş, 2022): physical bullying (fighting, kicking, punching, hitting), verbal bullying (teasing, swearing, threatening, spreading bad rumors), social bullying (intentionally excluding from a game or group, not inviting to class activities), sexual bullying (i.e. making sexual comments or engaging in sexual acts); cyberbullying (sending annoying and offensive electronic messages over the phone or computer).

Causes of bullying
The frequency of bullying varies depending on the country, culture and age of the victim. The personality of the child, his family and cultural characteristics, and the role of the social environment he is in also have an effect on bullying behaviors.
Studies indicate that bullies more often choose children and young people who have developmental, emotional, physical differences, health problems, and who do not have strong social relationships as the target of bullying.
The cause of peer bullying may not be the fault of the bullied child or conflict (Olweus, 1993). If the cause of peer bullying is not peer conflict, opinions on the question of what is the reason are varied. Peer bullying, for example, is thought to be for social dominance (Pellegrini & Long, 2002). Children with insufficient psychological resilience have difficulties in peer relations and may become bullies or victims (Gün & Gültekin Akduman, 2022).
Traditionally, bullying is maladaptive behaviors that result from functional deficiencies or inadequacies in the individual. Inability to cope with situations socially due to incorrect or limited social information processing, hostile association style, or poor emotion regulation skills leads to bullying (Hensums et al., 2023).

Effects of bullying
Being bullied by peers can cause significant problems in the physical, mental, social, academic and professional development of children and young people. Bullying causes the victim to feel bad and worthless in social situations and impairs academic achievement.
Bullying, which is a type of violence, involves repeated physical, psychological or social harm, usually occurs in schools, online and in environments where children gather (World Health Organization, 2019). Bullying can change both the victim's and the bully's self-perceptions. At the end of the bullying process, while the bully gains certain gains (intra-group status, money, goods, etc.), the victim experiences a sense of humiliated and helplessness (Gümüş & Talu, 2022). Being exposed to bullying has negative effects on children physically, mentally and socially. It can cause emotional effects such as depression, anger, anxiety, as well as somatic complaints, especially stomachache and headache (Akcan & Odabaş, 2022).
The three actors of bullying are the bully (bullying the other person), the victim/victim (the one who is being bullied) and the spectator (witnessing/bystander of the bullying). Studies have shown that bullying has serious consequences for the bully, the victim, and the audience. All actors of peer bullying experience academic, physical, psychological, social and emotional problems, and may face serious consequences such as low academic achievement, lack of selfesteem, social incompatibility, depression, inferiority, insomnia, and suicide in the short and long term (Özbek & Taneri, 2022).

Measures of bullying
Detecting bullying, providing solutions and preventing it are important for the healthy development of children and adolescents. Families and educators play a very important role in this regard. Current studies on bullying are based on the idea that peer violence is preventable. Preventive strategies for bullying are concentrated in the field of psychology and educational sciences.
This problem, which starts in the pre-school period, should be recognized first and preventive measures should be taken. In order to reduce bullying, it is important to acquire positive attitudes in children, identify risk groups and create anti-bullying programs.
Bullying in schools has a negative impact on children's perceptions of safety and school experiences. Since they spend a significant part of their days at school with their friends, most children experience the fear of being bullied by their peers with whom they are in constant interaction (White & Loeber, 2008).
Children who are victims of peer bullying need to receive early intervention (Erim, 2022). Measures of bullying; Families should be made aware of bullying, inclusive seminars should be organized about bullying, children who are prone to violence should be directed to social activities, studies should be made to establish empathy, punitive sanctions should be increased, informative publications should be made about bullying in mass media (Akpınar & Akpınar, 2022).

Method
The aim of this study is to develop a scale that can be used to detect bullying behaviors that can be seen in interpersonal relationships.

Data collection
In the research, a questionnaire for demographic information and the "Bullying Scale" prepared by the researcher were used as data collection tools. The scale consists of 50 Likert type items with 4 options. There are no reverse-rated items in the scales. Subscale scores were calculated by taking the mean of the items. The scale score was calculated by taking the mean of the subscales. All these scores range from 1 to 4.

