Lower secondary school pupils’ written descriptions of their experiences with bullying and the tendency to seek help

ABSTRACT The purpose of this study was to examine lower secondary school students’ (n = 223) descriptions of bullying and their experiences with the phenomenon. Students at three schools in Norway completed a survey containing open-ended questions. We employed a mixed-methods analytical approach in which responses were coded thematically and then quantified. Findings indicate that students viewed bullying in a manner consistent with prevalent definitions in the literature. However, they were often unsure whether social conflicts they had experienced would be considered bullying. School was clearly featured as the primary arena for bullying, although bullying in free-time activities and on social media was also prevalent. Most students who had experienced bullying had told someone about it, usually their parents or teachers. Students who had not sought help did so for fear of exacerbating the problem or because they felt that they were not in a position to do so.

While use of the term bullying varies across contexts and cultures (Nansel et al., 2004;Smith et al., 2016), certain features of the definition have been prevalent in the literature for decades. These include three principal criteria for describing negative social behaviours as bullying: (1) that the behaviour is intentional, (2) that it is repeated over time, and (3) that there exists an imbalance in the power relationship between the perpetrator and the victim (e.g. Farrington, 1993;Gladden et al., 2014;Olweus, 1993;Smith et al., 2002). Additionally, bullying is often categorized on the basis of the form that it takes, being either physical (e.g. hitting, pushing), verbal (e.g. insults, threats), or relational (e.g. gossip, humiliation, exclusion). It can also occur via digital communication (e.g. messaging), through audio or video recordings, or on social media or other Internet and gaming platforms. Bullying through digital technologies is commonly referred to as cyberbullying (see Smith & Slonje, 2010 for a review).
Logically, another central facet of the phenomenon concerns the position of the different actors involved, including the victim, perpetrator or 'bully,' and witnesses to the behaviour. These various roles have been refined by researchers to high degrees of precision. For example, perpetrators may be surrounded by peers who reinforce their behaviour, assist in causing harm to the victim, or attempt to defend the victim of bullying (Crapanzano et al., 2011). Thus, it is clear that bullying can comprise both individual and group behaviour and bystanders can have both active and passive roles. Group dominance can be one means of demonstrating power over non-group individuals or over other groups with lower social or physical status (Nesdale et al., 2008;Salmivalli et al., 1996).
One of the major challenges with respect to bullying research is how to reliably gauge its occurrence. Self-reports are the most common approach, but have obvious disadvantages, such as social desirability bias or inconsistent interpretations of bullying (Hymel & Swearer, 2015). One solution in survey research is to present a definition of bullying prior to asking questions pertaining to experiences, attitudes, or other areas of interest (e.g. Branson & Cornell, 2009;Cornell & Brockenbrough, 2004). Other studies do not provide participants with a definition of bullying, but instead employ descriptions of specific actions that fit with predetermined criteria in line with a researcher-chosen definition of bullying (e.g. Mischel & Kitsantas, 2020;Rose et al., 2015). A third line of research, which is far less prevalent, concerns direct assessments of participants' own verbal or written descriptions of bullying (e.g. Byrne et al., 2016;Vaillancourt et al., 2008).
In applying this approach, researchers have observed substantial differences in the way that bullying is described by different groups, including teachers, pupils, and parents (Bradshaw et al., 2007;Cornell & Brockenbrough, 2004;Demaray et al., 2013). Interpretations also vary depending on the age of those involved and their previous exposure to bullying (Byrne et al., 2016;Monks & Smith, 2006). A number of studies have similarly shown that both teachers and older students can have difficulties distinguishing bullying from other types of conflicts (Bauman & Del Rio, 2006;Mishna et al., 2005;Vaillancourt et al., 2008). Thus, it is important to consider how individuals and groups operationalize the concept in their everyday lives.
In particular, listening to the voices of lower secondary school students about their experiences with bullying is warranted for several reasons. First, there is a growing emphasis on the importance of recognizing pupils' rights and participation in school policies around the world (e.g. UN Committee on the Rights of the Child, 2009). Second, information gained from students may help to explain a number of unanswered questions. For example, research indicates that bullying increases after the transition to lower secondary school and then decreases again during upper secondary school (Bowes et al., 2010;Espelage et al., 2012;Hong & Espelage, 2012;Nansel et al., 2004). Though not fully understood, evidence suggests that this increase occurs rather quickly in the transition from primary to lower secondary school as youth form new social structures and hierarchies (Farmer et al., 2015;Pellegrini & Bartini, 2000). Inquiry into how students' interpret bullying can contribute to improving school environments and enhancing our understanding of why social relations among children may take a negative turn in this transitional period.
Although survey approaches are common, few studies have sought to assess students' experiences with bullying and their definitions of the concept by analysing written responses to openended questions. A unique study of this kind was conducted by Vaillancourt et al. (2008), in which over 1700 students between the ages of 8 and18 were randomly assigned to one of two conditions. Students were either provided with a 'standard definition' or allowed to use their own definitions when answering questions on bullying. Content analysis was conducted on children's definitions, which were then quantified and analysed across demographic and experience-related variables. The researchers found that students' definitions of bullying rarely included the three definitional criteria described above (i.e. intentionality, repetition, and imbalance of power). Moreover, their findings revealed that children in primary school placed more emphasis on physical and verbal aggression in their definitions, whereas relational bullying was more often described in responses from students in lower secondary school (Vaillancourt et al., 2008).
Along similar lines, a logical approach to gaining a deeper understanding of adolescents' experiences with bullying is through qualitative interviews. Eriksen (2018) conducted interviews with pupils and teachers at three lower secondary schools in Norway and found that the term bullying was seen by both groups as authoritative and rigidly consistent with standard definitions. At the same time, participants had difficulty applying the definition in practice. Teachers perceived that their students used the term too broadly and that its overuse in everyday contexts diminished its relevance with respect to bullying intervention (Eriksen, 2018). A study in Sweden (Hellström et al., 2015) found that adolescents placed emphasis on the victim's experiences as a central criterion for determining if bullying had occurred and were less likely to include repetition as a component of the definition. Dowling and Carey (2013) surveyed students' understanding of bullying and the help seeking behaviour of victims in six Australian schools. Their findings indicate that victims are likely to seek help from different people for different reasons. Students who reported having been victims of bullying perceived teachers as more likely to intervene, whereas parents were seen as more concerned, and friends easier to talk to. However, talking to friends was often in conflict with the goal of getting help to stop the bullying (Dowling & Carey, 2013). This may be one reason why the tendency to avoid seeking help is relatively high in this age group. Indeed, research suggests that as much as a third of adolescent victims of bullying do not tell anyone about this experience (Naylor et al., 2001;Smith & Shu, 2000).

