Gaining access to the field in medical ethnography: Reflections on ethical,methodological,and structural challenges in the study of long-term care facilities

Abstract

The article extends the discussion on the challenges in gaining access to the field in medical ethnographic research, focusing on long-term care (LTC) facilities. Medical institutions have been documented to be difficult sites to access. The reference, however, is to the recruitment of patients as informants. The challenges of recruiting practitioners as informants have not been investigated at all. The article presents the key issues that emerged in the process of gaining social access at the sites of two LTC facilities as part of a study on care workers’ identities. The main obstacles encountered during the fieldwork were organizational constraints and negotiating control over the process of recruiting the lower occupational tier of care workers with gatekeepers. The article presents the coping strategies implemented to overcome the ethical and methodological obstacles: continually reassessing the consent and cooperation of participants and developing a rapport with nurse’s aides during interviews.

Keywords

ethnography issues in research methodology organization of health services

Introduction

Gaining access to participants is crucial for ethnographic research (Patton, 1990). In the process of gaining entry to research sites, using gatekeepers to less powerful individuals, and recruiting participants, ethnographers encounter many ethical and methodological challenges (Cunliffe and Alcadipani, 2016). Accounts of gaining social access in medical ethnographic studies in which the informants are members of the medical staff are scarce. Furthermore, studies investigating ethical and methodological challenges in studies of clinicians tend to overlook power structures in medical workplaces and their effect on the recruiting process. These studies tend to ignore the situational reality of health practitioners, and seldom consider the embedded hierarchies and strains that characterize the organizations, or the dilemmas arising from power differences in clinical settings (Toffoli and Rudge, 2006).

The present study examines the ethical and methodological challenges I encountered during fieldwork as part of a research project that explored care workers’ identities at geriatric facilities in Israel, focusing on nurses and nurse’s aides, the significant personnel that maintains long-term care facilities. The article is divided into three main sections. The first section discusses the literature on gaining access and the context of medical workplaces as organizations with complex power relations. This section provides the background for the research project, describes the research sites and the participants, and provides data about access and recruitment. The second section is based on the author’s ethnographic fieldwork, and it illustrates the key issues that emerged during the recruitment process for the project in question. The third section presents the conclusions and implications.

Gaining access

One of the challenging and time-consuming processes in ethnography is gaining access to social settings (Van Maanen, 2011). It involves securing entry to a particular site and ensuring that individuals, such as employees, whom the researcher has targeted as informants, agree to participate and to cooperate. This process is usually regarded as a linear and instrumental task, consisting of two steps: (a) the initial entry to a social field, granted by the gatekeepers, which is a condition for step two, and (b) collecting data with the cooperation of the informants (Shenton and Hayter, 2004). Gatekeepers are involved when the researcher cannot approach participants directly, and their cooperation is crucial at impermeable sites and organizations (Mikecz, 2012), and when access is needed to populations that are unreachable otherwise (Lee, 1993).

Gaining access to low-tier informants in organizational settings has several ethical and methodological aspects that can influence the study. First, initial physical entry to organizations is usually achieved through high-ranking officials of the organization (Hammersley and Atkinson, 2007). In formal organizations, key personnel can grant or withhold permission for access altogether (ibid). At health-care, education, welfare, and other public institutions, the imperative of going through ethical review boards forces researchers to engage with top-level gatekeepers and become dependent on them (Herdman, 2000). Additionally, researchers may need to negotiate access with different influential gatekeepers at multiple entry points to the site, including strategic managers, operational managers, and informal influential figures among the research population (Hammersley and Atkinson, 2007). Gatekeepers who are managers are usually the only channel of communication to low-tier participants and can easily compromise ethical restrictions, such as confidentiality and consent, in many ways (Cunliffe and Alcadipani, 2016).

Qualitative researchers’ problem in gaining access to health settings has been documented before (Herdman, 2000; Zaman, 2008). Initial entry to health settings is extremely difficult for outsiders because of formal and rigid ethical restrictions (Boulton and Parker, 2007). Nevertheless, ethnographic research at long-term care (LTC) facilities has become prevalent (Allen, 2004), but the ethical and methodological challenges in gaining access and recruiting nurses or nurse’s aides as participants are seldom reported (Pope, 2005; Webb, 2002).