Data analysis
Varimax rotated EFA (exploratory factor analysis) was used as scale development statistics to determine sub-dimensions. To determine reliability, item-total correlation coefficient and item remainder correlation coefficient, Cronbach and Rulon coefficient were calculated. Similarly, t-test was conducted between the upper and lower quartiles to determine the strength of discrimination. As a result of the statistical processes, the scale consisting of 5 subscales has been proven to be valid, reliable and usable.
After the scale was structured with EFA (exploratory factor analysis) and item analysis, its construct validity was tested with CFA (Confirmatory Factor Analysis). CFI (Comparative Fit Index) and RMSEA (Root Mean Square Error of Approximation) fit indices were used to test the suitability of the DFA model. Considering these criteria, DFA models were created. As a result of the statistical processes, it was seen that the scale was valid, reliable and usable.

Results
The statistical method used to determine the construct validity of the scale is exploratory factor analysis. The data were found to be suitable for factor analysis with the KMO (0.976) and Barlett (Chi-Square=19264.100 df=1225 p=0.000) tests (Table 2). In the analysis, which was carried out in such a way that the components with an eigenvalue of 1 and above were selected, 6 components were formed. The 6 components created in the analysis using the principal components method explain 72.345% of the total variance (Table 3). Factors and the items they contain were determined according to varimax rotated factor analysis. It was determined that the scale had 5 factors (Table 4). Open Journal for Educational Research, 2023, 7(1), 1-12. ______________________________________________________________________________________________ 7 The 6 components obtained by factor analysis resulted in 5 factors. The factor structure of the scale is as follows; F1-1: Psychological direct bullying, F1-2: Physical violence, F1-3: Psychological indirect bullying, F1-4: Physically disturb, F1-5: Physical infliction (Table 5). The relationship between factor totals and scale total was tested with item-total correlation and item-remainder correlation coefficients, and internal consistency between factors was tested with Cronbach  (0.870) and Rulon (0.907) coefficients. According to the results of this analysis, it was determined that all the factors in the scale had internal consistency (Table 6). This process, which was carried out according to the scale total, was repeated separately for each of the 5 factors. Summary statistics on these analyzes of internal consistency are as follows: Cronbach's coefficients F1(0.970), F2(0.960), F3(0.943), F4(0.907), F5(0.793); Rulon coefficients F1(0.977), F2(0.973), F3(0.925), F4(0.904), F5(0.781). According to these results, it was seen that all factors and all items had internal consistency.
Participants were ranked in descending order according to the scale factor total, and it was determined which participants were in the upper or lower quartiles according to this ranking. The difference between the upper and lower quartiles was analyzed using the t-test statistical technique. This process, which was carried out according to the scale total, was repeated separately for each of the 5 factors. According to these results, it was seen that all factors (Table 7) and all items were discriminative between participants with high and low opinions on peer bullying problems. All Pearson correlation coefficients among all factors are statistically significant. Normality tests, kurtosis and skewness coefficients showed that the data were normally distributed. All relations between factors are at a significant level (Table 8). After the scale was structured with EFA (explanatory factor analysis) and item analysis, the scale's 5 sub-dimension structure was confirmed as a result of CFA (Confirmatory Factor Analysis). The model was validated by comparing the indices ((χ2(1158) = 3647.113 p = .000 χ2 /df= 3.149; TLI= .861; NFI= .820; CFI= .869; and RMSEA= .075 (90% CI 0.072 0.078); GFI = .700; AGFI = .670) with the criteria. All items loaded significantly on 5 factors of the scale. Figure 1 shows all the factors, items and values of the scale.

Conclusion
Statistical analyzes show that the scale has construct validity, discriminant validity and internal consistency reliability. As a result of all statistical analyzes, it was determined that the Peer Bullying Scale, which consists of 5 factors and 50 items, is valid, reliable and usable.
There is no reverse scored item in the scale. High scores on all factors indicate that peer bullying is a high-level problem. It is hoped that the scale will help to better understand the difficulties experienced especially in schools. Thus, it may be possible to make peer relationships healthier.