Purpose of the study
Progression from primary to secondary education corresponds to a transition from childhood to adolescence during a period in which many students are at an increased risk of exposure to bullying (Bowes et al., 2010). At the same time, the field has much to gain from listening to the voices of students, who have rarely been given the opportunity to describe their own experiences and interpretations of bullying (Pister, 2014). Such an approach offers the opportunity to better understand aspects of adolescents' everyday life that may answer important questions about changing patterns of bullying in this period of development and contribute to the knowledgebase about how to intervene and prevent bullying from occurring. Therefore, the purpose of this study was to examine lower secondary school students' descriptions of bullying, as well as their personal experiences with the phenomenon by giving them the opportunity to express themselves on the issue anonymously and in their own words.

Participants
Participants were 223 eighth grade pupils drawn from 3 lower secondary schools in southern Norway. The selection of schools was based on school rankings in the annual survey on pupils' learning environment and wellbeing from the Norwegian Directorate for Education and Training. Schools with a high (positive) score on wellbeing and learning environment were intentionally selected. The three schools were among the top ten ranked schools in the region with respect to students' rankings of their learning environment. All three were located in suburban neighbourhoods. The schools varied, however, with respect to the socio-economic and ethnic/linguistic background of pupils. One school, in particular, had a substantial percentage of students with immigrant and minority backgrounds, while the other two schools comprised primarily pupils from the 'majority population' in Norway with respect to cultural background. The schools also differed in size, ranging from approximately 200 to 500 pupils.
Consistent with grade level, 94% of participants were 13 years of age (range = 12-15 years); 122 (54,7%) were boys and 100 (44.8%) were girls. One pupil did not report gender. Prior to attending these schools (grades 8-10), participants had attended a total of 11 different primary schools. The majority (67%) of pupils had attended 1 of 3 primary schools prior to starting at their current school (approximately 40-50 pupils from each of these 3 primary schools), indicating that many pupils had known at least some portion of their classmates before entering their current school.

Questionnaire
In Norway, the use of laptop computers or tablets in school is commonplace, and by eighth grade, most pupils are proficient at typing and editing texts digitally. The study employed the use of a digital questionnaire linked to a password-protected database and posted on the internal network within each school. Brief response forms were used for demographic questions (e.g. age, gender) and questions aimed at gathering limited information about specific experiences. However, the 10 main questions included in the questionnaire were open-ended. Pupils' responses to four of these questions were analysed in the current study: 1). What do you think bullying is? 2). Does bullying occur at school or other places where you spend time? Explain.
3). Describe a specific incident that you have experienced or observed when it comes to bullying. 4). Did you tell anyone about what you saw or experienced? Why or why not? In order to ensure questions would be suitable for pupils to answer in written format during the time allotted, the choice of concepts and wording were developed in collaboration with school leadership and piloted with a subgroup of five adolescents between the ages of 12 and 14 who did not attend the three schools included in the study.

Data collection
We began with a meeting between the researchers and the three head-teachers (i.e. principals or school head) of participating schools. In addition to providing guidance with regard to the content of the questionnaire, they assisted us in determining technical solutions for data collection, and in gaining informed consent from parents. Parents were required to sign a written consent form that provided information about the study and how the information we collected would be used. Pupils were also given the option of not answering the questionnaire on the day that data was collected. All 8th grade pupils at the three schools were invited to participate (response rate = 86%).
Each class was given approximately 30 minutes to complete the questionnaire while researchers visited the school between early December 2018 and mid-January 2019. We began by explaining in broad terms the intention of the study and underlining that participation was voluntary. A script was used to ensure the conformity of information provided to pupils across classrooms and schools. A small number of pupils were exempted from the study due to disabilities or poor Norwegian language skills.

Ethical considerations
There were several ethical considerations to consider due to the sensitivity of the topic. We registered the project with the Norwegian Centre for Research Data (NSD) and followed the guidelines and recommendations of that agency. When participants are over the age of 12, NSD guidelines indicate that emphasis should be placed on gaining the pupil's consent. Whereas parents were asked to return signed consent forms, pupils were provided information about the study both verbally and in written format. Pupils were informed that their formal consent was given by completion of the questionnaire and instructed that they could leave questions unanswered if they did not feel comfortable answering. Pupils whose parents had not provided consent were given another written activity to work on while the survey was being conducted. Information provided to pupils and parents included statements that: (a) all data would be stored anonymously, (b) that they could withdraw at any time, and (c) that it was important to be careful not to use names or identifying information about themselves or others in their responses.