Organizational hierarchy at long-term care facilities

Power structure at healthcare facilities is manifest in the division of labor and in the hierarchical relations between practitioners, which are conceptualized by their occupational status (Abbott, 1988; Freidson, 1970). The practice of medicine at institutions is characterized by a system of role relations between various occupations based on patterns of subordination and close supervision (ibid). At geriatric and LTC facilities, where there is minimal physician presence, the professional nursing ideology is hegemonic (Kane, 1990). In the organizational structure of these facilities, different levels of staff have clearly defined roles and responsibilities (Jervis, 2002; Ostaszkiewicz et al., 2016). These facilities are usually described as highly hierarchical (Jervis, 2002), with interpersonal relations between registered nurses and nurse’s aides characterized as authoritarian (Beck et al., 2002). Registered nurses have the legal and ethical responsibility for the care of patients, whereas nurse’s aides usually perform most bodily tasks, and are tightly supervised by the nurses (Jervis, 2002; Yeatts and Seward 2000). Power relations between nurse’s aides and nurses are affected by the professional processes that nursing has undergone in the last two decades (Montayre and Montayre, 2017). Nurses extended their roles into domains that formerly were those of physicians, and they are required to perform onerous documentation and administrative functions (Ackroyd et al., 2007). Among other factors, these trends have contributed to the assignment of the “dirtier” tasks to nurses’ auxiliaries, such as nurse’s aides (Jervis, 2002).

Power dimensions vary between organizations in the degree and steepness of the hierarchy, and in the degree and pattern of segregation by gender and race (Acker, 2006). The occupational hierarchy in medicine reflects the social structure and ethnic stratification of society: women, migrants, and ethnic minorities are overrepresented in the lowest-paying occupations in healthcare, such as nurse’s aides and sanitary workers (Duffy, 2005) who perform undesired tasks that carry social stigma (Ashforth and Kreiner, 1999). Many care workers come from developing countries, producing what has become known as “global care chains” (Ehrenreich et al., 2003). In Israel, market reforms have privatized most social services for the elderly (Ajzenstadt and Rosenhek, 2000), exposing female service workers to the deskilling practices of the “contract state” (Benjamin, 2011).

Research and participants

The account presented in this article was derived from a research conducted at two LTC facilities in Israel, between January 2017 and April 2018. The research focused on two occupational groups, nurses and nurse’s aides, aiming to explore the work-related identities of nurses and nurse’s aides, including organizational and occupational identities (Dutton et al., 2010). Ethical approval was granted by the Ethics Committee of Bar-Ilan University and the institutional ethics committees of the two facilities. In Israel, institutional research ethics committees are supervised by the Ministry of Health.

I collected data at two sites: a ward in a nursing home for elderly patients and a geriatric ward in a hospital. Both wards provided LTC for patients over 75 years old. The wards had permanent work routines and sets of tasks. Morning and afternoon shifts consisted of two nurses and four–five nurse’s aides. Nurses provided medical care and supervised nurse’s aides who provided care and bodily services. The night shift was exceptional because the staff was limited to one nurse and one nurse’s aide, who worked together changing posture and diapers. Nurse’s aides worked in pairs, according to a prearranged duty roster, which assigned each pair to specific rooms, depending on the services the patients needed.

The staff in the wards was representative of the highly diverse Israeli workforce. Significant groups in the Israeli medical system, in addition to long-time Jewish residents of the country, are Jewish immigrants from the former Soviet Union (FSU), Israeli Arabs, and foreign workers. The healthcare system is one of the largest employers in the country, and it has traditionally attracted immigrants and minorities to fill positions at all levels of the professional hierarchy. Most registered nurses were Jewish women who immigrated from the FSU. Many nurses from the FSU started their careers in Israeli health facilities as practical nurses or nurse’s aides, and completed their nursing studies while working (Remennick, 2001). Many others, disadvantaged immigrant women, especially older ones, did not have the resources to improve their occupational status and continued to work as nurse’s aides.

