The Importance of Being Attentive to Social Processes in School Bullying Research: Adopting a Constructivist Grounded Theory Approach

School bullying is a complex social phenomenon in need of further exploration regarding its connections to contextual aspects, group norms, and societal structures. This calls for research approaches that can get closer to participants’ experiences and the different social processes involved in school bullying. One such approach is the constructivist grounded theory (CGT) approach, which aims to be attentive to participants’ main concerns and social processes through both analysis and data collection. This approach comes as a theory-method package with its use of a symbolic interactionism perspective. In this paper, I will show how CGT as a theory-method package, as well as symbolic interactionism and sociology of childhood, has been helpful in my research on school bullying (focusing on social structures, norms, and processes). More specifically, I give different examples from the whole research process, e.g., maintaining a focus on participants’ main concerns, the coding process, being guided by sensitizing concepts, addressing issues of social justice and equity — and overall forming and maintaining a theoretically and ethically prepared researcher role. I also suggest that this approach is helpful in dealing with ethical and theoretical challenges when researching topics known to negatively affect people’s lives and wellbeing — and when the social context makes it difficult for participants to address victimizing structures, positions, and processes.


Introduction
Research on how social structures facilitate bullying has revealed how social processes related to normativity and deviance, social status, and issues related to gender and sexuality (e.g., Eriksen & Lyng, 2018;Thornberg, 2015;Rawlings, 2016;Ringrose & Renold, 2010;Walton & Niblett, 2013) are crucial in school bullying. However, school bullying research has long been focused on the individuals rather than on the social structures that facilitate bullying, which some believe explain why school bullying persists (Horton, 2016;Walton, 2015). One way forward is to conduct more research that attends to participants' perspectives and that tries to grasp how they make sense of their social worlds and how their perspectives can connect to wider social practices and conditions (Charmaz, 2021). While the division between quantitative and qualitative research might be exaggerated (Bryman, 1999, Clarke, 2019, most research on bullying has used quantitative methods (Mishna et al., 2009;Patton et al., 2017). In relation to this, we have reason to carry out much more research on school bullying using qualitative methods, as these approaches might be more suitable for examining issues such as perspectives, interactions, nuances, and which social processes are important in school bullying (Mishna et al., 2009;Patton et al., 2017). One way forward within qualitative research on school bullying is to start with "interaction itself, attend to the social contexts in which bullying occurs, ask questions about meanings produced by such interactions and understand these interactions as not solely the province of young people" (Pascoe, 2013, p. 3).
Qualitative methodologies present an opportunity to develop a deeper understanding of the group processes of bullying and participants' perspectives on peer harassment. They are "capable of discovering important discourses and nuances" (Mishna et al., 2009(Mishna et al., , p. 1222. In a review 1 3 of qualitative research strategies used in school bullying research (Patton et al., 2017), some crucial contributions from this research are revealed. These contributions include being able to capture pupils' perspectives, highlighting contextual conditions and processes that influence bullying, diminishing power imbalance within research by positioning participants as experts, and disclosing divergent themes between adults and children. Qualitative research has also contributed with insights of how bullying relates to issues such as race, class, and gender, pointing to the importance of exploring social structures involved in bullying (Patton et al., 2017).
While qualitative research has given us crucial insights into some of the social structures and processes involved in bullying, we must continue to unpack the different social dynamics involved in bullying: for example, how gendered aspects play out in relational bullying (Eriksen & Lyng, 2018) and associated processes of teasing, harassment, and violence (Odenbring & Johansson, 2021). We also need to investigate how adults' assumptions affect what help bullied pupils receive (Bjereld et al., 2021). More research is also needed on which processes affect teachers work on school bullying. Recent studies have pointed to how school design, scheduling, and juridification affect teachers' work Strindberg, 2021) and how their responses affect the bullying path (Wójcik et al., 2021). Another aspect pointed out by scholars (Forsberg, 2021;Strindberg et al., 2020) discloses how pupils' bystander reactions are closely tied to the social dynamics of their social context, such as fear of being singled out. Pupils also reveal, however, examples of successful interventions (Forsberg, 2021). In future research, it is imperative to explore how pupils and teachers view bullying processes, how they are dealt with, whether there can be successful interventions and processes, and what they look like (Bjereld et al., 2021;Forsberg, 2021;Odenbring & Johansson, 2021). More research on these aspects could bring attention to factors of importance for affecting bullying.