Analysis
The three authors began by reading the entirety of responses to gain a general sense of what the text 'means.' We then discussed the data and came to agreement on the primary themes. Using a content analytical approach, the first author separated students' written responses into smallest meaningful units of information (Weber, 1990) and coded responses in relation to the broader themes identified by the research group. New themes and categories were added as they emerged, and items (i.e. units of information) were recoded to fit the evolving schemata. Additional meetings in the research group were held as the analysis progressed to discuss and adjust the categories and themes as needed.
It became clear early in this process that a procedure was necessary that would allow us to succinctly summarize the data, first and foremost because of the large number of participants. In addition, although most participants responded to all survey questions, their answers were often very brief, ranging from a few words to only a few sentences. Therefore, we chose to quantify the prevalence of different categories using an 'enumerative system' (Lincoln & Guba, 1985, p. 336). A separate database was created for each of the four questions taken from the questionnaire. Categories were coded under separate columns and observations were recorded as being present for a given response (1) or not being present (0). We then calculated the percent of observations based on the total number of participants (n = 223) or the total number of responses falling within a specific category (e.g. number of pupils who reported telling someone about an incident). In many cases, categories were not mutually exclusive, and a pupil's response could be coded in multiple categories. The words of students have been translated from Norwegian to English by the authors.

Defining bullying
Pupils' definitions of bullying were coded into 8 categories. These are presented with examples in Table 1. The two most common categories referred to bullying as (a) 'bothering,' 'harassing,' or 'teasing' others (34% of participants) and as (b) a behaviour that is 'repeated over time' (35%). Crossreferencing these statements, we found that pupils noted the requirement that bullying is repeated over time in 67% of responses for which the statement was also coded in the harassment/teasing category. The following statement illustrates the importance placed on harassment (e.g. 'being mean to someone') and repetition from the perspective of students, 'If a person is mean to you like three times in two months, then it's not bullying, but if it happens almost daily, then it's bullying.' The second most common set of categories comprised references to negative verbal (23%) and physical (23%) behaviours. Relational forms of bullying (e.g. exclusion, spreading rumours) appeared only slightly less frequently (19%). Bullying via social media, text messaging or other digital formats were referenced by 9% of pupils, and among 31% of pupils who referenced relational forms of bullying. A handful of pupils noted that groups as well as individuals (5%) engage in bullying others. The element of an 'imbalance of power,' as it is described in the literature (Farrington, 1993;Gladden et al., 2014;Olweus, 1993), is not easily set into words. Thus, only 4 statements (2%) made explicit references to a higher degree of power held by bullies. All of these referred to physical strength or age, such as pupils from upper grade levels bullying younger pupils.

Awareness of bullying
Only thirty pupils (14%) indicated that they were aware of, witness to, or otherwise experienced bullying 'in the places where they spend time.' In addition, 13% reported that they were unsure if bullying was present in these contexts. While many of these students provided statements that leaned towards a belief one way or the other (e.g. 'I don't think so'), they were far from resolute. Among the 30 pupils who reported experiencing bullying, 15 indicated that it occurs primarily in school (50%), whereas 10% stated that it occurs in free time (e.g. sports activities) and 17% noted that they experienced or were aware of bullying on social media. In stark contrast, 73% of respondents reported that bullying was not something that they experienced or had witnessed. However, among the 163 pupils who stated that they were not aware of the presence bullying, some offered caveats such as that bullying was common in their previous school (6%) or that they often witnessed 'teasing' (8%). In addition, 5 of these 163 pupils (3%) referred to verbal aggression (e.g. 'name calling') and 6% mentioned negative relational behaviours (e.g. 'talking behind someone's back'). In addition, 4% referred to fighting and 3% described other forms of social or emotional issues, such as feeling alone or isolated or seeing others in this state (e.g. "some kids are always on their own). As one student reported, 'I don't feel that bullying happens around me. There might be a few fights, but I never become a part of it. Sometimes it is hard to intervene if you see someone else is being bullied.'