To the best of my knowledge, there is no available accurate numerical information concerning nurse’s aides employed in LTC institutions in Israel. Demographic forecasts predict that by 2030 15% of the Israeli population will exceed the age of 65 (Brodesky et al., 2018), which suggests that there will be a growing demand for institutionalized care in Israel and a growing care workforce. As of today, the geriatric sector has difficulty recruiting staff among native Israelis, at all levels of occupational hierarchy, because of its low prestige (Haron et al., 2013). The sector has attracted undocumented workers from African, East European, and South American countries, who are employed under poor occupational terms through contractors. Every year, the Israeli government issues permits to selected labor contractors to employ a fixed number of foreign workers who have no work permits (Harper and Zubida, 2010).

Only a few of the nurses were born in Israel and spoke Hebrew natively; the vast majority spoke Hebrew as a second language. The majority of nurse’s aides from FSU were middle-aged Jewish women. African and South Asian nurse’s aides were young women and men, who were either foreign workers or asylum seekers. Most of them spoke basic Hebrew that allowed them to get by at LTC workplaces, although some spoke fluently.

Access and recruitment

Gaining access involved three levels of hierarchy in the nursing home and four levels in the hospital. At the initial stages of the project, I approached by email four heads of geriatric departments at large hospitals in Israel and seven managers of private nursing homes. I presented myself as a medical sociologist who studies care workers’ identities, and emphasized my background as a physiotherapist.

My first contact at the hospital was with an executive manager, who gave me permission to initiate the bureaucratic and institutional ethical process. After I received the necessary approval, the second level was the professional tier, the head physician who introduced me to the staff. The head physician authorized interviews with four members of the paramedical staff and the chief of nursing, who became the dominant gatekeeper to most participant observations. The chief of nursing also recruited all the care workers I interviewed, who were the focus of the inquiry.

My first contact in the nursing home was the general manager, who was the initial formal gatekeeper of the establishment. The general manager introduced me to the mid-level personnel of the facility. I interviewed two mid-level managers, two members of the paramedical personnel, the head of nurse’s aides, and the chief of nursing. The last two became the main gatekeepers in the nursing home to participant observations at staff meetings and social events, and access to care workers.

I offered to the gate keepers to volunteer at the facilities in order to absorb as much as possible the organizational culture and to build a rapport with staff members. The managerial rank in the hospital did not approve my volunteering in the ward; however, in the nursing home I met weekly with two women who were hospitalized to chat and discuss current events, for a period of 14 months. We usually met in their rooms, occasionally in the main halls. Participant observations in the nursing home included celebrations of religious and national holidays, and five staff meetings. Participant observations in the hospital included attending irregularly the weekly staff meeting (once or twice a month) and two social gatherings of staff members outdoors. At both sites, I spent many hours waiting to interview staff members in the corridors and open spaces, which provided me with the opportunity to observe staff members interacting and to have short informal talks. Professional staff meetings were held in the mornings, and most social events were in the afternoon. Most interviews were scheduled for the morning or mid-day, but I interviewed ten staff members, from both sites, during evening shifts because I wanted to understand the working atmosphere outside the rush hours and to have the opportunity to engage with staff members.

The original recruitment plan was to invite nurse’s aides and nurses to participate as interviewees, and recruit using the snowball sampling method (Noy, 2008). Voluntary participation is consistent with ethical considerations and enhances rigor in the research, based on the assumption that voluntary compliance produces authentic data. During the first staff meetings I attended, the gatekeepers gave me the opportunity to present the study and address ethical issues regarding interviews with potential participants. On these occasions I stressed the voluntary basis of the participation, the fact that I was an independent researcher, bound by ethical restrictions, and that the confidentiality of participants would be protected at all cost.

I interviewed almost the entire nursing staff of the hospital ward: 18 in total (eight nurses and ten nurse’s aides), except for one nurse’s aide who refused to participate. Two interviews were excluded from the study sample for reasons on which I elaborate in the subsection “arranged” and “surprise” interviews. The ward in the nursing home comprised a team of ten nurses and approximately 16 nurse’s aides. Of these, I interviewed ten nurses and eight nurse’s aides, following the principles of theoretical sampling and saturation, which aims to expand the sample as long as it contributes to the theoretical and conceptual insights of the research (Glaser and Strauss, 2017). Two nurse’s aides in the nursing-home refused to be interviewed.