Within qualitative research, there are, of course, many different approaches. In this paper, I will attend to the CGT approach (Charmaz, 2006(Charmaz, , 2014 and three of its key aspects (the focus on participants main concern(s); the analytical focus on meaning, actions, and processes, and the use of the symbolic interactionism perspective). These key aspects enable the CGT approach to offer a theory-method package (Charmaz, 2009;Charmaz & Keller, 2016; with methodological tools and starting points that have contributed to my own research on pupils' perspectives on school bullying. Throughout the paper, I will reveal how CGT has been helpful for me when aiming to be attentive to pupils' main concern(s) and social processes in school bullying. I will use some illustrative examples from three different papers in which I utilized CGT. I will also discuss some theoretical and ethical challenges in relation to the CGT approach when being guided by the main concern(s) of the participants.

The Constructivist Grounded Theory Approach: a Theory-Method Package
The CGT approach has developed from earlier contributors of grounded theory (for a more thorough discussion of versions/ variants see Bryant, 2019 such as Glaser and Strauss (1967), Glaser (1978), and Strauss and Corbin (1998). CGT therefore shares some commonalities with earlier contributors, such as coding procedures, data collection, being attentive to and grounding the iterative analysis in participants' main concerns, and viewing them as experts on their social worlds (Charmaz, 2014;Glaser & Strauss, 1967). The focus is on exploring participants' main concerns in relation to a phenomenon and what the participants view as most important. I have chosen to work with the CGT approach in particular because it highlights the role of the researcher as a co-constructor of research and reflexivity, encouraging them to take a clearer position on suitable methods for social justice research (Charmaz, 2014;Charmaz, 2018;Bryant, 2019). CGT also comes as a theory-method package and emphasizes symbolic interactionism, pragmatism, and an interpretative stance (Charmaz, 2009;Charmaz & Keller, 2016;. These underpinnings consider meanings, processes, and a back-and-forth (abductive) reasoning to be important in the research process. These characteristics of CGT make it a valuable toolkit for social justice studies and can aid critical inquiry for social equity (Charmaz, 2017(Charmaz, , 2021Duckels et al., 2019).
As previously mentioned, there is a need for more research that pays attention to how social processes and other social structures involving normativity and deviance, social status, and issues related to gender and sexuality are connected to school bullying (Odenbring & Johansson, 2021;Rawlings, 2016;Ringrose & Renold, 2010;Thornberg, 2015;Walton & Niblett, 2013). In CGT, it is possible to consider how social structural issues are associated with school bullying, as these questions can serve as analytical lenses throughout the research process that the researcher can use "with a critical and analytical eye" (Charmaz, 2021, p. 160; see also Thornberg, 2012) to explore how they might be relevant. This connects to the symbolic interactionist underpinnings and the idea of sensitizing concepts (Blumer, 1969). Sensitizing concepts involve the researcher incorporating questions and theoretical concepts as possible lenses. These concepts are used to explore the relevance of these issues, asking questions about them and whether they fit with the participants' main concern(s) (Charmaz, 2014). By utilizing symbolic interactionism as a perspective, attention is drawn to how perspectives on a phenomenon are constructed in social interaction and in processes. It also assumes that people act toward the world based on how they interpret and define it (Blumer, 1969). This perspective can also address how perspectives, processes, and interactions connect to wider social structural concerns and negotiated orders (Charmaz, 2014(Charmaz, , 2021Duckels et al., 2019). Indeed, this does not mean that symbolic interactionism is always used for theory construction. Whether it should be used as a theoretical perspective informing theory construction is explored throughout the research process. With that said, the symbolic interactionism perspective still informs analysis and data collection procedures by focusing on action and processes which are made explicit in the constructivist version (Charmaz, 2014). In sum, a grounded theory approach seems suitable and relevant for those who want to explore what social processes are made relevant in school bullying as it offers (a) focuses on the main concern(s) of the participants and (b) analytical tools to study meanings, actions, and processes. The CGT approach also (c) utilizes symbolic interactionism and previous knowledge as sensitizing concepts to inform the data collection and analysis. I will now attend to these three key aspects and how they have contributed to my school bullying research. Three of my own papers utilizing CGT (Forsberg, 2017(Forsberg, , 2019Forsberg et al., 2014) will be used to illustrate some of these key aspects throughout the paper.

