Why Did I Not Prepare for This? The Politics of Negotiating Fieldwork Access, Identity, and Methodology in Researching Microfinance Institutions

It has been increasingly recognized that undertaking qualitative research can pose many challenges for researchers. However, scanty literature focuses directly on the experiences of doctoral research students from developing countries studying in Western Europe and other similar geographic regions, and the challenges of doing fieldwork when they return “back home.” In this article, I use my experiences in the process of undertaking PhD fieldwork on two donor-funded microfinance institutions located in Zambia to demonstrate that doctoral students from specific regions (Africa in particular) undertaking research in their native countries can struggle to manage and make sense of the challenges and identity issues raised in their “familiar” environments. I also present a detailed discussion of how various gatekeepers and participants facilitated access, identity alteration, and the impact of insider–outsider positionality on collected data. It is concluded that organizational “politics” and local context can have significant bearing on power relationships, identities of researchers, and methodological preferences.


Introduction
Qualitative researchers have increasingly stressed the need for reflection on the positionality and identity of the researcher in the field (Berger, 2015;Ergun & Erdemir, 2010;Ezzy, 2010;Humphrey, 2007;Mannay, 2010;Soni-Sinha, 2008;Taylor, 2011;Turner, 2010). Other researchers, however, have highlighted the ethical and methodological challenges involved in working in developing countries (Cornet, 2010;Crossa, 2012;Geleta, 2014;Guevarra, 2006;Kiragu & Warrington, 2013;Mandiyanike, 2009;Rubin, 2012;Sultana, 2007;Turgo, 2012;Visser, 2000). Nonetheless, although these scholars conducted their research in the specific geographic regions, the research contexts and geopolitics of their work vary considerably. Many accounts of methodological challenges and dilemmas are based on research in South Asia and Latin America and biased toward the experiences of ethnographers and human geographers. However, there is a dearth in publications by African doctoral students studying in the West and other similar contexts who return "home" for purposes of research and are thus able to consciously reflect upon their fieldwork dilemmas and experiences. For example, Mandiyanike, researching in his native country, Zimbabwe, writes about being treated with suspicion by some of the participants in his research because of his "connections" with the United Kingdom as a study base. Geleta (2014), on the other hand, presents an ethnographer's perspective, reflecting on the fluidity of the "insider/ outsider" status in Ethiopia and how his methodological orientation complicated the process of identity negotiation. Kiragu and Warrington (2013) also highlight the ethical and methodological complexities they faced while conducting research with schoolgirls in Kenya. Others, including Rubin (2012), a researcher working in both contexts (South Africa and India), talk about layered identities and the complex ways in which insider-outsider status was experienced within and across different sites.
This article seeks to add to the growing methodological literature on experiences of natives researching back home in a developing country context by focusing on Zambia and its microfinance industry. Three main insights are drawn from 587560S GOXXX10.1177 1 University of Lincoln, Brayford Pool, UK this research, and the article therefore makes a contribution to the literature in three main areas. First, the importance of undertaking pilot studies is emphasized, especially where researchers have limited prior knowledge of those to be researched and their environment. Research takes place in varied contexts and unique organizational and community circumstances-meaning that even familiar places and spaces can appear strange. As Van Teijlingen, Rennie, Hundley, and Graham (2001) state, "Pilot studies are relevant to best practice in research, but their potential for other researchers appears to be ignored" (p. 289). Although I was familiar with Zambia as my home country, I had very limited prior understanding of the processes within the microfinance sector in the country.
Second, the article considers the gap between the assumptions underpinning the theoretical methodological literature on research processes and methods, and what was practically achievable in the field. While it is crucial to highlight the gap between approaches "in the textbooks" and practical experience "in the field," this study makes a more specific observation that the established literature primarily involves researchers working across the North-South divide. In addition, the literature focuses heavily on Western researchers' experiences of conducting research on development projects in Africa and rarely on the experiences of African researchers themselves. This gap in the methodological literature means that for inexperienced researchers and, in particular, those crossing research methodological divides, preparing for and managing fieldwork dilemmas can prove to be more complex than the process is often described to be.
Third, I highlight the difficulties of coming to terms with the reality of, despite being Zambian, not being an insider all the time, and the way in which my multiple identities were not always met with trust and support. Returning to Zambia to conduct fieldwork posed several dilemmas for me. As Sultana (2007) notes, The "field" versus "home" is a problematic distinction. Undertaking research at "home," therefore, resulted in different dynamics, particularly in relation to concerns about insider-outsider status and politics of representation. In writing this article, I aim to contribute to the growing literature on the challenges, politics, and methodological complexities of insider research (Cornet, 2010;Geleta, 2014;Rubin, 2012;Sultana, 2007;Turgo, 2012) in developing countries. Importantly, the contribution that the article makes is in part related to my insights as an African researcher.
To advance these claims, I begin by briefly discussing previous literature of research politics and my own research orientation. I then move on to talk about the research context and challenges of establishing initial access even in a seemingly familiar environment. This is followed afterwards by my experience in the field. Here, I deal with the politics of researching, where I detail my relationships in the field and how my identities as researcher were constructed, and the challenges of navigating the complexities that come with insider/outsider position in one's home country. In particular, I emphasize the challenges and dilemmas of navigating the ever-shifting field identities, further complicated by my naivety. Thereafter, I discuss how the politics of institution (re)shaped emerging relationships with senior management and loan officers and the data for which access could be granted. I conclude by first summarizing the major arguments set forth in the article, which underscores the importance of reflexivity and an understanding of how research participants impact the conduct of research in the field. I finally end with a couple of lessons for other scholars and doctoral students conducting research in developing countries.