At the hospital, all interviews took place in the office of the chief of nursing, which was a private place with no surveillance, and where there were no interruptions. The chief of nursing met me in advance and gave me the keys to her office, or I would pick up the keys from the ward secretary. At the nursing home, I conducted the interviews with the lower-rank workers in various locations, mostly in open and public places such as the physiotherapy hall, vacant rooms, the public porch, or the medication storage room. These spaces were covered by camera surveillance. The conversations were held privately because these spaces were deserted at nap times, but during most of the interviews we needed to stop once or twice if a coworker passed by.

I documented every occasion on which I visited the wards, after meeting the gatekeepers, after staff meetings and social events, and after personal meetings as a volunteer. During social events, I had the opportunity to get to know staff members, understand the relations between coworkers, the power relations between various actors, and other organizational aspects that relate to their identities as care workers. The volunteer work permitted me to meet staff members without the mediation of the gatekeepers, to form micro-relationships, and to be present in the wards during quiet periods. I accumulated a large quantity of notes that I labeled “before an interview with X” and “after an interview with X.” These notes concerned my observations while waiting for interviewees in the wards, as well as the contextual aspects of interviews, and other impressions I gathered during interviews. I attached these notes to the transcripts of the interviews and analyzed them as complementary data, providing context to the iterative process of reading and rereading the data. Reading and analyzing these observations contributed to the research project and to the emergence of the salient themes that reflected the process of recruitment. The names of all informants have been changed to ensure confidentiality.

Key issues in the process of gaining social access

Organizational constraints in interviewing nurse’s aides

Nurse’s aides, and to some degree, nurses, work under extremely harsh conditions, which entail strenuous physical tasks, tight schedules, overwork, and scarce free time. Nurse’s aides perform difficult physical tasks, such as transferring patients from bed to wheelchair. They are perpetually challenged by assaults from patients suffering from dementia (Morgan et al., 2002; Pitfield et al., 2011). Moreover, nurses and nurse’s aides work in shifts, which means that they need to adapt to working nights and weekends.

In many instances during my fieldwork, I was told that the limited time that nurse’s aides could invest in patients was the most valuable resource at LTC facilities. The time of the nurse’s aides was scarce, and everybody needed it: management faced persistent shortages in nurse’s aides, and the patients were in constant need of their time. In the mornings and after the patients’ afternoon nap, nurse’s aides were often observed rushing from one patient to another, bells ringing in patients’ room, with patients asking to be taken out of bed. Often, nurse’s aides had arguments between themselves or in their interactions with patients, and the overall atmosphere was stressed. At these occasions, nurse’s aides were especially inaccessible and could not spare a moment to talk with me. At times, nurses interrupted them during their breaks and they had to leave in the middle of their meal, which caused anger and frustration.

Arranging an interview with participants was challenging because of time constraints, among other reasons. Time is central in medical discourse as well as in healthcare organizations, which are based on the rationale of the “clock-time” (Deery, 2008; Simonds, 2002). Healthcare workers usually work under stressful timetables, which is evident in the day-to-day reality of care workers at long-term care facilities undertaking strictly scheduled body work as bathing and feeding. These are particularly stressful because the duration of bodily services is unpredictable, making it difficult to adapt to strict schedules (Davies, 1994; Dyer et al., 2008). The gatekeepers, who were extremely busy too, had difficulty sparing a nurse or a nurse’s aide. The time slots that gatekeepers allotted to nurse’s aides for the interview were usually short, about half an hour around 10 a.m., squeezed between two busy morning and midday periods. Because the time slots for a few interviews were very short, and interviewees appeared to be a little edgy to rush back to work, it was difficult to conduct a quiet conversation that could raise deep insights. Therefore, I tried to meet interviewees during nap time or evening shifts.

Many of my fieldwork notes consisted of observations I made while waiting for interviewees. There was usually an unexpected delay because of staff shortages or a patient admission that required their full attention. Often, gatekeepers scheduled nurse’s aides to be interviewed at the expense of their meal breaks, a predicament I needed to address.

