Lessons Learnt From Conducting an Instrumental Ethno-Case Study in a Mainstream Secondary School

Abstract

The aim of this paper is to provide an insight into the lessons learnt as the primary researcher during an instrumental ethno-case study research project that took place in a mainstream secondary academy that set out to examine the physical education (PE) experiences of adolescent, white, pupils in a ‘typical’ working-class secondary school (Ayrefield Community School – ACS). With the term instrumental ethno-case study used to reflect the non-continuous nature of the 34 days spent in the school over a 16-month period, the structure and content of the paper aims to provide a greater theoretical and practical understanding of this relatively nuanced and contemporary approach to qualitative research in education. Determined that the study design remained based in ethnography, consideration is also given to the impact of professional and personal time constraints on an ability to engage in a truly immersive ethnographical study. In this regard, the paper also aims to provide additional guidance on the design and implementation of the study and the recording and interpretation of the data for any researcher with similar aims and/or time constraints. In addition, the paper also utilises Norbert Elias’ sociological concept of involvement and detachment as the lens through which the juxtaposition between objective researcher and experienced mainstream secondary school teacher was explored and managed throughout the duration of the study. Overall, therefore, the paper aims to provide a theoretical justification for the use of the term instrumental ethno-case study design, additional guidance and lessons learnt on the design and implementation of this approach, and further detail on the extent to which the potential conflict between objective researcher and former school teacher was acknowledged and managed. The project itself primarily set out to examine the physical education (PE) experiences of adolescent, white, pupils in a ‘typical’ mainstream working-class secondary school (Ayrefield Community School – ACS) and in doing so also sought to explore the attitudes of the pupils in relation to their views on qualifications and education generally, how they spent their leisure time, and more specifically, how their own actions and relationships with their male PE staff came to influence the content and delivery of their PE provision. Therefore, reflecting on the use of the traditional case study methodologies of covert and overt observations, informal guided conversations, and focus group interviews with male pupils and school staff, the paper highlights and examines the practicalities and considerations associated with the selection of, and access to, a ‘typical’ white, working-class school, how care was taken to encourage ‘natural’ behaviour during covert and overt observations, the specific nature of informal guided conversations with pupils and staff, and the techniques utilised in order to facilitate focus group interviews with these male adolescent pupils.

Introduction

Recent data suggests that sports participation and levels of physical activity (PA) continue to reflect patterns of social stratification, with a clear gap existing between social class groups in the UK in relation to their participation in sport and active leisure (Sport England, 2022). The reasons for this participation anomaly have consistently been linked to the differing lifestyles and opportunities to which young people from working and middle-class backgrounds are exposed – a social process that has subsequently been shown impact upon rates and types of sport and PA participation in adulthood (Haycock & Smith, 2012; Wheeler et al., 2019). More specifically, studies show that working-class children are more likely to develop narrow, class-related leisure profiles and sporting repertoires during their childhood that serve to limit the likelihood of them remaining physically active in adulthood (Engstrom, 2008) – especially when set against the experiences of their peers from higher up the social scale. In relation to this, one of the main aims of physical education (PE) in mainstream schools is to develop the range of skills and knowledge for all pupils and widen their sporting repertoires as part of an attempt to promote long-term participation throughout their lives (Evans & Davies, 2010). However, not only has PE provision in British mainstream schools been shown to be unsuccessful in widening sporting repertoires with a view to promoting working-class pupils’ sporting/ability development and lifelong sports/PA participation (Evans & Davies, 2008), some suggest that the subject of PE in its current form may even be perpetuating the social class difference that has been shown to exist between social class groups (Dagkas, 2011).

As part of an attempt to explore the influences behind these claims, the study set out to examine the extent to which the wider social background of white, working-class male pupils combined with the actions and attitudes of, and relationships with, their PE teachers, came to impact on the way these adolescent male pupils influenced and experienced their PE curriculum and lessons. The study also aimed to examine the impact that school PE then had on their sporting repertoires and participation in sport/active leisure outside of school with a view to gaining a greater understanding of how and why socially stratified participation patterns continue exist between adults from different socio-economic backgrounds.

Selecting the Case

Several mainstream ¹ secondary schools were initially highlighted as potential cases through which to adequately and accurately explore the key objectives of the research question (Denscombe, 2010) and therefore provide accurate and sufficient opportunity to explore the short and long-term influences that were acting on white, working-class male pupils and the extent to which the short and long-term relationships of which they were a part came to impact on the ways that they viewed and experienced school generally, and PE more specifically. In this regard, care was taken to ensure the case school was not selected purely for convenient intrinsic and/or personal reasons (Simons, 2009) and so in order to avoid this, a structured analysis of key factors was utilised in order to generate a list of objectively suitable cases which focused upon the quantifiable data linked to the percentage of pupils on roll identified as white-British, the most recent Ofsted rating (1–4) of the school, the most recent GCSE (A*-C) pass rates published, and the ranking of the three most local lower social output areas (1 - 32,844) that made up the school catchment area. All schools within a 20-mile radius of my home address that emerged as suitable following this process were ranked in order of suitability with the top three contacted in order to ascertain their willingness to be involved in the study. Ultimately, ACS was selected from this range of possible alternatives (Rowley, 2002) as it emerged clearly as the most suitable ‘typical’ case due to the fact that it stood to epitomize a much broader range of cases (Bryman, 2012; Denscombe, 2010).