The Literature
Much of published research in microfinance has traditionally been guided by the positivist paradigm that assumes an epistemological stance where "reality" is singular and research is ideally claimed to be objective, detached, and unbiased (Bryman, 2001;Laws, 2003). Within this paradigm, there is an expectation of the researcher conforming to methodological exactness and being distant from those being researched. As Wall (2006) notes, traditional scientific approaches require researchers to minimize their selves, and put bias and subjectivity aside by denying his or her identity. I however use different lens and take a position that the field is a politically contested terrain (Clifford, 1988), thereby requiring, as Wall notes, a disclosure about the situatedness of the knower and the relation of the knower to the subjects of inquiry. My research has no claim to being fully ethnographic in design, except for the methods used such as participant observation and shadowing in studying the ways in which loan officers interacted with clients of microfinance institutions (MFIs). This is because microfinance and the role of loan officers are played out "live" in the field through social interactions. It is therefore informed by an orientation that reality was socially constructed and that there were multiple things to be known and not singular as implied by positivists (Bryman, 2001;Silverman, 2005). This meant that my positionality and identity were not given but multifaceted and emerging knowledge socially constructed (Crotty, 1998;Denzin & Lincoln, 2003) by actors in different ways. The researcher and the researched, in this view, emerge as active collaborators of meaning through their interaction (Lincoln & Guba, 2000;Tembo, 2003). Therefore, rather than taking a detached stance, my ways of inquiry connected with real people and their issues with accessing microloans.
This article is based on the author's fieldwork experiences in Zambia and discusses the complexity of doing fieldwork in one's homeland due to issues of access, identity, and power. In particular, I engage in reflexive recounting through autoethnographic personal narrative (Cole, 2013;Crossa, 2012), and also acknowledge my effect on the researched and how my thinking and "doing" research was shaped by field dynamics (Van Maanen, 1988). Alvesson, Hardy, and Harley (2008) state that "reflexivity is not only important to the understanding of what happens in research but also requires researchers to declare their authorial personality" (p. 497). Berger (2015) also notes, "Reflexivity means turning of the researcher lens back onto oneself to recognise and take responsibility for one's own situatedness within the research and effect on the setting and people, data being collected as well as its interpretation" (p. 2). In recounting my field experiences, I admit that the positivist perspective did not prepare me for what I then came to understand that, out there in the field, researchers inherently occupy tenuous positions because of the power that participants possess in defining their access and identities (Guevarra, 2006). In this study, loan officers and clients of MFIs were not simply passive recipients of my claim of authority and research agenda but turned out to be active agents who reshaped the way data were collected and the practicality of the research process. To facilitate this collaboration, I had to employ ways of inquiry that connect and dialogue with real people and their issues of accessing microcredit, as the key question of my research was the following: How does group-based microcredit actually work on the ground rather than the espoused narrative?

The Research Study
In writing this article, I use an autoethnographic account (Muncey, 2005;Sparkes, 2000;Wall, 2006) and draw on my own experiences of the process of undertaking fieldwork on two donor-funded Zambian MFIs to generate a sense of what was on the ground. A caveat should be mentioned here, though. This article is not specifically focusing on the detailed accounts of research methodology approaches, data analysis, and findings per se; these are reported in Ritchie (2012, 2013). Here, I reflect on the method in practice based on my doctoral fieldwork, the first undertaken as a pilot study between November and December 2003, followed by a longer phase between May and July 2004. My PhD research stemmed from a growing interest in, and optimism about, microfinance as an "endorsed" 1 development strategy for poverty reduction on one hand, and the unexplained variable takeup in Sub-Saharan Africa (SSA) on the other. The aim was to study microfinance in action and from the bottom-up-from the perspective of clients and loan officers, thus tapping into local knowledge of how microfinance works rather than how it should work. Very limited evidence has emerged, especially from Africa, about what loan officers and MFIs do and how they actually undertake their activities. Donor-funded research prior to my own study often involved impact assessments of microfinance, intended to highlight its virtues and positive prospects (see Copestake, 2002;Copestake & Mlotshwa, 2000, for examples of such work). Research by these scholars and consultants focused on making a case that microfinance does empower the poor to justify continued external funding. This research, however, aimed to examine how group-based microfinance works from the perspectives of loan officers and their clients. After successfully completing my doctoral studies, I undertook another piece of research in May 2010 with the same MFIs, this time focusing on the changing role of loan officers as microfinance becomes increasingly commercial.
Loan officers are frontline employees of MFIs who come to acquire context-based knowledge by going into the "field" and interacting with clients in their own milieus. My central research aims were therefore to examine the experiences and roles of these loan officers in the development of microfinance and the ways in which they interacted with clients of MFIs. To understand the practicalities of microfinance, the roles of loan officers and their experiences and interactions with clients, an applied ethnographic approach to the study design was deemed appropriate.
Over a total period of 5 months, I shadowed loan officers, attended staff/client group meetings, applied observation techniques, and coupled these with interviews. The choice of shadowing as a method seemed appropriate, though disruptive to research participants (Gilliat-Ray, 2011). Indeed, as I came to learn later, this technique was better suited to observing and documenting the work of loan officers, which (in the field) seems to be unstructured and constantly constructed through daily routines and interactions with clients. Microfinance and the role of loan officers are not played out "live" in the office but in the field. Shadowing therefore offered the opportunity to explore how microfinance actually works and allowed me to witness some of the challenges loan officers face as they attempt to deliver financial services to the poor. I came to appreciate loan officers' experiences and their positions as "foot soldiers" (Chua, 1998) of microfinance. In addition, shadowing offered significant insights that would have been largely unobtainable had I used questionnaires and interviewing as my only research methods. Opportunities for informal interviews with individuals, especially loan officers, were also sought, often taking place spontaneously to fit with their limited time.