A Focus on the Main Concern(s) of the Participants
The focus on the main concern(s) and social processes is crucial when using a grounded theory approach. For me, two key analytical tools have been helpful for making me maintain this focus: the back-and-forth movement between analysis and data collection, and the analytical tools in CGT referred to as initial, focused, and theoretical coding (Charmaz, 2014). In one of my studies (Forsberg et al., 2014), I explored what pupils viewed as important when they were bystanders to bullying. This was an interview study with pupils aged 10-15, using individual semi-structured interviews. The study focused on what bullying is, whether the pupils had seen it taking place, how they and others responded, and why they/others responded in that way. Previous studies had highlighted how bystanders could assume a range of bystander roles, and how different factors could influence their bystander roles (e.g., Salmivalli et al., 1996). This influenced how I approached the study and the type of questions I asked during the interviews. For example, I asked pupils how they had responded, how others had responded, and probed for more examples of how the pupils could respond as bystanders. To illustrate the back-and-forth movement between data collection and analysis, I started to conduct several interviews exploring the pupils' perspectives on this issue. I began the data coding soon after conducting a few interviews. In that way, I could start to conceptualize what was going on in the data and update the interview protocol for further data collection. One of the analytical tools that I found helpful for attending to social processes and the participants' main concerns was the initial coding. This is a word-by-word, line-byline coding that focuses on staying close to the data and being attentive and curious about the participants' actions, to begin to make sense of the data (Charmaz, 2014). In line with the influence of symbolic interactionism (on CGT), these codes focus on actions and processes using gerunds (Table 1).
The table below reveals how the codes stay close to the data and emphasize the actions and processes, and on what they depend according to the participant. From this small example of coding, we can already obtain a strong understanding of which processes might be of importance when being a bystander to bullying (e.g., social relationships, friendships, views of seriousness, situational caring). The most telling or comprehensive codes are selected and further explored. In the analysis, various questions are asked: What process(es) are going on? When does the bystander response change? Are there any specific words used that pupils attribute meaning to? How can we interpret what is going on in the data/the process? (Charmaz, 2014).

Familiarization to Explore Main Concern(s)
In line with the grounded theory approach/symbolic interactionist perspective, I have positioned the pupils as expert commentators on their social worlds (Charmaz, 2014). This positioning is crucial and is a reason as to why there should be a focus on participants' main concern(s) from start to end, as this is considered to be an ethical stance (Glaser & Strauss, 1967). An illustrative example can be used from my study on how 11-to 15-year-olds talk about their social worlds and social incidents such as bullying (Forsberg, 2019). Before conducting any interviews, I spent time with the pupils so that they could get familiar with me. Such an approach can also be crucial for getting participants to feel comfortable enough to open up about their main concern(s) (Charmaz, 2014;Mayall, 2000;Raffety, 2015;Ravet, 2007). This approach provided me with some tentative ideas for what to explore in the interviews. For instance, I could inquire about activities I had seen pupils engaged in during breaks. This type of inquiry also allowed pupils not only to reflect on activities they engage in, but also to experience that I am truly curious about their social worlds and perspectives. At one school I saw pupils engaging in a game in the corridor, something that was referred to as "becoming the gay." In this game, some boys were running around and touched each other in a playful manner, and whenever someone was touched, they said, "you are gay." Later, in my interviews, I asked about this and what kind of activity it was. This became the starting point for my study on how 11-to 15-year-old pupils frame different incidents such as bullying (Forsberg, 2019). Previous literature shows how pupils' definitions and understandings of bullying might be different compared to adults (Naylor et al., 2006). Pupils have also been found to view bullying differently compared to commonly used definitions and understand various interactions as drama or conflicts and not bullying (Allen, 2015;Eriksen & Lyng, 2018;Lyng, 2018;Marwick & Boyd, 2014). I was thus attentive to the presence of such processes. I started