Space and place in researching male early high school leaving in Orange Farm Township

I reflect on the methodological processes underpinning a dissertation that investigated male learners’ reasons for leaving high school early and the strategies they employed to negotiate everyday life. A qualitative case study was conducted with nine male early high school leavers between the ages of 18 and 25, as well as 12 stakeholders involved in the Orange Farm Township, south of Johannesburg. Purposive and snowball sampling techniques were used. Data were collected through in-depth interviews, document reviews, and observations. Narrative analysis revealed complications related to the notions of space and place of the potential participants and the researcher. In the first place, the difficulty was not in identifying participants, but in establishing rapport to the extent that they agreed to participate in the research. Assumptions about space and place gave rise to expectations that had to be managed, and consequently the researcher had to rethink the methodological choices. In particular, participants’ perceived real social positions and their relation to different social spaces had to be negotiated. It is suggested that relatively novice researchers, researching male early high school leaving in familiar spaces, can mitigate complications of space, place and stigma during fieldwork by using multiple sources of data and strategic, flexible interviewing techniques.


Introduction
Relatively novice researchers using qualitative methodologies and observing conventional research prescriptions may encounter situations that necessitate practical reflection beyond the available conventions. Such situations may be conceptualised in what Lee-Treweek and Linkogle (2000:24) term "dangers" encountered in the field or dangers that emanate from fieldwork relations. This paper is based on a research dissertation written as part of the requirements of a Master of Social Science in Sociology degree (Bingma, 2012) that investigated male early high school leavers in Orange 1 Farm Township, located south of the Johannesburg city centre. The aim of the paper is to outline the methodological processes that informed the dissertation and, in hindsight, to reflect on the processes and their implications on the research process. This reflection is important in providing points of methodological engagement for novice researchers working in the area of early school leaving and male early high school leaving in particular.

Background
Early school leaving is a key problem confronting school systems and it calls into question the school systems' organisational components (Coleman, 1988), curriculum and pedagogical approaches (Freire, 1970;Illich, 1970), and the very context within which schooling occurs. The context of schooling may relate to issues of school governance (Xaba, 2011). To understand early school leaving requires in-depth consideration of each case to unpack the decision-making processes involved for each individual and to map out the life chances of the many individuals concerned. It is necessary to employ qualitative methodology to identify the nuances and interplay of the social factors involved in the process of deciding to leave school prior to completing Grade 12. Qualitative methodology is also necessary to obtain an in-depth understanding of the problem. However, qualitative research targeted at early school leavers that have already left the school system is difficult when one cannot rely on official admission registers (Mahlomaholo, 2011). The difficultly lies not only in recruiting participants (Flisher & Chalton, 1995), but also in establishing a rapport with them to the extent that they will agree to tell their stories. The difficulty is compounded by the social stigma attached to early school leaving (Bingma, 2012). With its many contradictions (Harber & Mncube, 2011), schooling remains an important element in broader social discourses about developing a knowledge economy. SAPA (2008) and the Department of Basic Education (2011) provide perspectives on the official positions on the topic of early school leaving. Scholarly work dealing with early school leaving in South Africa includes research by Flisher and Chalton (1995); Inglis (2009) ;Motala, Dieltiens and Sayed (2009) ;Porteus, Clacherty, Mdiya, Pelo, Matsai, Qwabe and Donald (2000); Sibanda (2004);and Wegner, Flisher, Chikobvu, Lombard and King (2008).
The research dissertation under discussion was conducted with reference to the Orange Farm Township (Bingma, 2012). This township was established in the late 1980s as a direct result of the social disorganisation and growing conflict that characterised the dying years of apartheid (Crankshaw, 1993). Orange Farm was declared a township in 1997. Like many townships, it is characterised by social reconstruction and/or destruction. In fact, townships cannot be understood outside the broader national dynamics of a society playing out in local contexts (Mills, 1959).
The Orange Farm Township was chosen as a field site because observations had previously been conducted in the area for a small-scale Honours research report (Bingma, 2008). This area was also chosen because the researcher was familiar with the area and lived in this area permanently for eight years from 1997 to 2004. Although the researcher has not lived in the Orange Farm community permanently since 2005, the township remains a home to her as her family home is located in the community and she visits the area frequently.