Entry-First Stage: The Pilot Study and Lessons Learnt
One MFI was chosen for the pilot study because it was one of the few Zambian MFIs that had previously been the subject of research by other academics, with traceable publications in the public domain. In addition, this MFI had been in operation for 5 years at the time of my first fieldwork visit and was therefore in a position to throw some light on how microfinance was engaging with the poor. Only one branch of five located in the Copperbelt participated in the pilot study. 2 In negotiating entry into this MFI (to be referred to as MFI T), I made full use of my being Zambian as well as my academic position as a lecturer (insider), because both facets of my identity enabled me to establish my credibility as someone committed to the development of microfinance and my country, even though I was at the time located in the United Kingdom. Entry negotiations began with the chief executive officer (CEO), without whom the research would not have proceeded. I emailed the CEO, explaining who I was and identifying myself as an academic at one of the established Zambian universities, indicating that I was asking for permission to conduct research. I was not particularly known within the development sector and non-governmental organization (NGO)-led microfinance world, making negotiating entry a daunting undertaking. I therefore decided to emphasize the relevance of my doctoral studies to local developmental issues, after which I was asked to send an outline of my research proposal. Following exchange of several emails, permission was granted 3 months prior to my first field trip in November 2003. The CEO expressed interest in supporting the research because it resonated with the overall mission of the organization, reducing poverty by empowering women through microfinance. I had originally envisaged that once permission was granted from the top, access to sites of interest would be less problematic. Instead, however, I found myself having to renegotiate with the branch manager and then individual loan officers-the "real" gatekeepers 3 in bottom-up research into microfinance. As I explain later, negotiating entry and gaining permission were not one-off events (Cornet, 2010;Kiragu & Warrington, 2013;Tembo, 2003).
My research involved use of questionnaires, semi-structured interviews, and focus group discussions with loan officers. I also conducted interviews (in English) with the manager and accountant at branch level, as well as the operational manager and the CEO at head office. I had not thought much about observation and shadowing as useful ways of collecting data but lessons from conducting a pilot study provided space to reflect on the appropriateness of my methods with the reality of loan officers' work. Consequently, some changes to the research methodology were made after the pilot study stage because of the constraints imposed by "gatekeepers." For example, I was told: "Do your work but do not disturb loan officers." This resulted in my conducting shorter face-to-face interviews with loan officers (of not more than an hour), such that I instead relied more on observation and shadowing them in the field to better understand their work. This methodological twist provided useful insights and enabled me to build up a vivid account, detailing the context in which microfinance existed at grassroots level. I observed the harsh environment in which loan officers and most of their clients, especially women, worked. Loan officers often make lengthy journeys, at times coupled with very early starts. Marketplaces where most of their clients conduct their businesses tend to be noisy, busy, muddy, or dusty, making the environment unattractive to work in. In addition, holding conversations with loan officers as we traveled to meeting places and made follow-up visits to defaulting clients provided an opportunity to reflect on their attitudes to clients and for me to clarify certain observations.