Another organizational or occupational feature that may have influenced my ability to communicate with nurse’s aides in a casual and informal way was their lack of autonomy during work hours. Nurses were monitoring their whereabouts at all times through cameras located in public spaces, such as corridors and main halls. Nurse’s aides had no privacy during breaks either. Nurse’s aides, similar to other service workers, usually work in supervised spaces and do not have a private space, as does the managerial staff, which works in private offices.

Negotiating control with the gatekeepers

The gatekeepers were the main and almost exclusive sources for recruiting participants. Below I describe the strategies they used to exercise power over the sampling and data collection processes for the study. I rely on examples from my experience negotiating with gatekeepers in the process of recruitment of nurse’s aides, and to a lesser extent of nurses, and I discuss the ethical and methodological implications of this process.

The head of nurse’s aides in the nursing home was in charge of the employment, training, and supervision of the nurse’s aides at the facility. Because turnover was high, she was constantly busy with the nurse’s aides’ schedule, which was changing by the hour, and although she was cooperative, she was somewhat inaccessible. She connected me with the nurse’s aides whom I interviewed. The other two gatekeepers, the chief of nursing at both facilities, were extremely compliant with my efforts to recruit staff members for the study. Our similar background made it easier for us to communicate and build a rapport. Our ages were similar, between 35 and 45 years old, our professions were health-related, and we shared a similar academic background (M.A. degrees). They also sympathized with my efforts as a researcher because they had conducted research in the past.

“Arranged” and “surprise” interviews

At the initial stages of the fieldwork I did not estimate correctly the total control the gatekeepers would exercise over the recruitment process and the final sample of the study, and I accepted the suggested participants of their choice. Gatekeepers can provide researchers with a narrow selection of respondents (Fielding, 2001). “Arranged” or “exemplary” interviews with suggested, exceptionally collaborative participants were in contrast with the overall unfriendliness and hostility of most staff members. I believe that a total of four interviewees (one nurse at the hospital, and two nurses and one nurse’s aide at the nursing home) were hand-picked (arranged) at the beginning of the data collection process. Later, I learned that they were relatively senior staff members, associated with the gatekeepers. These interviewees were keen on making a good impression on me, praised their co-workers and managers, did not raise any problematic issues, exhibited exemplary organizational loyalty, and were at times inauthentic. These interviews were usually flat, in the sense that they contained many of the buzzwords that dominate the hegemonic discourses of medical care, and did not present any data that could be thought of as a thick description (Geertz, 1973). One slogan that was frequently repeated by interviewees was “teamwork,” a highly valued term in medical care, which seemed out of place in the context of the fragmented distribution of work and delegation of care tasks in the wards. Although I sensed that arranged interviews were flat, my first impression was confirmed only as data collection developed and the variation in interviews increased.

The inauthenticity of several interviewees was revealed in explicit expressions such as “What else do you need me to say?” (Alina, practical nurse), or “Don’t worry, I know exactly what to say” (Arin, practical nurse). At the end of the interview, Arin related a few stories, the details of which she presented “off the record.” The details, which concerned patients and staff members, were rather judgmental and revealed conflicted relations in the nursing home.

There is the possibility that because gatekeepers sympathized with my efforts to recruit interviewees, they selected interviewees they thought would be more cooperative. Often, following an “arranged” interview, I was asked by the gatekeeper who had arranged the encounter: “How was it?” or “How was she?” This may or may not have been an innocent question. I recognized that the gatekeepers had my best interest at heart, and wanted to advance my data collection project. But occasionally I interpreted these questions as attempts to obtain some kind of information from me about the interviews. Although developing a rapport with gatekeepers can facilitate data collection and achieving the research goals, it may also influence data collection and introduce a bias in the results (Cunliffe and Alcadipani, 2016). In the current study, the fact that I developed a rapport with the managerial ranks may have naturally been a factor in the liberty they took, in the first stages of the study, to ask questions about the content of the interviews, which could have compromised the participants’ confidentiality. However, I overlooked these remarks because I was committed to protect the confidentiality of my interviewees, and had no intention of sharing the data, transcripts, or notes with anyone other than my academic supervisor. In time, the gatekeepers’ remarks stopped. Furthermore, as the study progressed, I was more free to come and go at the facilities, my encounters with the gatekeepers decreased, and they had fewer opportunities to question me after interviews.