The Case

The case-study school (ACS) was situated in the northern village of Ayrefield and deemed suitably ‘deprived’ to meet the criteria of the study. The three Lowers Super Output Areas (LSOAs) ³ surrounding the school site were ranked between the range of 450–500, 825-875 and 1125-1150 respectively for social deprivation out of a possible 32,482 LSOAs nationally (ONS, 2014). Nearly a quarter of all residents in Ayrefield had no formal qualifications, and twice the national average of residents were in ‘bad’ or ‘very bad’ health. Further to this, nearly two-thirds of all households in the area were defined as deprived in either one or two of the four indicators of deprivation (e.g. one adult unemployed or no inhabitant has a level two qualification) (ONS, 2014). 11.5% of current residents had either never worked or were classed as being long-term unemployed and over twice the national average (NA) of adults were currently claiming key working age benefits (36% - NA 15%) with almost three-times that number on incapacity benefit (20% - NA 7%). There was also a strong prevalence of ‘white’ ² people living in Ayrefield with 1330 of the 1389 total residents classing themselves as white-British. Out of the 965 pupils on roll at ACS, only 1% of students considered their first language to be other than English: ‘most pupils were white British with a distinct lack of pupils from ethnic minorities on roll’ (Ofsted, 2012, p. 5). In addition, 45% of all pupils in Y11 at ACS were officially defined as being ‘disadvantaged’, half (49.4%) of all pupils had been eligible for free school meals in the last six years and the most recent Ofsted report stated that ‘the school had faced challenges in the recruitment and retention of teachers’ (Ofsted, 2012, p. 7) which appeared to be indicative of both the reputation of the school and the behaviour and attainment of some pupils.

Ethnography, Case Studies and Ethno-Case Studies

Despite the fact that the meaning and focus of ethnographic research has evolved and expanded in recent years (Gobo, 2011), the primary aim of ethnographic research remains to gain a clearer understanding of the complexity of peoples’ beliefs, attitudes, and behaviours (Silverman, 2011) within their real life contexts (Cohen et al., 2007) as part of a particular group and/or culture’s values and norms (Geertz, 1983). In doing so, the researcher should look to immerse oneself as ‘naturally’ as possible in the group or institution by employing relevant methods such as observations, group interviews, and open conversations (Delamont, 2009; Gillham, 2000; Stewart & Shamdasani, 2014) as part of a conscious attempt to get closer to the reality of social phenomena (Blumer, 1954) and describe and interpret the shared and learned patterns of values, behaviours, beliefs and language of those involved (Harris, 1968). Despite this broad level of agreement with regards to the typical research design of traditional ethnographical research, however, there remains significant debate in relation to differences between ‘true’ ethnography and what is often termed ethno-case study (or quasi-ethnographic) research. By extension, there is also a need to acknowledge the fine line that often also exists in both theory and practice between what is deemed ethnography and what could be termed a case study – an issue that in fact often leads to the two terms being used inter-changeably (Parker-Jenkins, 2018; Patton, 1980). In relation to this, case studies and ethno-case studies have consistently been used in a manner akin to ethnography due to the sharing of the same data collection techniques to answer research questions and explore, describe, and explain empirical data emerging from real-life contexts (Flyvberg, 2006; Gerring, 2007; Yin, 2014). Indeed, a case study will typically often rely on multiple sources of data collated via a range of research methods (Yin, 2014) which are typically (i) participant observation, (ii) semi-structured and open interviews and (iii) documentary analysis (Denscombe, 2010) and are commonly used in conjunction with one another to triangulate the data (Gray, 2004). Based on this approach, case-studies enable the researcher to explore a range of themes and subjects in a focused way within their natural context (Gray, 2004) in order to provide an in-depth account of the relationships, experiences, and social processes that are occurring in that particular instance (Bryman, 2012; Stake, 1995). In addition, they can also prove to be ‘invaluable in adding understanding, extending experience, and increasing conviction about a subject’ (Gray, 2004, p. 123) due to the fact that the case does not become divorced from its context (Yin, 2014). However, the key difference between the accurate use of the term case study and true ethnography relates to the amount of time that is spent in the field and the subsequent immersion into the community of a culture-sharing group (Hammersley & Atkinson, 1995) – something that many argue a case study often does not provide. In this regard, some academics suggest that the unearthing of genuine relationships, complexities, nuances and outcomes is far more challenging in many case studies due to the fact that the researcher is often not embedded in the research over a sufficiently prolonged period, meaning that they are unable to get a true awareness and understanding of longer-term patterns that change over time (Hammersley, 2006). However, other authors state that the significance of time spent in the field has evolved in recent years to be less of a constraining issue, with strong suggestions that ‘new ethnographies’ with different understandings and less rigid and restricted durations (Pole & Morrison, 2003) can be utilised in order to generate more fluid and varied time frames through which to conduct ‘new’ ethnographies (Bagley, 2009; LeCompte, 2002; Mills & Morton, 2013). Importantly, this means that contemporary ethnographies may now last months rather than years without any perceived decreeable risk to the quality, accuracy and depth of data (Ingold, 2014). In this regard, Parker-Jenkins (2018) utilises the term ‘ethno-case study’ as part of an attempt to convey the sense of an inquiry which concerns people and employs techniques associated with long-term and intensive ethnography, but also acknowledges and accepts the limitations of such an approach in terms of scope and time spent in the field. Whilst there is clearly a need to acknowledge the fact that contemporary ethnography or ethno-case studies still require on-going personal, engagement over a sustained and significant time frame (Emmel et al., 2007; Miller & Bell, 2002) in order to generate credibility, respectability and trust-worthiness (Lincoln & Guba, 1985), for Parker-Jenkins (2018, p. 24) ‘both ethnography and case study have value in the process of social science research and as the use of technology speeds up data generation we can avoid having tight boundaries around the two concepts’.