Research Site and Access: The Main Fieldwork
The main research site was in the Copperbelt province of Zambia, where I had lived and worked for over 10 years. The second stage of fieldwork (between May and July 2004) now involved two MFIs (T and P), both with branches in the Copperbelt but with headquarters in Lusaka. I chose the Copperbelt because I could speak the language (Bemba) used by loan officers in meetings with their clients, without requiring the services of an interpreter. Indeed, almost all clients with whom I interacted were comfortable with using it rather than English. I was also familiar with the local cultural expectations of the people in that region, which helped me to appreciate and understand what was culturally acceptable. While "familiarity," as noted by Cotterill and Letherby (1994), can reduce initial problems of access, it can nevertheless create other problems; social class barriers can make one a "stranger" in any locality, even a "known" locality. Accordingly, Hawkins (2010) and Crossa (2012) make an observation about the difficulties of studying populations with backgrounds similar to those of researchers at one level yet different at other levels. In this case, I was both an "insider" and an "outsider" (Acker, 2001;Crossa, 2012;Merton, 1972;Paechter, 2013;Rubin, 2012;Turgo, 2012). This reinforces the notion that boundaries between "insider-outsider" are in practice porous and complex. For example, in this research my being insider came about because I was in a place I called "home" and shared a common local language. But classifying someone as an "insider" based on the ability to communicate in native language is simplistic as everyone's "home" can acquire multiple interpretations based on researcher's relational experience and how their identity is crafted by research participants. For example, older female clients of these MFIs would occasionally refer to me as "one of them," therefore an "insider" because I was, at one level, like thema woman, mature and married. On the other hand, I was unknown, an "outsider" or a "stranger" to MFIs because I came from outside the microfinance sector. And yet to other participants, I was constructed as the "other" through class privilege and was clearly not "one of them." As I later explain, this insider/outsider binary became highly dynamic in time and through space (Mullings, 1999) as my multiple identities meant different things in different contexts (Rubin, 2012).
I believed, rather naively, that research participants would be free to talk to someone "native," especially as, in the past, most research within microfinance had been conducted by non-nationals. However, as I soon discovered, loan officers in particular did not easily engage with the "politics" of MFIs and, even where they did, it took some negotiation before they "opened up." In the later stages of the study, however, loan officers came to be less wary of me and talked openly. Indeed, one remarked that: our lending methodology is rigid and the relationship between the donors and our organisation is that of "specialist" and "learner" and yet it is supposed to be the other way round. Because of such relationships, donors tend to dictate and the management here fails to challenge some of the instructions given to them. Management would rather agree with them [donors] than get our ideas or suggestions. Their defence is "this has worked elsewhere" so it will have to work here as well though management knows that some aspects of the methodology don't suit our environment. They are not bold enough to put it across to the donors. The population we are dealing with is different from that found in Asia. (Male L/O) The Politics of Negotiating and Maintaining Access Gummesson (2000) identifies three different access types: physical access, meaning the ability to get close to the object of the study and really be able to find out what is happening; continued access, referring to maintenance of ongoing physical access; and mental access, referring to the ability to understand what is happening and why. Central to these types of access are gatekeepers (Hammersley & Atkinson, 2007;Mandel, 2003;Reeves, 2010), who can help or hinder research depending on their personal views on the value of the research to their organizations. Gatekeepers at various levels needed to be approached for this research, indicating that gaining access was not going to be a one-off event but, as Van Maanen (1988) describes, " a continuous push and pull" (p. 144) and therefore part of an ongoing process to be revisited every day over the course of the fieldwork (Reeves, 2010). For this fieldwork, access could not be negotiated on one single occasion but involved continued negotiation and renegotiation (Ahrens, 2004;Bryman, 2001;Burgess, 1991;Hall & Hall, 1996;Schatzman & Strauss, 1973) and the development of ongoing relationships. As Mosse (2005) notes, For outsiders access to the workings of development agencies (or institutions) is difficult. For one thing, such agencies operate within a nexus of evaluation and external funding which means that effective mechanisms for filtering and regulating the flow of information and stabilising representations are necessary for survival. (p. 12) Consequently, access to quantitative data by an outsider was not always welcome and often restricted. Peil (1993;cited in Mandel, 2003, p. 203) notes that "even experienced researchers often assume that once top authorities have given their permission, everyone lower in hierarchy will fall into line." I had my letters of introduction written at head office and approved by senior managers, and I thought to myself, "I am in!" My excitement was temporary as I quickly realized that this was just the beginning. I was given a condition-"provided your research does not disrupt the work of loan officers." Lee (1993, p. 124) considers conditional access to be a situation where gatekeepers often "allow" researchers into a setting but then use formal agreements and procedures to control their activities. In this case, the injunction was the following: "Do your research but do not disturb us." Others also note that formal access does not guarantee rapport with individual informants (Cornet, 2010;Laurila, 1997;Lee, 1993;Mukeredzi, 2011). Branch managers, as well as loan officers, were powerful gatekeepers to observation of how microfinance actually works in the field. Operational managers concerned about meeting targets expressed concern that I would "get in the way" of loan officers and wanted to know "how much time" the research required. As Clark (2011) notes, not all gatekeepers will agree to research requests and in some instances may even attempt to block access to some parts of organizations. For example, in the initial stages of my research, some loan officers wanted to block access to client groups with repayment problems because they feared bad exposure. More importantly, they thought I was only interested in the "good stories" of the poor about microfinance. One loan officer commented, "When donors and these consultants come here, we are very selective in which client groups they get to visit and talk to." In this research, I had to negotiate with multiple gatekeepers (Hammersley & Atkinson, 2007) because the participants in question could not be approached directly and an intermediary was required at each level to facilitate access. Gaining access was therefore a "social process of negotiations" (Bondy, 2013, p. 1) and situationally specific, resulting in what Lee describes as "seemingly unlimited contingencies."

The Politics of Field Identities
As researchers we spend much time unpacking our research questions, developing methodologies, and worrying about how to explain our aims so participants can understand our interests and questions. Yet what we too often fail to consider is how much the questions we ask, and the answers we receive, remain contingent on who participants assume we, as researchers, are. (Razon & Ross, 2012, p. 497) While the prospect of going back home to do fieldwork was exciting and challenging, I gave limited thought to how people in the field would construct my identity. As Razon and Ross (2012) rightly observe, I did a good job of developing methodologies, polishing up research questions, and constantly rehearsing how I would explain my research aims to my participants, but underplayed the "politics" of field identities, as the narratives that follow demonstrate.