The above circumstances made the researcher both an insider and outsider in field relations. The term insider denotes the familiarity of the researcher with the physical space, with some of the participants, as well as with the broader social experiences of the community members under study. Insider did not include the everyday lived experiences of the participants as members of a category of people that the researcher was interested in understanding. The latter point involves being an outsider -the researcher is not a male early high school leaver and does not share this group's experiences of negotiating everyday life. The researcher can also be considered an outsider because of gender and education level in relation to the target population (Bingma, 2012).
The methodology used in the dissertation included a qualitative case study designed to answer the research questions. Data were collected through in-depth interviewing techniques and observations. The data were analysed through narrative analysis. The above decisions were informed by the following research questions: • Why did male learners leave high school prior to completing Grade 12? • How did male early high school leavers negotiate their everyday lives? The objective of the research was to explain male early high school leaving through personal accounts by the participants and this objective directed the methodological choices. A decision was made to conduct 15 to 20 interviews with male early high school leavers in different sections of the Orange Farm Township. However, only nine primary participants between the ages of 18 and 25, as well as 12 stakeholders including teachers, a ward councillor, youth groups (for example, LoveLife), health service representatives, and community members were interviewed. These participants were selected through the use of purposive and snowball sampling methods. The final report used pseudonyms for the primary participants and stakeholders to ensure anonymity. Information that could be used to identify the participants was not included in the report. Extensive observations formed part of the data collection process to supplement existing literature on Orange Farm and photographs were an integral part of recording the activities of the males throughout the township (Bingma, 2012).
Over and above reflecting on the methodological processes that underpin the dissertation (Bingma, 2012), this paper is a modest contribution to the literature on early school leaving and it highlights the potential complications of space and place when researching male early high school leaving in a familiar context, as well as the methodological decisions that facilitated and/or hindered the process.

Conceptual framework
Conceptually, the dissertation is embedded within the sociology of education subfield. This subfield considers education processes and experiences in relation to numerous other local and national social processes (Ballantine & Spade, 2007). This paper draws mainly on the work by Goffman (1963) in dealing with stigma and the ideas explored by David Harvey (2004), Herbert Gans (2002) and John Urry (2001) that address issues of space and place. The dissertation used the notion of insider and outsider mentioned above to address the dilemmas encountered during fieldwork. Stigma was used as a broader conceptual tool through which male early high school leaving could be understood alongside social capital (Bingma, 2012).
In hindsight, the notions of space and place seem more lucid in framing and aggregating the processes that unfolded. Also, the dissertation under discussion used stigma to read, albeit superficially, the relations between the participants and the researcher (Bingma, 2012). Stigma is central to this paper and is important in illuminating the analysis of fieldwork relations in general. Goffman (1963:3) asserts that stigma is a discrediting quality. He differentiates between "discreditable" characteristics, which are not readily noticed, and "discredited" characteristics, which are objectively visible. One can argue that male early high school leavers are a discreditable group whose characteristics are not readily noticed because their 'differentness' is not immediately evident. Stigma must be established in society for it to exist (Goffman, 1963) and the role of power is central to its social reproduction (Link & Phelan, 2001). When the dividends of completed schooling in the context of socioeconomic opportunities are considered, it may be said that early school leavers occupy a lower position on the "status hierarchy" (Link & Phelan, 2001:379) because of the value placed on a high-school certificate. Early school leavers are labelled "outcasts and losers" (Panday & Arends, 2008:1), as well as dumb and/or lazy.

Stigma
A person who is stigmatised then has a "sense of not knowing what the others present are really thinking about him" (Goffman, 1963:14). The problem becomes one of dealing with the facts of one's shortcomings and whether "to tell or not to tell" (Goffman, 1963:42). In a context where social constructions of masculinity are linked with the ability to provide materially (Ratele, 2001;Connell, 2005), a male early high school leaver's social identity is discredited in that he may struggle to meet the societal expectations of being a provider because of high unemployment levels amongst the youth in South Africa (Gordhan, 2011).