In relation to all this, therefore, the term instrumental ethno-case study is used as a way of acknowledging the long-term but intermittent nature of data collection alongside the consistent attempts to accurately explore, describe and explain the values, behaviours and culture of the school. More specifically, whilst a more continual period of time spent in school over the 16 months was not possible for me due to a range of personal and professional commitments, the weeks and days spent in school were purposively selected (where possible) in order to create and develop suitable relationships and to observe what were perceived to be key lessons and events, whilst maintaining a degree of spontaneity that would enhance the chance of experiencing both ‘true’ and serendipitous events. Overall, therefore, despite the personal and professional constraints imposed on me, all efforts were still made to ensure that the study remained led by ethnographical principles and that I was actively involved in many aspects of school life (Emmel et al., 2007) for a ‘fairly lengthy’ period of time (Hammersley, 2006, p. 4) so that I was able to become and remain immersed in the case as much as possible in order to record and interpret what people were doing and saying (Hammersley, 2006) as accurately as possible.

The Study Design for ACS – an Initial Overview

In the early stages of the research process, I made the conscious decision to attend the University Research Ethics Committee (UREC) meeting in person due to the fact that researchers have often experienced significant institutional constraints with studies that have involved covert participation observation (Roulet et al., 2017) as this study did. However, the design and aims of the study were all adequately justified and the principles of autonomy, non-malicience, beneficence, and justice were all considered (Greig et al., 2013) meaning that full ethical approval was granted by the UREC. In relation to the structure and implementation of the data collection period, an initial ‘familiarisation phase’ took place at the start of the calendar year where I aimed to familiarise myself with the school and become a more ‘accepted’ presence around the PE department with both students and staff. Following this, a covert observation approach was adopted with male pupils across 45 PE lessons during February, March and April that also involved approximately 60 guided conversations with both pupils (40) and PE staff (20) with all PE staff aware of my presence and broad focus of research at this time. During these initial covert observations, certain ‘types’ of male pupils began to emerge which meant that the majority of male pupils in each year group at ACS were allocated to one of three named groups based on their attitude, behaviour, engagement at school as well as their relationships with staff and peers. These groups (not formally recognized in any way by the school) were subsequently named the Performers, Participants, and Problematics and emerged as significant part of the study design, approach and findings going forward. As a direct result of the emergence of these groups and following the process of ‘breaking cover’ with the pupils at this time (May), a total of eight focus group interviews were then conducted over a period of several weeks. These focus group interviews were aimed at clarifying and/or validating some of the evidence that had emerged from earlier data collection via overt and covert observation (McNaughton-Nicholls et al., 2014) as well as exploring some of the issues that had emerged from the literature review process. A focus group with the male PE staff (n-4) was also conducted around the same time for similar reasons. All focus groups (staff and students) were conducted alongside continued overt lesson observations in mainly practical PE lessons in order to continue with data collection and further examine issues that had emerged from focus group interviews. This led to a total of 65 PE lesson observations (50 Practical and 15 classroom) being conducted during the duration of the entire study. It is to a more detailed and chronological overview of the process of data collection and lessons learnt throughout the process that this study will now turn.

Lesson 1. ‘The Gatekeeper’

Once the case school was confirmed as being the focus of the study, the challenge of gaining initial access to senior members of staff initially and then the school more broadly was addressed (Bryman, 2012). This was instigated by identifying a ‘gatekeeper’ who was in a position to facilitate initial access to a senior member of staff in order to explain the nature of the research and gain formal permission to access the school in order to conduct research with staff and pupils (Corra & Willer, 2002). This initial ‘gatekeeper’ (a member of pastoral staff at ACS who emerged as being a friend of a former colleague) was able to organise a telephone conversation with the head teacher during which I was able to provide an overview of the nature of my research and some information pertaining to my own personal and professional background. Following this, a more formal, face to face meeting took place where the more significant ethical considerations relating to the covert nature of initial observations of the pupils, my temporary role as LSA in the school, and the need for access to changing facilities were discussed and then justified in relation to my desire to minimise changes in pupil behaviour during the early stages of data collection. Following this meeting, where a copy of the approved ethics application from the University Ethics Board was presented, the head teacher expressed her full support for the aims and design of the study dependent on a full LA-specific DBS check being completed prior to the commencement of data collection. The following day, a meeting with the head of PE took place during which the nature, duration, and focus of the study (‘an examination of working-class PE’) were explained.