Personal experiences
Pupils were asked to write about a specific incident or situation that they had experienced with respect to bullying. Examples from the various descriptions and how they were coded are provided in Table 2. About half (52%) of the 223 participants described an experience, whereas the other half wrote that they had never been involved in or witnessed bullying -or simply left the question blank. Of the 116 incidents described by participants, 45% were coded as 'unclear' as to whether they represented bullying. The following statement illustrates the challenge of coding a specific incidence as bullying: We were in art class. Suddenly several students started saying mean things to this one guy. I was so shocked I couldn't speak. (. . .) It was like, a lot of people against one. I should have said something, but I couldn't. I guess I'm not really used to these things. They kept going at him for a long time, but calmed down eventually. The person just sat there, laughed a little every now and then, but I could see that he was sad.
In other instances, the decision to code an experience as 'unclear' was due to pupils themselves stating that they were uncertain if the situation was bullying. Other statements described aggressive behaviour or other conflicts, such as fighting (19%) or 'outbursts' that occurred in a single incident (25%) and which did not appear to be part of prolonged situations (e.g. 'called someone a name'), or other social-emotional issues (4%), such as loneliness or difficulty making friends.
Among the 116 pupils that described an incident (including those coded as 'unclear'), 45% portrayed a scenario in which they were themselves the victim. In 46% of cases, pupils described someone else as the victim, while in only 4 responses (3%) did they describe themselves as the person who engaged in bullying (i.e. 'I sent some mean text-messages to someone'). In addition, a number of pupils described multiple aggressors or the involvement of groups of children in the bullying of others (19%). These data are presented graphically in Figure 1.
As was the case for the previous questions, responses were further coded on the basis of location and type. Percentages are again based on the 116 pupils who described an incident, rather than the entire sample. Regarding the question of context, results were similar to those described above concerning the students' 'impression' of the existence of bullying. Approximately half of participants (47%) described situations that occurred at school, 15% referred to digital interactions (i.e. social media, images, text messages and so on), whereas 16% described conflicts that occurred during freetime activities (e.g. football practice). Table 2. Examples of student descriptions of bullying and how these were coded.

Example
Coding I know there has been some bullying in my little brother's class. There was a kid who had to change classes.
• Role: other/witness • Location: school I haven't experienced bullying, but once I saw a boy standing in the middle of a gang of boys and they took away his backpack and his hat and were really cruel to him. Someone I know has been hit a bunch of times (not hard, but just like irritating) for a while and it made him afraid to go to school and now he has really big problems. That person is now in treatment at the hospital and is in a lot of pain. Not with his body, but emotionally, his mood, feelings and body are damaged in a different way. You can't see it on the outside, but the inside is hurt, and he may never be as good as he was. It's not like a bruise that disappears after a while. I feel like the adults should have done something about this before it got so bad.  Finally, with respect to types of bullying behaviour, the most common was verbal aggression (57%), followed by relational forms of bullying (36%). A total of 33 responses described physical aggression, comprising 28% of these 116 participants. It should be noted that some pupils reflected on events from the time that they were in primary school. Moreover, pupils often wrote about situations that occurred across multiple contexts and in different forms. Although categories are not mutually exclusive, results nonetheless indicate consistent trends across responses, as presented in Table 1 and Figure 1.