Another phenomenon that reinforced my impression that interviewees did not participate entirely voluntarily or that they may have been coerced by the gatekeepers to some degree was that of “surprise interviews.” Because gatekeepers would not give me the phone numbers of their staff members, I had to schedule interviews through them, on the phone or after a participant observation. I was always assured that the workers themselves agreed to participate voluntarily. But when I arrived at the scheduled hour, on four occasions the interviewee was not aware of our meeting at all. Among these, two nurses (one of each facility) declined to be recorded and were suspicious of my taking notes during the interviews. Although all participants agreed orally to the interview and signed informed consent forms, the behavior and low cooperation of these participants made it clear that they were coerced to some degree to participate. The data these informants provided was minimal, and they answered questions laconically, without starting any thread of a personal perspective or association that was not directly related to the question asked, as interviewees often do in the course of in-depth interviews. Two hospital nurse’s-aides (among the four “surprise interviews”) seemed to be afraid of me during the interview. Because my efforts to reassure them failed, I cut the interviews short and these interviews were excluded from the study sample.

These interactions caused me great discomfort regarding the free and voluntary consent of the respondents to participate in the study. I believe, however, that the gatekeepers had no intention to sabotage my efforts to ensure the participants’ voluntary consent, and the “surprise” interviews were mostly the result of the gatekeepers’ lack of time to set up the meetings with me according to the rules I devised.

Blocking certain interviewees and participant observations

In their account of problems gaining access to caregivers, Groger et al. (1999) stated that the reasons for not interviewing some of the people had to do with gatekeeper bias and with direct refusal. In the present research, I cannot claim such well-defined reasons for not being able to interview certain workers, but I can offer some insights, based on the process of data collection, regarding possible reasons why gatekeepers may have prevented me from interviewing potential participants and from conducting certain observations.

The gatekeepers employed various tactics to prevent me from approaching two potential participants (a nurse at the hospital and a nurse’s aide at the nursing home) and from attending some staff meetings at the nursing-home. For example, Adele, the chief of nursing at the hospital, noted that certain nurses and nurse’s aides were “hard nuts to crack,” meaning that they would not cooperate with me during the interview and would not be forthcoming. Adele was also under the impression that interviewing nurses and nurse’s aides from the FSU was a difficult mission because of what she called the “Soviet paranoia” of these staff members. She repeatedly prevented me from interviewing Lena, a registered nurse. Whenever I asked Adele to invite Lena to participate, she always answered that “she wouldn’t be a good interviewee.” I had observed Lena at staff meetings and was aware of her feisty attitude. Despite the reputation for introversion of workers from the FSU, Lena was blunt and did not spare her criticism of her colleagues and of the administrative staff. I interpreted Adele’s resistance as an attempt to prevent me from hearing criticism of the ward. Naturally, Lena’s remarks aroused my interest in her as an interviewee, but she was greatly hostile and did not want to even listen to me (she was what I call an “unapproachable interviewee”). Finally, when Adele asked her, she agreed to participate, and as expected, she was fluent and expressive in the interview.

At the nursing home, there was another potential participant that the gatekeeper did not want me to interview. Joseph was an Eritrean work immigrant in his thirties who worked as a nurse’s aide. I asked to interview him several times to diversify my sample, but the gatekeeper always evaded me. I found out from other interviewees that he was considered unreliable and had many clashes with the management, but they kept hiring him because they had staff shortages. Maybe the gatekeeper thought he would speak ill of the management.

As I collected information about the socialization procedures of nurse’s aides at the nursing home, I became aware of periodic orientation meetings for them. Officially, these were training sessions aimed at teaching how to enable basic activities of daily living (ADL). The staff held these meetings on a certain day of the week, but only “when necessary,” that is, when new workers had been hired or when the senior staff felt that the nurse’s aides needed to “refresh their knowledge of the regulations.” The head of nurse’s aides elaborated that occasionally the staff did not pay attention to basic requirements, such as the water temperature, when they bathed the patients. I was interested in attending these trainings, which were scheduled only two days in advance. I frequently asked the managerial staff whether there was an upcoming training. Yet, despite our ongoing communication, I always missed the trainings or found out about them in retrospect, so that it became clear that I was not welcome as an observer. Later I discovered that there were cases of conflictual interactions between nurse’s aides and patients, and of misconduct on the part of the nurse’s aides, which were addressed at these trainings.