to compare the different incidents described by the pupils to see what possible patterns and processes could be identified. In this case, I also conducted an initial coding, as well as an incident-to-incident coding, where I compared various incidents (Charmaz, 2014). In the preceding data collection and analysis, I asked more deliberate questions about what bullying is, how one identifies bullying, when something is not bullying, and if so, what. When analyzing and comparing how pupils framed different incidents, I was able to construct a pattern of three different incidents and a process of a contextual definition of harm. Different elements made pupils perceive the incident as bullying, quarrelling, or diffuse incidents. For an incident to be considered harmful and bullying, it had to have taken place repeatedly among non-friends, and the target had to experience both social and emotional harm. Independent of how they framed an incident, they interacted by being grounded in normative identity constructions that use both social categories such as gender and sexuality and locally produced social categories. For instance, the "becoming the gay" was referred to as a normal activity and non-harmful. However, these incidents might still contribute to a social climate in which different ways of being are constructed, and only some pass as normal. This has practical implications if the occurrence of bullying is to be addressed. In order to grasp and address the context and processes in which bullying are produced, it is crucial to explore also those activities not defined as harmful.

Attending to Meanings, Actions, and Processes Utilizing Symbolic Interactionism
In this section, I will show how symbolic interactionism has been helpful in maintaining attention to meanings, actions, and processes. Returning to the (previously mentioned) bystander study (Forsberg, 2014), I found it striking how pupils' bystander responses were talked of as being dependent on the situation. The participants frequently responded that "it depends." My ongoing conceptualization of their perspectives on being bystanders eventually led me to conceptualize five social processes on which bystander reactions depended (relationships (friends and social hierarchy), defining seriousness, victim's contribution to the situation, social roles and intervention responsibilities, and distressing emotions). This fitted well with incorporating a vital concept from the symbolic interactionism perspective -the concept of a "definition of the situation" (Thomas & Thomas, 1928). This concept highlights that "If men define situations as real, they are real in their consequences" (Thomas & Thomas, 1928, pp. 571-572). This means that how people interpret the characteristics of a situation (e.g., social relationships, roles, norms) affects their actions in that situation -in this case, how pupils react as bystanders to bullying. This conceptualization reveals how different perspectives are taken into account, which are situational and processual.
In much of the previous research on school bullying and bystanders, attention has been paid to individual characteristics -in a way putting the blame and responsibility on the young (Horton, 2016;Lunneblad & Johansson, 2021). In my study (Forsberg, 2014), the participants' main concerns, and which aspects they see as important when responding as bystanders, were instead related to the situation they encountered. This highlights the difficulties and the many different social aspects involved when they are bystanders to bullying. Furthermore, it also draws attention to how being a bystander is a fluid position depending on how the situation in question is defined. Rather than remaining in a single position, pupils move between positions due to situations encountered and interpretations of these situations.
Carter (2019) argues that the theory-method package in CGT is important in two ways. It draws attention to how knowledge and perspectives are tools for coping with the world and for offering clues about what to attend to. Carter states: Interactionist theory might lead a researcher to see actions, interactions and relationality as central to human life; see the social world as dynamic and processual, rather than as being deterministically caused by static traits or conditions; understand all human action in the context of the situations in which it occurs; see people as active, thinking, interpreting, meaning-making beings; and understand actions and interactions as both constituting and being shaped by the meanings human ascribe to objects. (Carter, 2019, p. 555) This last part is of crucial importance in school bullying research if we are to attend to social processes rather than viewing people as static categories. It is also crucial in relation to children and young people being the participants in my research and the ways they are considered active meaning-making beings (Corsaro, 2017).