Help seeking
Data on pupils' help seeking behaviour is based on responses to the question of whether they told anyone about the experiences described above. In sum, 84 (72%) of pupils who reported an incident (n = 116) stated that they had told someone, whereas 31 (27%) wrote that they had not. Thirty-two percent of the 84 pupils who had told someone did not specify to whom they reported the situation, 39% stated that they had told their teacher or other school personnel (e.g. school nurse), and 16% had told friends or siblings. The most common group to whom pupils turned was their parents, with 45% of those who had told someone (84) reporting that they had told one or both of their parents.
Few students gave an indication as to why they chose to seek help from a given person. Although we did not ask this question directly, the tendency among students to seek help from adults implies that pupils see them as being more capable of resolving the problem than peers. At this stage of schooling, adults still retain a large degree of authority. Thus, many students appeared to equate 'telling someone' with reporting the issue to an adult. Other statements reflected the importance of having trust in the person to whom they confided, 'I always tell what happens to my mom and dad, I always tell them because I trust them.' The feeling of trust likely played a role in the decision to tell friends as well. For example, one student wrote, 'I needed to tell someone, and that person is a very good friend of mine who supported me and was there for me.' Albeit rare, an alternative reason for telling peers was to influence their social relationships: 'I think everyone who's with him should know that he has a sick temper and is unstable,' wrote one student about the need to warn others about the aggressor in the situation that he described.
Thirty-seven of the 84 pupils who wrote that they sought help, also wrote statements about the outcome of having told someone about the incident. These were coded into 4 categories of outcomes. Three of these concerned statements that the person they told (a) helped resolve the conflict (19% of 84), (b) did not help the situation (9%), or (c) made the situation worse (4%). In addition, 7 of these 37 responses indicated that the problem resolved itself, for example, because the children involved changed schools.
Forty-eight (40%) of the 123 pupils who described an incident, wrote why they did or did not seek help. Among the 30 pupils who wrote that they had reported the incident, these explanations were generally brief. For example, 16 were coded under a category termed, 'getting access to help' and 9 were connected to pupils feeling that they were obligated to do so (e.g. 'bullying is not right'). Five students (16%) wrote that it was necessary to avoid the potential negative outcomes if the behaviour did not stop. Among the 18 pupils who wrote reasons why they did not report the situation, the responses were quite varied. We broadly grouped them into three categories: (a) the situation was not that serious, it resolved itself, or others had already reported it (8 responses), (b) the belief that one should not 'tell on others' and try to resolve issues on their own (4 responses), and (c) fear of making matters worse and/or feeling 'too weak' to take action (9 responses). Responses from some students fell into multiple categories, thereby exceeding the number of pupils who did not seek help.

Discussion
Findings indicate that Norwegian students in their first year of lower secondary school view bullying in a manner that is mostly consistent with traditional definitions of the phenomenon (e.g. Farrington, 1993;Gladden et al., 2014;Olweus, 1993;Smith et al., 2002). However, the aspect of an 'imbalance in power' was largely absent from their descriptions and there was a high degree of uncertainty as to whether their own experiences would be considered bullying. While most students reported little direct experience with bullying, most of those who had exposure to it had told someone, usually their parents or teachers. Reasons for not seeking help were primarily related to the fear of making the situation worse or because they felt that they were not able to do so. Perceived outcomes of seeking help were mixed, where several pupils described adults' reactions as having no effect or a negative impact on the situation.

Descriptions and experiences