"Who Are You Doing This Research for?"
Embedded in this first question were a whole host of other queries, and only later did I come to understand why I needed to make this clear. First, research within the microfinance sector in Zambia has traditionally been undertaken and sponsored primarily by donors from developed countries. At the time, Western donors were funders of the two MFIs I studied. Second, knowing for whom I was conducting the research later determined the dynamics of "what" information was disclosed or withheld and "how" participants then conducted themselves. For instance, branch managers and loan officers suspected, initially at least, that I was funded by management to "spy" on them. There was a fear that "deviant" activities and practices would be reported and that perhaps my research was intended to uncover discreditable information about them. In the initial stages of the research, loan officers were therefore interested in telling me about how they "loved" their jobs, as well as highlighting the virtues and empowering power of microfinance for its clients, the poor. Clients, on the other hand, wanted to know whether I was collecting information for the "owners" of MFIs-the donors. However, some clients saw the research as providing space to expose the "misdeeds" of their loan officers. The above question and the ones which follow were aimed at positioning me and enabling the subjects of my study to feel at ease and engaged in conversations "Are You Doing This for Donors?" Donors represented a "privileged" category within the development sector and microfinance in particular because they were synonymous with external funding. They mattered to institutions' continued survival and, in practice, were treated as insiders. As such, there was a natural line of upward accountability to them (Dixon, Ritchie, & Siwale, 2006). The power implications in this MFI-donor relationship were obvious. One long serving loan officer put it this way: Things would have been a lot easier for you if your research was connected to one of our donors. You see, donor sponsored researchers and consultants do not have to negotiate access with anyone because they are regarded as part of the system that funds MFIs and no one probes them anyway.
There was, however, a downside to this privileged donor position as I later learnt from frontline employees of the MFIs. Some loan officers suggested that there was pressure from top management to impress donors with "good" data to justify their continued funding and to make a case that microfinance does indeed help the poor. Establishing whether I was "one of them" (donor sponsored researchers) therefore had the potential to alter relationships and data made available to me. Realizing both the potential privileges and the disadvantages of being identified with donors, I had to ensure that I distinguished my research project as purely academic with no donor involvement. Despite all my assurances, some loan officers were still suspicious. Comments were made such as "who is behind this work?" and "why us?" Having established my independence from donors, I hoped to move on but repeatedly found myself interrupted by other questions from participants.

"Is It for Management Then?"
This question was particularly puzzling initially as I had not realized the extent of mistrust between management and those with direct contact with MFIs' clients-loan officers. Once I had successfully distanced my research from donor influence or support, there was-in the minds of my interviewees and participants-only one option left to account for my presence in the field: that I had been hired by senior management to spy on branch managers as well as assess the performances of loan officers (Bowling, 2002;Bryman, 2001). Given such preconceptions about my identity, gaining the trust of loan officers and their supervisors was challenging and time-consuming in both MFIs. Although I had achieved physical access, the need for "social" access required me to gain the trust of loan officers (Marshall & Rossman, 1999) and also develop credibility with them, by maintaining some distance from the management. In this study, interpersonal relationships were important in negotiating "social" accessibility (Marshall & Rossman, 1999;Sixsmith, Boneham, & Goldring, 2003). I had to prove that I was a reliable and trustworthy researcher and that all I was concerned about was gaining knowledge about their roles and how microfinance actually works, not with evaluating their work performance. Being "visibly around" and shadowing them was somehow threatening as, at the time, they did not know where my loyalties lay. In the very first few weeks of my fieldwork, I accompanied a female loan officer to her group meetings. Along the way, I picked up a conversation, and part of it went as follows:

JNS:
So what has kept you going? L/O: It's just God. I love my job and want to see poor people change. JNS: Would you not put it down to good conditions of service? L/O: I don't want to comment because you might report me to senior management.
Loan officers in particular feared that their dissenting views and ways of "doing" microfinance (e.g., not fully complying with lending methodology) could be exposed. For example, contrary to lending policy, some loan officers admitted to including their relatives in groups they managed. They told me that this was an indirect way of helping their extended families but could become a dilemma when these relatives defaulted on their loans. Others ignored the rule that members of the same family should not be in one group, so as to guarantee that peer pressure worked in repaying back group loans. In other instances, loan officers revealed that, to cope with pressure, they found it "convenient" not to turn up for meetings with client groups that were up to date with repayments to focus on those in arrears. Effectively, group meetings with clients were often reduced to mere collection points instead of spaces where basic training in book keeping and methodology was offered, as stipulated in official documents.
Loan officers worried more about repayment targets than training clients and making physical visits to clients' businesses. In addition, multiple borrowing across MFIs was widespread, creating more debt pressure on already indebted poor clients. Interestingly, loan officers knew about this but never warned clients against the practice because loan officers only cared about repayment and not where the money being repaid came from. From their field experiences, loan officers knew that group lending methodology was not working and that many of its rules were not adhered to by clients or themselves. In one group meeting, clients openly commended their loan officer for flouting some of the lending rules to enable them to access further loans. Although these "field tactics" were not explicitly harmful to clients, they nevertheless undermined the long-term sustainability of groups on which continued lending was premised. The advantage I had in accessing these "hidden" practices was my ability to understand and speak the local language (used in meetings with clients), providing a space where a Zambian gaze had the potential to generate richer context-based knowledge than that of a Westerner.
Shadowing loan officers meant that I soon came to know their everyday work away from the official space-the office. This realization became problematic, hence the interest in establishing "whose side I was on" (Becker, 1967) before they could trust me. The message was clear: "If this is for management then you are not on our side because we loan officers and management do microfinance differently and they [management] neither trust us nor understand the realities of working with the poor." Failing to build rapport with loan officers could have resulted in them keeping perspectives from me that addressed the very questions my research sought to explore. Striking a balance between management and loan officers without being viewed as trying to undermine either was fraught with difficulties. For example, managers offering to collect completed questionnaires for me made loan officers suspicious of my reasons for being there, while arranging to interview loan officers away from organizational premises or outside office space made some senior managers suspect my motives-and occasionally asked me to share my "field" experiences with them. At this point, it became clear to me that the suspicions loan officers had about me "spying" on them for management had been valid. Take an example of an earlier conversation with a female loan officer who refused to comment on their conditions of service for fear that I might report her views to senior management. This was a real dilemma as I risked being drawn into organization politics, with a danger of turning my research into some kind of a "surveillance" project. Managers wanted me to comment on the practices of individual loan officers when in the field. I purposely destructed them by talking about their clients-the poor, instead of loan officers. Because this research was about how microfinance actually works (from bottom-up) and not how it is managed from the top, it was easier for me to keep away from senior managers without compromising my data collection. Managers neither collected completed questionnaires for me nor did I discuss individual loan officers' conduct in the field, as doing so would have confirmed my "spy" identity and undermined the trust with loan officers.
Introducing my research, I presented myself as the learner that I was by explaining to loan officers and branch managers that I was there to understand how microfinance actually works in empowering the poor. In doing this, I enabled them to occupy the "expert" position, with its empowering effect, and thus put them at ease with the realization that I was there to draw upon their knowledge. This sense of empowerment was important for them because, unlike Western or donorsupported researchers, I was not perceived as a "power figure." In general, loan officers were enthusiastic and pleased that I had chosen to work with them, stating that mine was the first research to focus on the role they played in the development of microfinance. They felt that their importance was at last being highlighted and that someone was giving them a "voice." Once this threshold was passed, loan officers granted access to client groups, cautioning me to dress simply (i.e., not in expensive clothes associated with the West or the United Kingdom), and in a similar style to the older female loan officers. In short, "be decent, no showing off!" I also decided to converse with clients in Bemba (the local language) to be seen at least as "one of them." I thus only used English if they felt comfortable with doing so.
Nevertheless, this granting of access to clients and their meetings came with dilemmas. On a number of occasions, loan officers used my ever-shifting identities to their advantage. For instance, a loan officer made this announcement at one of the meetings: "Today we have a visitor [referring to me] from head office to see how we conduct our group meetings." This caught me off guard and, after the meeting, I inquired of the loan officer why he had introduced me as "one of them." He replied, "To instill fear and help them get serious and attentive because then they think that with good behavior you will recommend them for larger amounts of loans." How do you deal with such manipulation of the poor when you clearly know they are being taken advantage of? What would have happened had I interrupted the loan officer at that point and gone on to make my own introduction? Was this an indirect way of asking me to reciprocate his "favor" in granting access? Loan officers were in control of client groups and exerted undue influence of how individual clients accessed loans and subsequent repayments. In this case, the loan officer presented me as a "power figure" from head office and not a researcher to serve his own agenda-loan repayments. Moving on to another meeting, in a different township, I explained to the loan officer the risk of making clients believe I was an official from head office, as this amounted to deceit and could jeopardize my credibility.