Being Guided by Sensitizing Concepts
Scholars within the CGT approach have pointed out the importance of not dismissing issues related to social equity and justice (Carter, 2019;Green et al., 2007;Roffee & Waling, 2017). A prerequisite for this is that the researcher has knowledge, self-awareness, and is sensitive in not dismissing issues related to social justice (Green et al., 2007). In the CGT approach, social justice issues may be explored by incorporating previous knowledge as sensitizing concepts (Blumer, 1969;Charmaz, 2014;Thornberg, 2012). I find that incorporating questions related to issues of gender or ethnicity, in combination with the analytical questions offered by Charmaz (2006Charmaz ( , 2021, can spark theoretical insights into these issues. This involves asking questions on how aspects matter and how, what a silence means, whether there are inconsistencies, and from whose perspective something matters (Charmaz, 2021). In my own journey within school bullying research, using gender as a sensitizing concept has helped me understand and conceptualize gendered social processes in school bullying. I have tried out different questions during data collection -such as "Some people have acknowl-edged…" or "I have heard about this before. What do you think about who engages in different types?" -to try out my conceptualizations and see whether various issues are relevant.
An example of how gender became relevant in one of my studies was when I analyzed my data on bullying and bystander roles (Forsberg, 2014) and noticed how some bullying incidents were more commonly addressed by girls. I thus started to take notes on this, as I thought the gendering of incidents might be relevant. My sensitivity to the relevance of gender was, of course, influenced by how previous literature highlighted its importance in school bullying (Rawlings, 2016;Ringrose & Renold, 2010;Duncan, 2012). It was something I brought with me into conducting the study, as a sort of sensitizing concept (Blumer, 1969). Through data collection and analysis, participants also pointed to its relevance, and I started to more deliberately include questions on these issues in the analysis (Hadley, 2019). This eventually led to a follow-up study of 40 high school girls' perspectives on bullying (Forsberg, 2017), exploring the issue in greater depth. The study focused on school girls' perspectives on bullying, what it is, whether they could exemplify bullying, and whether and how they thought gender was relevant in bullying (Forsberg, 2017). In my analysis and data collection, the issue of self-confidence stood out as important and connected to different types of bullying, a gendered order, and how pupils tried to secure social value at school. The conceptualization of the data pointed toward the importance of negotiating self-confidence and revealed how girls tried to manage normative standards, resisting being in a victim position, in order to secure social value. This analytically drew attention to how securing a socially valuable position is important, how this also relates to issues outside school, and how gendered aspects are incorporated and negotiated in relation to bullying. This study contributes to problematizing gendered assumptions of "girls bullying in a certain way" (Lunneblad & Johansson, 2021;Eriksen & Lyng, 2018). Instead, bullying is constructed as an identity process located in a complex and negotiated gendered order.
In the study above, I was able to include a focus on gender because it fitted with the main concern(s) of the participants. My sensitizing concepts had thus "earned their way" (Bryant, 2019, p. 118). However, some aspects might be more hidden (Charmaz, 2021), as people might have difficulty talking about issues related to race, gender, and other identity issues (Clarke, 2008;Walton & Niblett, 2013). In a study exploring social categories and social differences related to bullying, Walton and Niblett (2013) show how participants engaged in a sort of self-policing of their statements, where the participants positioned themselves in line with an awareness of the politically correct way to express while still attending to how social difference matters in bullying. Laying out such an analysis highlights how the researcher must be attentive to what is said and what is not said, how it is said, and how aspects related to social categories can be important. As Clarke (2008) suggests, if people cannot easily talk about these issues, this "places a greater burden on the researcher to gather pertinent data" (p. 237). To be able to draw attention to how a theoretical concept or perspective is useful assumes a well-informed researcher, and this can be viewed as ethically important due to how these issues are also found to be crucial (Carter, 2019;Roffee & Waling, 2017).