Lesson 2. Building Rapport

The decision to immerse myself in an initial familiarisation week provided me with a highly effective opportunity to become more familiar the with day-to-day structure of the school day and work towards developing a plausible long-term ‘identity’ within the PE department as a part-time learning support assistant (LSA). This identity was initiated and developed as part of an attempt to become a visible, plausible and accepted presence (Pitts & Miller-Day, 2007) both around school generally as well as in the PE department specifically and achieved by arriving at school before students, ensuring that I was seen in the PE staff office at appropriate times, eating lunch in the school canteen with other PE staff, and attending extra-curricular activities. More specifically, every attempt was made to formulate and develop my identity as a part-time LSA from around the local area so that participant observations could be conducted in a relatively covert manner and ‘natural’ behaviour could be promoted as quickly as possible. This assumed identity was more formally supported by an official ACS ‘support staff’ identity card/lanyard and the wearing of school PE staff kit, as well conscious efforts being made to replicate and accentuate the ‘local’ accent when in conversation with pupils, emphasise any knowledge of the local area when needed, and taking every opportunity to demonstrate an interest in local and national sport (e.g. League One football).

In relation to the PE staff and other relevant staff around the school, despite there being less of a need for me to perpetuate a false identity, I made consistent attempts to engage in appropriate professional conversations with staff, ensured that I conducted myself professionally at all times, and took every opportunity to demonstrate my knowledge and previous experience of having worked in education and PE. In order to facilitate this, I was also able to provide all PE staff in the department with an overview of the research design and focus (stated as being an ‘examination of PE in working-class schools’) so that they felt clear and comfortable with the reasons behind my presence in the department. Following this, I became focused on creating and developing open and trusting working relationships with the PE staff by taking a passive role in a range of pre-confirmed PE lessons that were complimented by a range of informal conversations with relevant pastoral staff (e.g. head of Y11) if/when appropriate. During this stage, I found it very important not to use any form of communication that might have suggested a form of professional judgement on my part, even when the staff might have questioned their own actions and professionalism (Teacher - “I know that they’re playing football again, but they just will not play anything else” – Me - “I get it. It definitely means that they’ll take part and who needs the battle when they leave in 10 weeks”).

Lesson 3. Preliminary Observations

Participant observation is a research method that generally occurs over a relatively sustained period and involves watching, listening to, and asking questions of, people as they follow aspects of their day-to-day activities (Payne & Payne, 2004) with the aim being to provide a uniquely humanistic, interpretative, and ‘natural’ approach to data collection (Yin, 2014). Given the immersive nature of this approach, participant observation can provide a more accurate portrayal and ‘picture’ of the group/case that is unlikely to be gained simply by speaking to people (Simons, 2009) and if done effectively can ascertain people’s ‘true’ actions and attitudes (Gobo, 2011) in their respective ‘natural’ environments. In relation to covert participant observation more specifically, this occurs when researchers become embedded in the group while the researchers conceal ‘their true identity and purport to play some other role’ (Vinten, 1994, p. 33) in order to provide access to otherwise unavailable data (Lauder, 2003). In doing so, however, it also allows the researcher to capture rich data that would often otherwise be unobtainable (Roulet et al., 2017) as well as reducing the risk of those under study modifying their behaviours in the presence of a known researcher/outsider (Denscombe, 2010; Oliver & Eales, 2008). As a result of these benefits, covert studies – that inevitably entail some element of deception - have played a prominent role in the development of the social sciences (Roulet et al., 2017) and have been employed in prominent sociological studies across a variety of groups or settings, including factories (Bernstein, 2012), asylums (E. Goffman, 1961), gangs of adolescent boys (Parker, 1992), football hooligans (Pearson, 2009), and young men on the run from the police (A. Goffman, 2015). However, not only is the justification of covert methods context-dependent (Oliver & Eales, 2008), it brings with it many potential criticisms and potential risks to participants and researchers (Roulet et al., 2017) that include the ethical issue of participants being ‘manipulated’ and ‘conned’ (Erikson, 1995, p. 9) and in doing so, claims can be made that the use of covert observation transgresses the core principle of any research (Bryman, 2012) in that participants should always be able to make an informed decision on whether to participate (Denscombe, 2010). Indeed, the Social Research Association (SRA) guidelines (2021, p.121) specifically state that ‘inquiries involving human subjects should be based as far as practicable on freely given informed consent of subjects’, the British Sociological Association (BSA) guidelines (British Sociological Association, 2017, p.3) state that ‘as far as possible, participation in sociological research should be based on the freely given informed consent of those studied’, and the Economic and Social Research Council (2022, p.31) are clear on the fact that covert observation constitutes deception which ‘should only be used as a last resort when no other approach is possible’. In the relation to this final point, however, the American Psychological Association (APA) (American Psychological Association, 2010, p. 807) recognizes that ‘the use of deceptive techniques [can be] justified by the study’s significant prospective scientific, educational, or applied value’ when the use of ‘non-deceptive alternative procedures are not feasible’ and in the UK, Bryman (2012) states that such a data collection method may be justified in certain circumstances, as the BSA ‘leaves the door ajar’ for the use of covert observation by the use of the term ‘as far as possible’ in relation to attaining participant consent.