Students' definitions of bullying focused largely on the idea that negative social behaviour and aggression must be repeated over time to be considered bullying. While not always stated explicitly, the concept of 'intentionality' was also present in participants descriptions of bullying (e.g. teasing for no reason). References to an imbalance in power were far less prominent; however, the concept was evident in a handful of statements concerning observable characteristics, such as physical strength, size, or age differences. Although some studies indicate that students' definitions of bullying are largely consistent with the way it is traditionally defined in the literature (O'Brien, 2019;Pister, 2014), most studies suggest that adolescents rarely include all of these definitional criteria (Byrne et al., 2016;Cuadrado-Gordillo, 2012;Vaillancourt et al., 2008).
Arguably, variation in understandings of the concept of bullying influence perceptions of negative social behaviour and can lead to either an increase or decrease in students' willingness to seek help. For example, Vaillancourt et al. (2008) found that the definition used by students had an impact on the prevalence of bullying reported, where students who used their own definitions reported higher rates of victimization. Thus, adherence to a strict definition of bullying may prohibit students from acknowledging harmful actions associated with difficult-to-define negative, social behaviours. In the current study, it seems that pupils had learned the primary features of the 'standard' bullying definition. Nonetheless, there was little coherence between these features and their descriptions of personal experiences with bullying. Approximately one-fourth of descriptions focused on conflicts such as fighting or isolated acts of aggression, and the expectation that the behaviour must extend over time was mostly absent.
An unexpected finding of the current study was the large percentage of students (73%) who reported not experiencing bullying or being unaware of its existence at school or in other arenas where they spend time. That being said, about a third of responses to this question offered examples of other negative behaviours that were witnessed regularly, such as teasing, name calling, or other relational and social-emotional stressors (e.g. isolation and loneliness). Studies based on self-reports indicate that the prevalence of bullying in lower secondary school varies between approximately 5% to 25% depending on the region, time frame (e.g. within the last week or month), and manner of measurement used (e.g. Baly et al., 2014;Espelage et al., 2012;Hymel & Swearer, 2015). Given that the survey question posed to participants was open with regard to time-period, context, and specific form of bullying, we would assume the percentage of those reporting the existence of bullying to be considerably higher. This finding appears to support the conclusion that students applied a strict definition of bullying when assessing its prevalence.
Consistent with previous research (e.g. Harris et al., 2002;Hicks et al., 2018), students' reported experiences with bullying as occurring primarily within the school context. Although bullying via social media or during free time activities were prevalent, these interactions were often an extension of the social processes taking place at school. The significance of school is not surprising, given that this is where students spend most of their time and they often have little control over with whom they are grouped on a daily basis.
In addition, as Farmer et al. (2015) point out, the transition of young people entering school from different elementary schools creates a mix of students with shifting peer affiliations and social stressors that may give rise to bullying. While teachers and parents are often more concerned about physical and cyberbullying, evidence suggests that these are typically the least frequent forms of bullying reported, and that social and verbal forms of bullying are the most common (Hymel & Swearer, 2015;Mischel & Kitsantas, 2020;Vaillancourt et al., 2008). Our findings are consistent with previous research in this regard. Even after accounting for overlap between digital and relational forms of bullying, verbal bullying was referenced at nearly twice the rate. This finding has implications for bullying prevention, as attention can easily be misdirected away from common forms of 'face-to-face' bullying because the ubiquitous nature of the behaviour may make it more difficult to recognize.