The notion of managing a discreditable attribute of one's social identity is useful in making sense of narrative data. In examining subjectivity as it emerges in narratives, Bloom (1996:193) argues that respondents evaluate their own narratives through a process that provides opportunities wherein desires for "positive selfrepresentation" are replaced by attempts to understand the self. Bloom (1996:193) refers to that process as "nonunitary subjectivity in narrative representation". Nonunitary presentations of the self underline how individuals make sense of and negotiate contradictions and ambiguities in their life course. According to Newman (2002), the negotiation of what can or cannot be spoken, in part, is influenced by the meanings individuals attach to social interactions with others.

Space and place
Space and place are interesting concepts in as far as they transcend disciplines as analytical tools. Although Harvey (2004), Urry (2001) and Gans (2002) have different points of departure regarding space and place, this paper connects ideas emerging from their works to aid methodological thinking regarding conducting research on male early high school leavers. As a point of departure, the literature consulted states that space can be conceived of in multiple ways (Harvey, 2004;Urry, 2001;Gans, 2002). In Space as a Keyword, Harvey (2004:2-4) identifies the concepts "absolute space", "relative space" and "relational space". The interest lies in the link between absolute space, which is fixed, and relative space, which depends on the subject of the relativisation and the person doing the relativising (Harvey, 2004).
An example of these different types of space is a street corner -streets are absolute public spaces that are used by everyone. In many South African townships, street corners are particular spaces where young men gather and are likely to be seen. However, young women who stand on street corners are not considered 'respectable'. Therefore street corners as absolute spaces are gendered (Urry, 2001:13) and are appropriated and imbued with particular meanings by collectives. Urry's (2001) three features of space are useful in expanding on the idea of street corners as absolute spaces that are permeated by particular gendered meanings. Urry (2001:11) identifies "spatial practices" which are constituted by, amongst other things, individual routines with the possibilities of being entrenched in the landscape over time. This point captures the idea that street corners are spaces where young men congregate. The perceived unrespectability of young women who occupy the same space is also related to Urry's (2001:11) concept "spatial practices". For example, street corners are occupied by gangs and are also used as gambling spots in different sections of the Orange Farm Township. These practices are not generally associated with young women.
The next feature of space is related to the "representations of space" (Urry, 2001: 11) that order and represent the space. Here gambling can be used as an example of a practice that has become representative of street corners. To illuminate further, it can be argued that the dominant representations of space, coupled with stigma, complicate fieldwork processes (Bingma, 2012). The representations of space and the types of stigma associated with the spaces involve the different meanings bestowed on spaces by individuals and collectives. Gans (2002) argues that a space can have varying effects on different people. For example, the impact of a building may vary for different people and may even be minimal or nonexistent for some people. Here Gans asserts the multiple ways in which groups can "turn natural space into social space and shape its uses" (Gans, 2002:329).
In addition to spatial practices and representations, there are "spaces of representation … symbolic differentiations and collective fantasies" (Urry, 2001:11). The notion of 'home' can be used as an example of "spaces of representation" or "collective fantasies" (Urry, 2001:11). Hooks (1991, as cited in Urry (2001:15), notes how a home can consist of multiple locations. Urry (2001:15) also notes the importance of "mobilities" or the ability to move between these multiple locations. This point captures the dilemma of untangling the role of the researcher who is considered an insider during field observations and general field relations. This dilemma is emblematic of what Bourdieu (1990:56) calls "embodied history" resulting from the researcher's past lived experiences of the researched space -in this case the township.
Considering the above discussion on 'space' and 'place', for the purposes of the following discussion, 'space' will refer to the physical and tangible aspects of a location, such as streets and buildings, and 'place' will be used to identify one's social position or location which is intangible and fluid, but that has real consequences for individuals' everyday lives.