Ultimately, therefore, because the risk to participants was perceived to be relatively minor (SRA, 2021) and the use of covert observations is relatively widespread in studies of this kind (Bryman, 2012), covert observation was deemed to be ethically justified and practically effective in observing and recording the genuine behaviours and responses of this particular group (Simons, 2009). By utilising and reinforcing the range of approaches used in the very early stages of the study in order to create and develop a new identity (Roulet et al., 2017), a range of inevitable questions came from some pupils that were not only linked to my professional role in the school (“Do you work here now then sir?”) but also more personal questions such as “what football team do you support?” and “I suppose you want to be a PE teacher like these do you then sir?”. Despite the seemingly suspicious nature of these questions, a strong feeling of acceptance and trust was evident from an early stage in data collection - an outcome that appeared to be facilitated by the use of a strong local accent, apparent awareness of the local area when asked, and a willingness and ability to be involved in all aspects of the practical lessons being ‘observed’. In relation to the latter, involvement in the practical aspects of the lesson was not only an inevitable (and at times unavoidable) outcome of lesson observations (Yin, 2014), but perhaps more importantly served to help me gain greater levels of credibility and acceptance amongst the male pupils, and to some extent the PE staff. In this regard, I was ‘actively encouraged’ by pupils on several occasions to ‘actually participate in the events being studied’ by adopting ‘a variety of roles’ in the active research process (Yin, 2014, p. 111) by either taking on roles that no other pupil wanted to do (e.g. goalkeeper in football games) or demonstrating a degree of sporting competence (e.g. partner in badminton lesson) which also afforded me the opportunity to remain flexible and reactive to relevant incidents as they occurred (Simons, 2009).

Despite the fact that an element of deception underpinned the covert observation of pupils at this stage, every effort was made to view and record the pupils’ actions equitably with attempts consistently made to avoid personal bias and preconceived ideas (Evans, 2012). More specifically, not only was consideration consistently given to the potential and actual impact that my presence was likely to have on the actions of the pupils (McNaughton et al., 2014), but there was an on-going awareness of my own personal response to all observations and verbal responses from pupils and staff (McNaughton et al., 2014) in order to minimize the impact of any personal assumptions or perceptions on the collation and subsequent analysis of the data (Evans, 2012; Gobo, 2011).

As the relationship with the PE staff at ACS evolved and levels of trust increased, it was evident that the actions and responses of PE staff in lessons became increasingly ‘natural’. This was enhanced as a result of engaging in aspects of sympathetic and empathetic responses to certain situations that linked closely to my knowledge and awareness of the challenges that working in such an environment can bring. Therefore, whilst conscious to minimise the impact of my presence on the actions and/or responses of PE staff with regards to their lessons and relationships with colleagues and/or pupils, it was common for me to use statements such as “I know exactly what it is like” or “I would have done very much the same then” as part of a genuine attempt to empathise with the PE staff in relation to the challenges that they were facing and it ensure (where possible) that the PE staff did not feel that they were being professionally or personally judged. Overall, therefore, conscious and consistent efforts were made to formulate relationships where the PE staff felt that they were not directly involved in research (Payne & Payne, 2004) – an outcome that was aided by making a conscious effort to remain professional, flexible, and act as a full-time member of PE staff wherever possible. In relation to this, during this more focused period of lesson observations, not only were there no obvious signs of any suspicion amongst PE staff that could feasibly be expected in this context (Sharp, 2000), it was also apparent that PE staff at ACS were willing to act as key informants (Bryman, 2012) by providing ‘tip-offs’ and additional information (Hammersley & Atkinson, 1995) on specific pupils, groups, and lessons that they deemed to be relevant to my research at that time.