Perceived risks of seeking help
Despite difficulty operationalizing bullying, the majority of pupils (72%) had nonetheless told someone about the situation they had experienced. Students most often sought help from their parents (45%) and teachers (39%). Dowling and Carey (2013) similarly found that students primarily turned to parents as a source for help. This is a positive finding, given research suggesting that support from parents is vital to preventing bullying and reducing its impact on victims (Holt & Espelage, 2007). Moreover, parents of adolescents often have limited knowledge about what happens at school and bullying typically occurs in contexts where there is little adult supervision (Hymel & Swearer, 2015). Thus, students' willingness to seek help from their parents can serve to make these concerns more visible, remediable, and preventable.
In contrast to the current study, Dowling and Carey (2013) found that students considered teachers among the hardest sources to talk to about bullying. A study by Oliver and Candappa (2007) also found that students were reluctant to seek help from teachers primarily due to the fear that teachers would fail to respect their confidentiality.
These and similar studies have found that as students grow older and transition from elementary to secondary education, they tend to resist seeking help from adults (e.g. Boulton et al., 2017, April;Trach et al., 2010). Conflicting findings from the current study may be related to the fact that participants were still in an early phase of secondary education and some of their experiences reflected events from primary school. Consequently, it is not certain that students would have sought help in their current school or will continue to do so as they grow older.
While findings regarding students' inclination to seek help are generally positive, it is concerning that almost one-third of participants who described bullying-related experiences had not told anyone about the incident and more than one-sixth of these students had only confided in friends or siblings. While close friends are often considered the easiest to talk to about bullying and may make victims feel better (Naylor et al., 2001;Oliver & Candappa, 2007), they are less likely to be in a position where they can help the victim to resolve the conflict. It is revealing that the two primary reasons students gave for telling someone about their experience were to help resolve the issue or because they felt that they had a moral obligation to tell (i.e. 'it was the right thing to do'). On the other hand, several students reported the desire to find a solution on their own for fear of making matters worse or because they lacked the confidence to seek help. Clearly, there are many factors at play under these conditions and students face both emotional and social risks by drawing attention to bullying, particularly when they are the victim.
Although the number of students who reported on the outcomes of having sought help were limited, a substantial portion of these responses indicated that telling an adult did not have an effect on the situation or actually made it worse. Thus, it is plausible that students' reluctance to report bullying is related to their prior experiences. Cortes and Kochenderfer-Ladd (2014) found that students were less likely to seek help from teachers regarding bullying if they believed that teachers would punish the perpetrators, potentially due to fear of repercussions. On the other hand, teachers may fail to intervene because they see bullying as a normal part of adolescence or because they have difficulty assessing the severity of the problem (Kochenderfer-Ladd & Pelletier, 2008;Mischel & Kitsantas, 2020). Evidence suggests that adolescents are more likely to seek help from their teachers when they believe that teachers will become actively involved, inform parents, and separate the students (Cortes & Kochenderfer-Ladd, 2014). Furthermore, help seeking is more likely if pupils believe that teachers will be effective and fair in resolving conflicts (Aceves et al., 2010). Together with these findings, the current study underlines the importance of establishing trust between teachers and students and ensuring that the risks that students face when deciding whether to report bullying do not outweigh the potential benefits to all students involved.