Researching familiar spaces
Social science research often requires the researcher to enter foreign spaces. The researcher may not be aware of the "threats" involved in such research until they have entered the space (LeeTreweek & Linkogle, 2000:10). Whilst these threats may be negative, LeeTreweek & Linkogle (2000:10) contend that these threats can be "methodologically and theoretically productive", as in the case of the research dissertation under discussion. The interest here lies in when researchers enter their current and/or former social worlds and are then forced to negotiate the "mobilities" (Urry, 2001:15) between their current and former worlds. In the following section, the possibilities and limitations of researching a familiar space, in relation to recruiting potential participants, as well as methods of data collection are considered. Complications of stigmatisation with regard to the intended research population are also considered.

Establishing rapport and data collection
When conducting research in one's current and/or former social world, the issue of establishing a rapport with the research participants may be taken for granted. With regard to the dissertation under discussion, potential participants were found, however the target was not met in terms of the number of interviews conducted as mentioned above. Some individuals did not honour the arrangements of the scheduled interviews. Furthermore, these potential participants were again approached and although they voiced their interest in participating in the research, they did not arrive for the interviews. On reflection, it was found that there were two factors at play -one factor in relation to place and the other in relation to space -as well as an intersection between the two which was complicated by stigma (Bingma, 2012).
One has to locate the participants' mistrust of the researcher in the broader context of the stigma associated with early school leaving in general. Early school leavers occupy a particular place in people's imaginations as they are labelled as "outcasts and losers" (Panday & Arends, 2008:1). Goffman's (1963) notion of the management of information was evident during the process of recruiting participants for the study. Individuals who were approached wanted to ascertain what the research entailed and what the risks and benefits of the research were prior to identifying themselves as early high school leavers and agreeing to take part in the project (Bingma, 2012). Selfidentification as a male early high school leaver forced the individuals to place themselves in the social world in relation to others and in relation to the researcher in particular. The thought of being judged as a failure amid increased public calls for a focus on the knowledge economy made the potential participants feel uneasy and ambivalent. Although early school leavers are a diverse group, assumptions about their reasons for leaving school early seem to centre on their perceived academic failure. As some of the individuals under study knew the researcher, they felt obliged to participate in the study when they were approached to do so. However, these individuals were not necessarily interested in telling their stories. The researcher's perceived academic success in relation to the potential participants was perhaps counterproductive to the data collection process (Bingma, 2012).
Studying familiar spaces can also provide many opportunities in as far as the research process is concerned. These opportunities involve knowledge of the physical space and social expectations in relation to local etiquette. Familiarity refers to being aware of the unspoken rules pertaining to the social setting, where to go and where not to go. The researcher was aware of the spaces which she could move in and out of, as well as those that, unofficially, she could not enter during the fieldwork process in the Orange Farm Township because of gendered "spatial practices" (Urry, 2001:11). The target population consisted of, amongst others, young men who potentially took drugs and could be found in backrooms and secluded areas in the township (Bingma, 2012). Such spaces were not visited which resulted in the elimination of a section of the target population because of safety considerations.
Many of the potential participants congregated in social spaces to smoke dagga and to use other drugs (such as nyaupe), as well as to gamble and handle scrap material to be sent to the scrap yard, were not visited. Thus assistants were helpful in facilitating interviews to connect with individuals in such groups. Four participants were recruited through assistants. All of the assistants were male and had reasonable access to the target population because of their gender. The rest of the participants were located by the researcher while walking around the township either alone or with a female friend. The objectives of the research were explained to the potential participants and when an individual agreed to participate, an interview was conducted on the spot to eliminate further hindrances preventing data collection. It can be said that whilst familiarity in this situation was useful, it could also be posited that familiarity hindered the exploration of the possibility of transgressing local etiquette, which would have offered new ways of straddling gendered local spaces (Bingma, 2012).
When an individual agreed to participate in the research, language became an important element in the negotiation process and in the process of bridging the perceived differences between the researcher and the participant. The researcher is proficient in most of the languages spoken in the township. The researcher is also familiar with the local Tsotsitaal which she learnt when she lived in the township and which was also learnt during an Honours small-scale research project involving interviewing three male early high school leavers and a female early high school leaver (Bingma, 2008).