"We Too Want to Know Who You Are": MFI Clients
Clients asked critical questions about where I was from, my intentions, and how the research would benefit them. Interestingly, clients were just as fascinated by my identity and research as the loan officers. Are you from head office? Who are you doing this research for? Is it for the donors? These questions partly formed the basis for building up relationships and trust before clients would engage with me as a researcher, while enabling me to establish an appropriate field identity. I did not, however, anticipate that interacting with clients would challenge the "fixed" identity I had in my mind. I was a doctoral candidate/student from the United Kingdom and an academic affiliated to a local university. I was little prepared for the interrogation that followed. At a weekly group meeting for clients, soon after I was introduced by their loan officer, the following unexpected exchange ensued: Group leader: Who are you and where are you from? JNS: A researcher from Copperbelt University undertaking an academic piece of research. Group leader: Oh! Are you a student then? You are wasting our time because we have had too many of them and nothing happens. Group treasurer: Even these White people have been here and gone and nothing has changed. Group member : Well maybe it's because loan officers and branch managers usually warn us to only say the good things about our experience with microfinance so that donors can continue to give money to the poor.
The messages I was receiving were "not another student with more questions again," "we're over-researched here!" (Clark, 2008, p. 956), and "if you are a student, then know that we are in control and it's within our power to deny you our time for your research." These comments occurred when clients learnt that I was not working with donors funding MFIs. In addition, some clients and group leaders were distrustful about the value of my research, commenting that previous work with them had done little to address their concerns about aspects of the lending methodology with which they were not happy. This conversation provided an opportunity to reflect on why, in the early stages of my research, loan officers offered to find potential clients for me to interview (probably the "best" clients) in line with trends set in previous research work. From the perspective of the clients, research was not persuasive and did not make much sense if it did not address their immediate concerns.
Realizing that giving myself a student identity was not going to work, I went on to introduce myself as a lecturer from a local university as an alternative way of representing myself. To my surprise, clients did not believe that either. A group of women sitting right in front of me commented: "She looks too simple to be one." Another wave of scrutiny followed. Why would you choose to associate with the "low" class? What is your interest in all this and what benefit are you deriving from doing this research? Is it private work for which you are being paid and if so why should we help you with our time? Most people I talked to (not just clients) interpreted research work as meaning consultancy that came with a lot of money. In this context, I emphasized that the research was for educational purposes with no personal financial gain, something my participants struggled to appreciate. Throughout the fieldwork, it was necessary to consistently establish my identity as an independent local researcher from the Copperbelt University, not a hired consultant, to gain some trust and "acceptability." Among the clients, my position as a doctoral researcher did not make sense as most of them had very low education levels. On the other hand, the student label was disempowering. I therefore downplayed my U.K. doctoral student status in my conversations and emphasized instead my links with the local university as a lecturer. Having access to a vehicle with a visible Copperbelt University School of Business logo during my fieldwork provided credible proof of my identity, showing that I was "genuinely" local and not one of "them"-the donors.
So how did my inability to disclose where I really was based at the time (the United Kingdom) structure my interactions with participants-especially the MFI's clients? In Zambia, as in most developing countries, working (or even studying) abroad signals a better life and a different status. Mention of the United Kingdom could therefore have sent a wrong signal that I was someone well off and with lots of money to give out. It would also have been very difficult for me to deny any links with donors, something I needed to do if this research was to achieve its aim. Thus, I was, to them, a Zambian female researcher. About this, they commented, "We have not had women of your class showing interest in what our 'world' is like except for White females sent by donors." Consequently, my "student-researcher" identity did not work for me because it commanded very little power over the negotiations, and I was perceived as a time waster. To most of my participants, my most prominent identities were my nationality and my gender-being a woman made me non-threatening. To female clients, I was categorized as "one of them," and thus expected to empathize with them more than Zambian males and non-Zambians in general would. Srivastava (2006) argues that "field identities are multiple and continually mediated constructs in response to the anticipated or experienced perceptions of how participants receive, accept or reject the researcher's positionalities vis-à-vis their own" (p. 214). Ergun and Erdemir (2010) also stress that the identities of researchers in the field are in a process of constant negotiation and shaping of knowledge produced, while Kusow (2003) characterizes them as "frequently situational" (p. 592). Others have noted that insider status can be challenged by the research process itself (Crossa, 2012;Humphrey, 2007;Taylor, 2011). I reached the realization that, although native (and therefore an insider), I was an outsider to MFI's clients, the poor. I was never fully "matched" (Merton, 1972) to my research participants at a grassroots level. I became acutely aware of my class and educational privilege. I belonged to a different class with limited practical understanding of the daily struggles of the poor and therefore on that basis was not very different from non-native researchers. However, on reflection, I think that the fluidity of my insider/ outsider status in this study gave me certain advantages that an insider (from MFIs) might not have.
One incidence demonstrates this quite clearly. At one of the MFIs, I asked the finance manager for Portfolio at Risk (PAR) figures to help me triangulate loan officers' account of the institution's overall performance. I expected him to decline the request but, instead, this conversation ensued:

FM:
Do you want "donor" figures or "real" figures? JS: What is the difference between the two? FM: The PAR figures we give donors are "dressed up" and for public relations and marketing. The "real" figures are internal, for our own consumption and represent the reality on the ground. JS: I want the "real" figures please. FM: No problem, I will email you the spreadsheet.
I went in not as an expert but to learn from the main actors about how microfinance works. For example, as I began shadowing the loan officers, I was struck by discrepancies between stated high loan repayment rates, as reported in the microfinance literature, and the "doing" of microfinance in the field. It became apparent that the majority of clients did not repay their loans from the income generated by their small enterprises but became entangled in multiple borrowing to service their loans. The reported high repayments were in some cases the result of loan officers "massaging" some figures to meet targets. My outsider positionality and a questioning approach toward what I later observed, heard, or experienced helped loan officers to be self-critical. The situation represented a learning opportunity missed in earlier work, which mainly focused on clients rather than examining the role of loan officers.

The Politics of Institution-Researching Microfinance Institutions
Entering into organizations can be difficult, more so if the research is perceived as intrusive but also the timing of it. Why was everyone uneasy about my presence? From the gatekeeper's perspective, I was carrying out a sensitive piece of research with the potential to expose a darker side to microfinance. Individuals and institutions therefore had to be selective in the ways in which they engaged with me and the type of data for which access could be granted. Research aiming at exploring how microfinance actually works together with employing methods of shadowing and observation was considered threatening, especially by employees in branch offices with closer client interaction. It only became clear later that difficulties in gaining access reflected underlying tensions and perspectives toward research. MFI T had just survived a near collapse, and donors had stepped in to forestall its complete failure. At the time of the fieldwork, this MFI was going through "a restructuring," which included change at the top management level and the sacking of several loan officers. There were reports of large-scale financial fraud in both MFIs (Siwale & Ritchie, 2013), leading to donors withholding further funding. A blame culture characterized all conversations, and consequently, relationships at all levels were characterized by fear and mistrust. I too was initially not trusted, especially by frontline employees-loan officers. In studying microfinance in action, I too became suspicious that something was not right. In both MFIs, a level of mistrust pervaded both vertical and horizontal relationships. There was "fear of scrutiny" (Payne et al., 1980, in Lee, 1993 that an "outsider" was about to expose what was really going on within, because I was visibly around. Consequently, detailed financial information was placed "off-limits" to keep the crisis out of the public domain. I was therefore researching these MFIs at very difficult times, with one emerging from and another on the verge of a crisis (which later engulfed it and led to its failure).