Possibilities and Challenges with Theoretical Playfulness
Another element in the CGT theory-method package is theoretical playfulness (Charmaz, 2014;Thornberg & Dunne, 2019), using abductive reasoning to consider various ways to theoretically understand the data (Charmaz, 2020;Thornberg, 2012;Thornberg & Dunne, 2019). Theoretical concepts that give the best fit are after careful consideration and exploration applied (Thornberg & Dunne, 2019). Theoretical playfulness can also spark theoretical insights into how the participants' perspectives and main concerns can be connected to larger social and political structures (Charmaz, 2020). For example, people might view themselves as responsible and blame themselves, which might make social causes and solutions invisible. In such circumstances, the researcher has an important role to play in locating the participants' meanings and actions in relevant circumstances that might affect their life conditions, but of "which they may be unaware" (Charmaz, 2020, p. 131).
Returning to the study on school girls perspectives on bullying (Forsberg, 2017), I tried to explore with various theoretical concepts how best to construct an analysis of the data and the girls' negotiations. I first tried to elaborate with concepts from symbolic interactionism, but they did not help me to capture the negotiation and social structural issues. I found the concept of negotiated order (Strauss, 1978) and thought it fitted well with my analytical storytelling. This perspective emphasizes how social order is actively negotiated and ongoing, but that actors are constrained by social structures while negotiating. Applied to my data, the girls' perspectives seemed to be constrained by both a gendered structure and a normative peer structure. These structures identified self-confidence and fitting in as being crucial, but the combination was difficult to achieve and affected their bullying practices. Another dimension in their negotiations of social positions was that the participants tried to avoid positioning themselves as being affected by the gendered order and lacking self-confidence. A kind reader of my paper suggested that this might be understood as an example of face-saving practices (Goffman, 1967). I thus also included this concept to elaborate further on the girls' negotiations.
Scholars have pointed to how gender and other social equity issues are crucial in school bullying processes. However, it can sometimes be a theoretical challenge to attend to these concepts in the analysis because they do not make a fit with the main concern(s) of the participants. From a grounded theory approach, this might be considered a good thing as those theoretical concepts we use should fit the participants' main concern(s) (Charmaz, 2014;Glaser & Strauss, 1967). For researchers who want to contribute to how gender or other social equity issues are relevant in school bullying, this may to be explored in future studies. Even if one study is unable to attend to the relevance of social equity issues, a grounded theory is always viewed as modifiable, as multiple realities and perspectives are assumed. These issues may crop up or other studies might demonstrate how they are relevant and modify the grounded theory.

Ethical Challenges when Participants' Main Concern(s) Are Sensitive and Vulnerable
In the CGT studies exemplified in this paper, I have utilized in-depth interviews. Interviews have been found crucial in voicing participants' experiences of bullying (Mishna et al., 2004;Patton et al., 2017). Overall, interviews can be viewed as a "site of exploration, emergent understanding, legitimation of identity and validation of experience" (Charmaz, 2014, p. 91). As we have seen throughout the paper, in a grounded theory study, the researcher explores the participants' main concerns and the social processes. This informs the whole research process and the final output. It should never lose its connection to the participants' main concerns. The theoretical output should provide a useful perspective for the investigated practice (Charmaz, 2014;Glaser & Strauss, 1967) and be a conceptual contribution that is relevant for the participants (Charmaz, 2014). However, as previously indicated, people can have difficulty talking about certain issues. This can be because of an issue's perceived sensitivity (Walton & Niblett, 2013) or perceptions of one's own experiences as not qualifying as an example of bullying (Mishna et al., 2004;Roffee & Waling, 2017). The strong emphasis on being attentive to participants' main concerns raises an ethical challenge (Priya, 2019) between exploring the participants' main concerns and refraining from gathering "juicy data" (Charmaz, 2014, p. 66). I am interested in exploring participants' main concern(s), and social processes. It is therefore imperative to think about what questions to ask and how during interviews. Questions might make the participants disclose painful events or more information than they might have wanted (Allmark et al., 2009;Mishna et al., 2004;Priya, 2019;Roffee & Waling, 2017).