According to the Oxford Dictionary of Sociology (Scott & Marshall, 2009), language can be used to exclude and defend insiders against those that are considered to be outsiders. The above understanding of language is significant as the participants in the study used phrases and words that carried meanings that were fundamentally different to the everyday meanings associated with these phrases or words. For example, the term 'church' means the coming together of young men in a group and the term 'praying' means to smoke dagga (Bingma, 2012). Such prior knowledge of the meanings of these terms facilitated the extent to which participants could 'open up' to the researcher. Also, the researcher's familiarity with the high-school culture in the township made this culture easier to explain by being able to draw on her own experiences. These factors made the process more meaningful and engaging.
Flexible interviewing, as a data collection strategy, provided the means through which the abovementioned factors could be negotiated. Participants were asked about their peers at different points in the interview. They spoke openly about their observations and at times conceded that they did not exclude themselves from the broader problems such as drug abuse associated with young men in the township. It was found that it would therefore not be unusual for individuals to represent themselves in ways that reflect less negative attributes because assumptions about early school leaving seemed to centre on apparent academic failure (Bingma, 2012). Goffman's (1963:3) notion of managing the "discreditable attribute[/s]" of one's social identity is useful in the analysis of the phenomenon of early school leaving. Individuals managed the details surrounding their "discreditable attribute" (Goffman, 1963:3) through a process of selecting what could and could not be said about this attribute (Bingma, 2012). Bloom's (1996) work may also prove useful here and it can be argued that presentations of the self in a context of stigma are non-unitary. Individuals negotiate contradictions and ambiguities in their life course as they tell their stories and place themselves in relation to others.
Another complication of space and place is related to the venue for conducting interviews. An office could not be secured at the local multiple-purpose centre, so space was rented at the Orange Farm Community Library. The library is used for many community projects and specifically for projects focusing on young people. However, only two interviews were conducted at the library hall (Bingma, 2012). On reflection, it was found that the library's spatial location was a problem and that it was also layered with notions of place. The fact that the library is located next to the Orange Farm Police Station was also unhelpful.
Whilst two participants were comfortable with being interviewed at the library, most of the potential participants did not attend the scheduled interviews (Bingma, 2012). Their reluctance may be linked to the relationship between the police and the young men in many South African townships. Steinberg (2010) discusses how young black men living in townships in post-apartheid South Africa continue to be targeted by the police which thus blurs the lines between criminals and young men in general. This occurrence is a familiar one as the researcher has witnessed young men being indiscriminately picked up by police vans and locked up overnight at the local police station.
In relation to the library as an interview space, Gans (2002) notes how a space can have varying meanings for different people. Gans (2002) states that a building can have different effects on different people and that it may even have no effect on some people. Whilst the Orange Farm Library was a neutral space for the researcher and its proximity to the police station had additional safety benefits, it did not hold the same meanings for the potential participants. The police station possibly represented the individuals' tenuous relationship with the police in the community -demonstrating the link between space and place.
At this stage, dilemmas of space and place for data collection are considered. As the researcher was an insider, she could observe occurrences with a 'familiar eye' and as a consequence, she may, for example, have failed to observe what an outsider might have been able to observe immediately. The observations and the stakeholders and participants' narratives often reinforced each other in ways that produced specific readings of Orange Farm (Bingma, 2012). The researcher's familiarity with the physical and social space, as well as her own place in the social space meant that particularities and peculiarities that existed in the community could have been easily overlooked. The researcher saw the township through 'familiar eyes' and possibly took the many significant physical and social changes in the community for granted. Through a process of reflection and close reviewing of the photographs that were taken on each visit, it became clear that more time should have been spent on observation.
The abovementioned factors underline the "mobilities" noted by Urry (2001:15). It could be said that Orange Farm is both a current and former world to the researcher. It is a current world because of the researcher's continued connection with the township through family ties and friendships. It is also a former world in terms of the researcher's current everyday lived experiences that are situated outside of the observations made on the many continued visits. In essence, the researcher relied on the "embodied history" (Bourdieu, 1990:56) of the social space and, to an extent, the absolute space (Harvey, 2004). As a corrective measure, a friend accompanied the researcher on subsequent visits and noted, for instance, possible 'facts' that were missed by the researcher on earlier fieldwork trips (Bingma, 2012).