Some Reflections and Conclusion
As I have demonstrated in this article, researching donorfunded development projects in Africa can be frustrating and challenging even to those assumed to be "insiders" by virtue of being Africans or, in my case, Zambian. Prior to undertaking the research process itself, I had not fully reflected on how aspects of positionality such as being Zambian, educated, middle class, and a woman would affect access and relationship building. I was rather naïve in thinking that being Zambian would automatically grant me privileged "insider" status and therefore easy access. On the contrary, the process was harder than I had prepared for. It became clear through conducting this fieldwork that undertaking development research was problematic because of the donor dependency "culture," which conditions organizations to be more receptive to consultants whose research/reports create more opportunities for further funding. Typically, within the development sector in Zambia, donors are not only the "primary stakeholders" but also perceived as the "real investors." As a result, relationships are developed based on expected flows of money. This constructed image can be an added challenge for any other researcher negotiating access, regardless of ethnicity.
I have also shown that "insider/outsider" positionality when researching at "home" situates the researcher with epistemic advantages as well as challenges and unique dilemmas. However, this binary positionality can also provide an opportunity to question "indigenous" knowledge, a process able to enrich data (Al-Makhamreh & Lewando-Hundt, 2008). For example, I could not be sure whether research participants would have told different stories or shared other types of information with me had I been a foreign, White researcher. Neither was I sure that I fully represented how microfinance is experienced bottom-up by virtue of being Zambian. Being reflective, I have therefore been able to question whether I did introduce bias into my research as a result of being a Zambian researcher who initially did not question my own "Zambian-ness" as playing a role in how the research was to be conducted and understood by participants. Managing both positionalities and constantly reflecting on how they enable or constrain social space and interaction with research participants can be extremely challenging. Conducting research at home may not be a "comfy" affair. As a native in the field, a researcher at "home" almost always occupies a shifting position and identity construction can at best be conflicted. I have further highlighted the fact that native scholars "negotiate and experience different positionalities in the field stemming from their ethnic, linguistic, gendered, educational, and class backgrounds" (Jacobs-Huey, 2002, p. 799). The story also points out the power that informants can have over the "fieldwork life" of insider researchers (see also Turgo, 2012).
Prior to engaging in fieldwork for this research, I had hardly considered that my identity would be problematic and reconstructed in the course of the fieldwork. Reflecting on the fluidity of my identities as the research progressed therefore became a reality I had to deal with. I actively learnt to use these multiple identities in ways which helped facilitate the research access process. Other researchers, especially doctoral students interested in researching "back home," might want to reflect on this before entering the field. This discussion has shown how the organizational "politics" of those being researched and the nature of research can place restrictions on methodological approaches adopted and knowledge generated. Context and communities to be researched matter and, as was the case with this study, what was achievable was contingent on research participants' understanding and interpretation of the research objective and my perceived identity. I conclude by reiterating that the field is an uncertain place and a researcher always occupies a shifting position, while being mindful about ongoing speculation around their identity. Based on my field experiences, I have learnt that it helps to get a degree of confidence from the methodological literature, but that once in the field, one may have to be open to differences between expectations and practice. In particular, managing the plurality of identities, especially for those researching "back home," can be a challenge as there is neither a comfortable clear-cut insider nor an outsider position. You are constantly surrounded by ambiguity of status and yet seemingly at "home." I also point to the power that research participants can exert not only on identity construction and experienced positionalities in the field (Turgo, 2012), but more importantly, over access to the field and data collected. In addition, it is important to bear in mind that the research process is political even if you do not want it to be, thereby producing surprising uncertainties. The big lesson learnt is that there can be no "perfect" piece of advice related to doing research in one's native land, so plan to develop contingency strategies as you go along (Feldman, Bell, & Berger, 2003) and learn from your naivety.

Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding
The author(s) received no financial support for the research and/or authorship of this article.

Notes
1. Microfinance as a development strategy for poverty reduction has been endorsed by a number of international organizations such as the World Bank, United Nations, G20, G8, the Consultative Group to assist the Poor (CGAP), and other donor agencies. 2. In the main fieldwork, three branches from this particular microfinance institution (MFI) (T) were shadowed, while questionnaires were distributed to all the loan officers from seven branches in total. 3. While the term "gatekeeper" can be used in a number of different ways, gatekeepers within the research process are typically described as the individuals, groups, and organizations that act as intermediaries between researchers and participants and have the power to directly or indirectly facilitate or inhibit researchers' access (De Laine, 2000;Mandel, 2003).