Lesson 4. Engaging in Guided Conversations

A prominent and effective aspect of this observation process was the use of guided conversations during the school day that generally occurred before, during, or immediately after PE lessons. As is common with studies of this kind, guided conversations often occurred informally and spontaneously, and were generally utilised in order to provide insights into staff or pupils’ attitudes/behaviours, or as a way of clarifying issues and incidents that had been observed (Yin, 2014). In this regard, therefore, for both PE staff and pupils, questions were much more fluid rather than rigid (Rubin & Rubin, 1995) and in many cases related to unforeseen situations or incidents that have emerged from the process of observation. With the PE staff specifically, these guided conversations typically took place in a range of different environments and situations in order to gain a better knowledge and understanding of a wide range of issues. This may have been a conversation at break time (“Am I right in thinking that you’ve got that same group today as I saw on Monday”), a chat on the way to a lesson to determine its focus or address any issues to be aware of (“What are you planning to teach today” or “Is there anybody in this lesson it might be worth me keeping my eye out for”), or brief conversation in the changing rooms following a specific issue or incident (“Has that ever happened before?” or “Is he generally like that in lessons?“). The specific aim of these types of questions was to attempt to gain the member of staff’s interpretation and feelings towards an incident that we had both observed and in some cases been directly involved in (Patton, 1980). Therefore, not only were ‘how’ and ‘why’ questions often utilised with staff, but two-way discussions were often common (Simons, 2009; Yin, 2014) as part of a conscious attempt to understand the immediate and long-term context of a particular incident or on-going issue (“Why do you think that most of the them (male pupils) seemed to completely lose interest in the game towards the end?” or “You say they only seem interested in football, why do you think that is?”). In addition, more specific attempts were also made to gather the individual attitudes, views, and interpretations of staff which became easier as the study progressed and enhanced to some degree by the relative freedom of an unrecorded, one-to-one conversation (Yin, 2014).

With the male pupils, guided conversations were more likely to occur during or after lessons in order to clarify something that had been either observed or overheard (“You seem to enjoy Badminton, do you every play outside of school” or “Did you have a question about the rules, I’m happy to help you”) in order to gain a greater understanding of issues/incidents or to develop a clearer interpretation of staff and pupils’ views and values (Yeo et al., 2014).This was often achieved by appearing nonthreatening and bordering on ‘naïve’ regarding the issue or incident whilst using ‘how’ as opposed to ‘why’ questions in a conversational style (Simons, 2009; Yin, 2014). In many cases, there was a relative reluctance or ambivalence to the posing of questions amongst many of the male pupils so questions posed to pupils were often very short in nature in order to ascertain their willingness to respond (Yeo et al., 2014) (“Di you enjoy that?”) followed by lengthier and more detailed questions where appropriate in order to assess a pupil’s general attitude towards a sport (“Do you like tennis”?), determine their level of knowledge and/or understanding in a lesson (“You won the last point then so what’s the score now?”) or to gain a better understanding of their previous or current involvement in a sport or activity (“Do you ever play tennis outside of school?”). In many cases, the questions were also delivered with a false level of ignorance or naivety in order to try to relax and empower the pupils towards a confident and realistic response by ‘hiding’ a question within a brief conversation (“What’s travelling/double dribbling again? I’ve never played much basketball”). Both during, and following on from these guided conversations, the consideration that responses were often spontaneous and could be misinterpreted and/or forgotten (Bryman, 2012) was consistently acknowledged. In terms of recording the responses and content of these guided conversations, a conscious effort was made to ensure that the nature, context, and content of any conversation was recorded as soon after the event as was practically possible often by “slipping away” into an available room (Festinger et al., 2008) or completing more detailed notes in the relative privacy of the PE staff room. All staff were made fully aware that any responses that emerged from such conversations could be included in the final thesis and all subsequent quotes that were included in the final thesis were shown to relevant staff before inclusion, and all responses from pupils were anonymised ahead of publication.

Lesson 5. Defining and Naming the Pupil Groups

During the initial stages of covert observation, it was apparent that the school academic banding system in place at ACS was imposing and perpetuating a range of different social and academic expectations on the male pupils. In this regard, the peer groups with whom they were taught and the different relationships that they had with staff and peers, led the vast majority of male pupils at ACS to demonstrate a range of different attitudes and behaviours in most aspects of their school lives. In relation to the observations that took place in PE lesson more specifically, the differing actions and attitudes that existed between the male pupils in each year group related to how certain male pupils interacted with PE staff and peers, how they conducted themselves whilst changing, and the extent to which they were willing to actively engage and participate in PE lessons. As a result of these observations, it was evident that three relatively distinct groups of male pupils could be distinguished within mixed-ability, core PE lessons where all academic bands arrived together. These were a group of (i) top-band, high attaining students who engaged positively with PE staff and participated regularly in all lessons, (ii) a second group of middle-band academic pupils who were also content to engage and participate in PE but were relatively reserved and withdrawn with PE staff and many peers, and (iii) a group of low-band academic students who consistently displayed challenging (sometimes confrontational) behaviour with both peers and PE staff meaning that they often refused to engage or participate in some PE lessons. As a result, I created the terms (i) Performers, (ii) Participants and (iii) Problematics as a way of identifying and assigning male pupils in each lesson (although at no stage were the pupils themselves made aware of these). Mindful that not all students could be assigned to a group, Weber’s (1904) concept of an ‘ideal type’ was utilised as an analytical construct in order to develop a range of broad criteria that could be used in order create a ‘measuring rod’ (Coser, 1977, p. 223) for assigning male pupils to one of the three aforementioned groups (Performers, Participants and Problematics). These criteria were essentially built upon the assumption that component actors will/would behave in particular ways in certain social situations and whilst a ‘full empirical embodiment’ of either type of male pupil (Coser, 1977, pp. 223–224) was not possible, very obvious and relatively specific traits did emerge across various areas of school (language, attitude to learning) that not only allowed pupils to be allocated to one of the three aforementioned groups with relative confidence, but for the data collection to become more strategic and focused towards certain groups as a result. In relation to this, a desire on my part to conduct focus group interviews emerged with pupils across all the three groups as part of a conscious and strategic attempt to explore and triangulate any issues that had been observed and recorded from data collected so far. In order to select pupils for participation in these focus groups, there was a period of discussion with the PE staff regarding the concept and process behind the creation of the three groups generally, and the accuracy of pupils that I had initially assigned to each group more specifically. Follow this two-way discussion with PE staff, final participation lists for the focus groups were generated with the aim being that the most prominent and ‘typical’ male pupils had been assigned to each group so that more specific issues and incidents could be explored.