Photographs became useful as a methodological corrective measure through which observations could be checked and the collected data could be studied more rigorously (Bingma, 2012). Photographs were important not only in capturing the landscape, but also as a form of visual representation of the space and "spatial practices" (Urry, 2001: 11). Photographs also visually represented the young men engaged in activities such as gambling on a daily basis, as well as young men collecting scrap metal. Ultimately, it was evident that utilising multiple sources of data was invaluable as it allowed for crosschecking of information and made it possible to identify similarities, contradictions, misrepresentations and biases.

"Danger in the field"
Researching a familiar space does not only pose a danger to the nature and quality of the data collected, but it also impacts on the researcher. In the research process, a researcher's wellbeing can be under threat because of field relations (LeeTreweek & Linkogle, 2000). The danger may be exacerbated when the field site has "personal meanings" for the researcher (LeeTreweek & Linkogle, 2000:16). The social expectations of the stakeholders and participants blurred the role of the researcher.
As an insider, the researcher did not anticipate the personal emotional impact that researching her home environment would have on her and the progress of the research. It became increasingly difficult to deal with the emotional impact of the participants' stories as the researcher knew, recalled, and identified with much of what they recounted about daily life in Orange Farm. As the researcher had grown up in the area, she experienced a set of similar challenges to those narrated by participants. Ramphele (1993:139) notes that on entering the world of migrant labourers in her study of hostels in Cape Town, her exposure to the environment highlighted the limit that "one's psyche can take in such an intense setting". Although the research here focuses on young men, it was, in many ways, a reflection on the researcher's own past and broader experiences of the township. Becker (1967) notes similar issues when he points out that researchers may sympathise with those they are studying and they may believe that those under study may be treated unfairly as they are "more sinned against than sinning" (Becker, 1967:240). These points made by Becker (1967) indicate the need to reflect on relations with participants and how these relations can have an impact on the data reading process. There is a need to manage the unmanaged and unspoken expectations of the communities and participants carefully, particularly when researching socially vulnerable and/or stigmatised groups.
As an insider, when the researcher introduced herself and explained the project to different stakeholders in the community, these groups often praised the researcher (Bingma, 2012). However, this praise was often loaded with enquiries into what the researcher would do to improve the situation. Similarly, at the end of each interview, each participant asked what the researcher would do to help. It was clear that the stakeholders and participants placed the researcher in what Newman (2002) describes as a position of power in the social interaction. Associating the researcher with a position of power involved not only the researcher's perceived academic success in relation to the target population, but also the assumption that she understood everyday life in the township.
As a relatively novice researcher researching a poverty stricken area, the expectations of change and assistance resulted in a growing burden of searching for what could be done. The spoken and unspoken expectations of those being researched became critical to contain. The researcher required debriefing sessions with the supervisor and a friend. These sessions were useful in dealing with the emotional impact of the research process and also helped the researcher to reflect critically on the thought processes involved in the project. Debriefing was important as conversations with others increasingly indicated the researcher's frustrations, which made proceeding with the study increasingly difficult. Visits to the researcher's family home were limited and the data were read in the context of Orange Farm particularly.

Conclusion
Whilst qualitative research investigating male early high school leaving is necessary to understand why male learners leave high school prior to completing Grade 12, contextual dynamics must be noted in as far as such complications can impact the recruiting of potential participants, fieldwork relations, and data collection. Space and place are instructive in making sense of the researcher's and the participants' place in the social space. It is suggested that an awareness of 'give and take' and the constant negotiation of the power struggle between researchers and the research participants cannot be stressed enough. Furthermore, for researchers conducting research in familiar spaces, qualitative methodologies present challenges and opportunities through which "dangers in the field" (LeeTreweek & Linkogle, 2000:24) can be mitigated. The integration of multiple sources of data is perhaps the most critical measure to counter unintended impacts on the research process with regard to novice researchers undertaking research in familiar spaces.