In my CGT studies discussed in this paper, I have tried to prepare strategies in advance (Allmark et al., 2009) for how to deal with this challenge and avoid not gathering "juicy data" (Charmaz, 2014, p. 66) while at the same time exploring participants' main concern(s), and social processes. This means that I have tried to put the participants' well-being first. However, doing so might affect the possibility to fully explore a social process. This could cause a theoretical-ethical dilemma for the constructed theory if the researcher is not being able to explore fully the main concern(s).
While it might be difficult to know in advance what will happen during data collection, being theoretically informed is one strategy I have used. Being theoretically informed can be considered an ethical stance (Roffee & Waling, 2017) and will also be helpful when doing research with minority or, possibly, vulnerable groups. For me, this means to have a sense of what might be sensitive, relevant areas to explore and what might be a good order to ask questions: for example, asking more general questions like "have you seen bullying taking place" instead of "have you been bullied" to let the participants decide what route to take. I have also always tried to be attentive to the emotional state of the participant. This means that if I get an indication of them not being comfortable talking about something, I have not probed for it. Up until now, I have not found myself in situations where any participant got upset or sad, and where I had to stop collecting data. However, I have always tried to show empathy and build alliances. For instance, some participants have talked about their own experiences of bullying. I have, therefore, at times validated experiences, provided support, or suggested where support can be found (Mishna et al., 2004). I have engaged in a "validation of the children's experiences" as practiced by Eriksson and Näsman (2012) in their interviews with children who had been exposed to violence. I found it important to adopt this strategy, since I did not want to be positioned as a neutral adult accepting bullying behaviors while also demonstrating concern for the participants' well-being (Eriksson & Näsman, 2012).
In my studies, the participants have talked about sensitive and vulnerable experiences, and even if I have not experienced a "person who has fallen apart" (Morse et al., 2009, p. 85), I had to think about how not to end up "leaving the person in a vulnerable place" (Morse et al., 2009, p. 85). I have on a few occasions asked the participants if they have someone to talk to and provided information on where they otherwise can receive guidance and support. I have also asked them what help they would prefer (or if they want it). On these occasions, the schools have been aware about the situations and dealt with it. However, I also brought information cards to BRIS (a chat, email, and telephone helpline where children can get support from trained counselors on any matters. It is free, and the children can be anonymous). In addition, I think it is important to ask children about how to go about dealing with any situation (Mishna et al., 2004), as it might be considered a risk to disclose victimization (Bjereld, 2018;Mishna & Aggligio, 2005).
More generally, insights from the sociology of childhood have been helpful for me in framing and finding an ethically prepared researcher role. The researcher role, influenced by the sociology of childhood, can be of importance in attending to the participants' main concern(s) and how young people go about doing or talking about certain things (Christensen, 2004). In my CGT studies on school bullying in this paper, I have been influenced by the sociology of childhood, as school bullying is an issue that we know affects the health and well-being of many pupils around the world (Chester et al., 2015;Rawlings & Stoddard, 2019). I am also an adult, whereas the research participants are children. Doing research with children has been under discussion in terms of how children are viewed, approached, and involved in research (Christensen, 2004;Mason & Danby, 2011), and how status differences between adults and children might sometimes pose ethical challenges (Mishna & Allaggio, 2005). The sociology of childhood raises important questions about what role the researcher ought to have when gaining access to and trying to understand the social worlds of children (Christensen, 2004;Mayall, 2000;Raffety, 2015). My researcher role can best be described as a reflexive, friendly researcher strategy (Ravet, 2007) or unusual adult (Christensen, 2004), where I have approached the participants non-judgmentally, with an openness and curiosity to their perspectives. I have viewed them as experts on their social worlds, with an avoidance of authority, and, when invited, with an engagement in their activities (Mandell, 1988;Mayall, 2000;Thorne, 1993).

Concluding Remarks
Throughout this paper, the aim has been to show how the CGT approach has been useful for me when aiming to be attentive to social processes and main concern(s) in school bullying. I have also raised how being attentive to participants' main concern(s) can be seen as a way of considering ethics in research and contributing to theoretical development. Furthermore, I have highlighted some ethical and theoretical challenges when investigating main concern(s) and how I have approached these.