Lesson 6. Conducting Focus Groups with Adolescent Male Pupils

In order to begin the organisation and completion of focus group interviews, all male pupils who had been directly or indirectly involved in the study were informed of my true identity as researcher via a debrief session at the start of all relevant PE lessons (Lauder, 2003). The process of ‘coming out’ as researcher once trust has been gained with participants is rare in ethnographically-based studies due to the fact that most studies are generally focused on either covert or overt observations (Denscombe, 2010). Therefore, care was taken to explain the specific nature and aim of my role in school, reassure pupils about the use of any data previously collected, confirm that I would still remain as a member of PE staff (LSA), and would answer any questions that they may have. I was also keen to confirm that I would be happy to discuss the focus and nature of my research with any parents should they require it although this was not required. The decision to use focus group interviews at this stage was primarily based on the fact that such an approach allowed for a relatively large number of participants to be interviewed in order to cross reference some of the behaviours and responses that had emerged during observations and guided conversations and also assess the extent to which pupils responded ‘to each other’s views’ and interacted with one another during the asking and answering of questions (Bryman, 2012, p. 502). There was also the opportunity to probe responses from participants in order to gain a deeper understanding of how pupils’ responses may have been socially constructed (Wilkinson, 1998) as well as the opportunity to ask specific questions of certain groups (“I’ve noticed that you really enjoy playing football in PE. Why is that?“). Given the wide range of potentially emotive topics covered, care was taken to ensure that at least 90 minutes was allocated in order to complete each focus group and perhaps most importantly, these interviews were scheduled to take place during coursework catch up sessions as part of the pupils BTEC Sport Studies or GCSE PE revision lesson (“They’ll vote with their feet if you make them miss PE. Most will not turn up and if they do and they will not be happy”).

In terms of the practicalities of conducting these focus groups, pupils were informed in advance as to how long the interview would take, what would happen to their responses, and how and why each pupil should respect the response of others at all times. During the interview itself, a conscious effort was made to remain nonintrusive and facilitate the appropriate involvement of all participants in order to ensure that all discussions remain relevant (Bryman, 2012). As these interviews often took place in morning sessions (before lunch time) and over a prolonged period, soft drinks and biscuits were made available to the male pupils in each of the focus group venues primarily to avoid boredom and hunger whilst being mindful of any bias that this might cause (“Grab a quick drink to start lads and if you feel a bit peckish at any time just grab a biscuit by all means”). In relation to the specific responses from the pupils that had the potential to lead to culturally expected views (such as girls, football, fighting and behaviour at school) (Morgan, 2002), care was taken to pose questions in a manner that may minimise this issues (“Try to think what you actually believe here. If you disagree that’s fine and please consider what other people say”). In addition, there were also several cases where students had to be warned about their reactions to other pupils’ comments (“That’s fair enough Will, but you’ve got to accept that’s what he thinks”) as well as the need to also remind pupils about the nature of what they were saying and/or the language used in their responses (“Ok, but we need to avoid those types of words if we can”). Much more common was the requirement for greater depth or clarity in responses (“Can you just give an example of that from PE”) as well as the need to limit the over-bearing nature of some pupils whilst ‘actively encouraging’ responses from other group members who appeared more reluctant to contribution (“hat do you think about that Seth?”). Although it is difficult to ensure that their responses were not influenced or constrained by the presence of others in the room (Bryman, 2012) or that some individuals might have suppressed their responses if they appeared to be counter to the views of the rest of the group or potentially controversial (Finch et al., 2014), overall responses from pupils in all focus groups appeared to be predominantly honest, candid and mature.

Lesson 7. Approaching the Research as a Relative Insider

The term insider is used in this paper in order to reflect the fact that the study was designed and implemented from an informed, emic, perspective that was based on my previous ‘inside’ experience (Headland, Pike & Harris, 1990) as a secondary school PE teacher in a large working-class school nearby, as well as having a relatively close personal relationship with the village of Ayrefield. As a result, it was evident that my previous knowledge and experience of teaching in a working-class secondary school, my knowledge and awareness of the local area surrounding ACS, and my own upbringing within a similar working-class community not only allowed me to minimise the likelihood of feeling like a ‘stranger entering an unknown culture’ (Bell, 1999, p. 22) when arriving and spending time at ACS, but also prevented me having to acquire ‘the language and behaviours that go along with the need to acclimatize to a totally new environment (Tedlock, 2000, p. 455). However, notwithstanding, these beneficial and positive outcomes of being an ‘insider’ during the research process, the term ‘relative insider’ is used in this paper in order to reflect the fact that whilst there was no sense of me feeling like a true outsider, there was also no need to manage the ‘additional layers of complexity’ that researching amongst former colleagues would have brought as a genuine insider, as was the case for Unluer’s (2012) return to a previous site of employment. Therefore, the term ‘relative insider’ has been used to reflect the fact that I was able to see myself as part of a continuum between insider and outsider (Mercer, 2006) rather than a dichotomy between the two. This position and term enabled me to acknowledge the fact that I felt part of an environment to which I been previously accustomed at ACS – thus saving me time in having to absorb the culture and structure of the secondary school environment (Gillham, 2000) - but also avoided the need for me to manage the potential challenging issues of being a true insider by conducting research with former or current colleagues.