Limitations
There are several limitations to the study that are worthy of note. Although students' responses were anonymous, social desirability bias is possible and may have influenced results. In addition, the use of open-ended, written responses limited the participation of students who struggle with writing or who are less willing to express themselves in a narrative format. The mixed-method approach, in which qualitative, thematic categories were quantified represents both challenges and opportunities. An advantage of this approach was that it allowed us to broadly summarize the perspectives of pupils. However, there are clear limitations in interpreting the variability and nuances of responses, as well as estimating the validity of these findings for other populations. Moreover, differences among the three schools were not explored in the current study as our goal was first and foremost to ascertain a broad picture of students' perceptions of bullying within this region. More in-depth analysis of school differences and the influence of contextual variations is recommended for future research along these lines. Finally, students included in this study may not have had time to experience the social and developmental changes that contribute to an increase in bullying in lower secondary school. We recommend that future research pursue a longitudinal approach to capture the trends and patterns of bullying, as well as potential changes in attitudes or perspectives that are likely to occur as students move from childhood to adolescence.

Conclusion
This study explored social, contextual, and personal aspects of bullying in lower secondary school as they pertain to students' understanding of the concept, as well as overall trends regarding how bullying is experienced by pupils. Findings contribute to the knowledgebase and are relevant for teachers, researchers, and policymakers. From a policy perspective, our findings suggest that low rates of school-level bullying recorded on self-report measures do not necessarily imply a healthy or safe school environment. As our findings illustrate, many closely interrelated social and behavioural challenges faced by young people do not fall within traditional assumptions of the bullying construct. Consistent with previous research (e.g. Byrne et al., 2016;Cuadrado-Gordillo, 2012;Vaillancourt et al., 2008), our findings demonstrate that many pupils find it hard to recognize bullying in their daily life. Thus, we support the long-standing recommendations of Arora (1996) who advocates for emphasis on whole-school environments that focus on eliminating negative behaviours that may cause young people distress, regardless of whether they fall under a given definition of bullying.
A positive finding of the current study is that most students told others about their exposure to bullying. Of greater concern, however, is the perception among several of these students that sharing this information may have worsened or done little to improve the situation. Unfortunately, the data collected in this investigation only give a broad picture of the chain of events regarding parents and teachers' responses to being told that a child or adolescent was bullied. More in-depth research is needed to ensure that adolescents in this vulnerable period continue to seek help and, more importantly, that the help that is provided does not cause more harm than good.
Most important, the current study contributes to the knowledgebase by emphasizing the voices of students on the issue of bullying, an approach that has been largely underutilized in the research literature (Pister, 2014). Despite confusion about the 'right' definition, when and how to access support, and the causes of bullying; there is nonetheless a sense of clarity and sincerity in their descriptions that cannot be ignored. The words of one such student summarize the urgency and emotional strain caused by bullying within this context: Bullying sucks. Like, when you're in secondary school and feel that everybody hates you, that they are against you, that is pretty rough. When people stare at you, call you names, laugh and point at you even though you haven't done anything funny, just because you're different.