More specifically, I have pointed out how three key aspects especially have contributed to my research (the focus on participants' main concern(s); the analytical focus on meaning, actions, and processes; and the use of the symbolic interactionism perspective). The CGT theory-method package with its use of a symbolic interactionism perspective (Charmaz & Keller, 2016;Charmaz, 2009; has been important for me in my research on school bullying, as a toolbox and in order to offer perspectives: "directions along which to look" (Blumer, 1954, p. 7). Methods are tools, but methods still influence what we see and, together with our previous knowledge and experience, also what we can see (Charmaz, 2006, p. 15). The attention on the participants' main concern(s), the analytical focus on meaning-making, and the symbolic interactionism perspective further allow for exploring perspectives on processes, social concerns, and identities, and how these are connected to school bullying. Such attentiveness to social processes is important when contributing to school bullying research, as more work is needed on which social processes are important, how contextual issues matter, and how this connects to social categories and structural issues (Eriksen & Lyng, 2018;Pascoe, 2013;Rawlings, 2016;Ringrose & Renold, 2010;Thornberg, 2015;Walton & Niblett, 2013).
As I have illustrated in my paper, I have focused on the participants' main concern(s) by using broad interview questions, and tried to pay attention to the participants' wordings or expressions and how they define their social worlds. This can be viewed as an ethical stance in conducting school bullying research, and especially with children. This is also where the analytical tools in the CGT approach, such as initial coding and the use of previous knowledge as sensitizing concepts, (Charmaz, 2014) have been helpful. CGT positions the researcher as someone who constructs the analysis, while the methodological tools help ensure attentiveness to social processes and main concern(s) as part of the data collection (Charmaz, 2014). I have shown concrete examples of how these tools helped me to be attentive to social processes in the analysis, to be able to include a focus on social equity issues, and to be both theoretically and ethically prepared.
Some ethical challenges and ways to deal with these when exploring the participants' main concern(s) were also pointed out. I discussed how the emphasis on the main concern(s) could evoke ethical challenges where it is important to put the well-being of the participant before the gathering of data on their main concern(s) (Charmaz, 2014). This could, of course, cause a theoretical challenge for the constructed theory if it is not able to explore the main concern(s). Another theoretical challenge raised was how participants might not easily talk about certain issues, which makes it crucial for the researcher to be theoretically informed to be able to attend to social equity issues. For me, the symbolic underpinnings of CGT and the sociology of childhood have been helpful in positioning the participants as expert commentators of their social worlds. These aspects have also informed my researcher role and been crucial tools for attending to social processes and main concerns(s). It has also given me the opportunity to get familiar with the participants' wordings, processes, and social relations. Furthermore, these aspects helped me to place the participants' perspectives within a social context, encouraging me as a researcher to understand, explore, and maybe then empathize with the difficulties that many young people face at school. This also helped me to understand how these difficulties might connect to their dealing with various social processes and dynamics.
The empirical insights drawn from my research in this paper need to be considered with theoretical sensitivity (Glaser, 1978). Findings are contextual, situated, and modifiable and build on the participants' and researchers' co-constructed perspectives (Charmaz, 2014). However, in terms of usefulness (Charmaz, 2014;Glaser & Strauss, 1967), the concepts and social processes that I have exemplified from my CGT studies are relevant and have practical implications for schools and educators. The studies highlight how social norms, group processes, and identity constructions are a part of school bullying. Furthermore, they address the importance of fitting into the peer landscape, and how definitions of bullying and bystander responses are fluid and dependent on how pupils interpret their social contexts. This indicates that bullying is a complex phenomenon that interacts with a number of social dynamics. In sum, my conceptual contributions highlight how it is imperative to focus on social dynamics if we want to combat bullying. For schools and educators, social dynamics must be at the core of anti-bullying work.
Funding Open access funding provided by Linköping University.

Conflict of Interest The author declares no competing interests.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.