By extension, a conscious and consistent consideration in relation to this level of immersion and acceptance in the school and the familiarity with the surrounding area was the need to acknowledge and avoid the very real risk of ‘going native’ (Van Heugten, 2004). Not only can this lead to the researcher ‘losing his or her distance and objectivity’ by effectively becoming a ‘member of the group and forego[ing] his or her academic role (Fontana & Frey, 2000, p. 655), but also that the demands of the research project itself often fail to be met due to the ‘very close and emphatic identification with the subjects of the research’ (Pollard, 1985, p. 219). As a result, there was a need for me to separate my own professional and personal experiences and views from the data collection and analysis process in order to acknowledge the fact that whilst ‘inside information’ brought several key advantages to the research process, this stood to potentially constrain my level of detachment that would usually be expected of most research studies (Kanuha, 2000). Indeed, a constant balance had to be sought between creating sufficient distance to maintain a position as an objective researcher whilst becoming overly distanced from the study to the extent that there was less opportunity to attain ‘thick’ descriptions of what were often complex phenomena (Geertz, 1973). More practically, and as a specific example, when working with the PE staff specifically, the tendency for assumptions to be made regarding my intimate knowledge of the situation or culture by other teachers (‘you obviously know what I mean’) was always counteracted by a request to clarify or elaborate ‘just for the record’.

Lesson 8. Considering Involvement and Detachment – Theory into Practice

In order to acknowledge and address the potentially problematic position as ‘insider’ researcher, Norbert Elias’ concept of involvement and detachment emerged as a consistent consideration throughout the research process. Just as Bryman (2012, p. 22) argues that ‘there is a growing recognition that it is not feasible to keep the values that a researcher holds totally in check', figurational sociologists state that researchers cannot be completely detached in their work. Indeed, whilst this does not mean that it is desirable, or possible, for them to be completely involved, from a figurational standpoint, the research process should involve a combination of both involvement and detachment. When examined in more detail, notions such as “ultimate truth" and “complete detachment" have no place in Elias’ approach to research (Murphy et al., 2000). Instead, a figurational approach to research involves a clear recognition that sociologists should strive for an appropriate blend between involvement and detachment whilst simultaneously acknowledging that it is impossible for any sociologist to achieve complete objectivity or ‘detachment' in their research. After all, for Elias, it is evident that unlike the chemist studying chemical reactions, the sociologist is an inescapable part of the phenomena that are human relationships (van Krieken, 1998) due to the fact that ‘social-scientific knowledge can only develop and emerge within the society it is part of, and not independently of it' (van Krieken, 1998, p. 176). When considering the direct involvement and role of the researcher as an ‘insider’, therefore, it is ‘crucial to recognise the centrality of the researcher in the process of data generation and analysis' (Perry et al., 2004, p. 139) and that such as level of involvement and detachment is inevitable, especially where the aim is to develop ‘a more reality-congruent picture of complex aspects of the social world’ (Perry et al., 2014, p. 139). Ultimately, this is precisely why figurational sociologists prefer the concept involvement and detachment because it more accurately reflects the reality of the personal situations of social researchers compared to traditional conceptualizations of objectivity and subjectivity (Rojeck, 1986). In relation to all this, therefore, my aim throughout all stages of data collection was to recognise the range and complexity of my involvement with ACS as well as the village of Ayrefield more broadly. More specifically, consistent and specific care was taken to acknowledge the impact of my time working as a PE teacher in a mainstream, working-class school and my own working-class upbringing. By acknowledging these experiences and relationships as well as their potential to influence my recording and interpretation of the data, my consistent aim throughout the study ‘was to strive to distance (my)self as far as is possible from one’s values’ (Bloyce, 2004, p. 149) in order to facilitate a better, more reality-congruent understanding of the issues related to the area of research (Bloyce, 2004). This was done by acknowledging the fact that ‘the sociologist as participant must be able to stand back and become the sociologist as observer and interpreter' (Maquire, 1988, p. 190) in order maximise ‘the chances of obtaining secure knowledge'. In order to achieve this, I consistently considered the recording and interpretation of the data emerging from at all stages of the study and also engaged in considered and open discussions with a range of external people including academic staff/supervisors, former school colleagues and well-informed friends in an attempt to highlight and explore any potential biased viewpoints and/or issues as they began to emerge.