Coping strategy: Reassessing consent and cooperation

The first pressing ethical problem I had to address during data collection was consent: the potential interference of the gatekeepers in the participants’ decision to freely agree to take part in the research. Gatekeepers have the power to determine how participants are informed about the study, which in turn influences their willingness to participate. Although I participated at staff meetings and was introduced to most of the staff, I had no control over the image the gatekeepers painted of me, and could not know how the interview was presented to them. Therefore, I was not able to assess correctly the possible effect of the superior-subordinate relations on the respondent’s participation in the interview. Because I was dependent on the gatekeepers to schedule participants, I stressed repeatedly throughout our prolonged engagement that I did not want the employees to feel coerced to participate in the interviews, that I was bound by ethical constraints, and that the research would benefit more from interviews given voluntarily.

Tyldum (2012) argued that social scientists overestimated the goodwill of participants, and overlooked or underestimated possible pressures exerted on their respondents.

The collaboration of participants does not necessarily provide researchers with the certainty that they are cooperating willingly. For example, Meyers’ (1993) experience with conducting action research in a healthcare setting caused her to reflect on ethical issues pertaining to the collaborative relations formed in this type of research, and came to the conclusion that even the egalitarian paradigm of action research cannot blur the asymmetry between the researcher and participants.

The ideal standard ethics review process does not always correspond to the practical ethics that researchers must observe in the field (Franklin et al., 2012), and the standardized rules for maintaining informed consent and confidentiality fail to recognize the ongoing nature of ethics in research (Hemmings, 2006). At times, researchers need to tailor their research design, resorting to a “situated ethics” approach (Perez, 2019). Miller and Bell (2002) proposed to conceptualize informants’ consent as an ongoing process, in which consent is negotiated and reassessed. In the current study, I aspired to reassess and enhance consent at numerous points after it had been given to the gatekeepers, so that I could regain control over observance of the ethical standards. First, when I met interviewees who were scheduled to be interviewed without prior notice (i.e. “surprise interviews”), I would “start over.” In other words, after I approached them and had a moment alone with them, I would explain about the research and make sure that the interviewees understood that they were not committed to participate but were indeed willing to take part in the study. Second, I explained about their rights as informants and about anonymity, I offered them the option not to be recorded, and I reassured them that they could speak freely. I told them explicitly that I had no connection with their superiors, and I also offered the informants several exit points at which I could reconfirm their consent.

Whenever I sensed an act that suggested coercion in the smallest degree or even a gesture that betrayed a feeling of discomfort of a participant, I explained to the participants about their right not to be interviewed at all, and that there will be no negative consequences for them if they refused. Any sign of possible coercion would raise my awareness of the participants’ reactions. The two nurses who declined to be recorded were naturally granted that option, and in the course of these interviews I used all the techniques noted above (constant reassessment, reassurance of confidentiality and anonymity, and providing exit points). I used the same techniques with the two nurses whom I identified as fearful of the interview situation, but when I failed to reassure them, I cut the interviews short in a friendly way. The data provided by the two interviewees who had originally declined to be recorded were minimal and did not contribute to the research project, but the documentation of these incidents in my fieldwork notes was valuable for advancing my understanding of the contextual features of the researched sites and of the depth of hierarchy between staff members. Moreover, these incidents indicated that I should use the fieldwork notes concerning these interviews. Because the authenticity of data was in question, I used the multiple research methods for triangulation to increase rigor (Patton, 1999).

Coping strategy: Developing a rapport during interviews

Establishing a rapport with the participants is crucial for obtaining their cooperation in qualitative studies and building trust with informants facilitates data collection (Lincoln et al., 1985). Usually, the process of developing a relationship during interactions is depicted as the precursor to the recruitment of participants. In the current study, however, traditional rapport-building techniques did little to gain the cooperation of nurse’s aides, and I needed to adopt other measures. I wished to establish a rapport with interviewees by immersing in the field and becoming a familiar face to the care workers. During my initial encounters with the staff, however, I sensed a measure of alertness on the part of some employees, as one would expect (Lincoln et al., 1985). Bogdan and Biklen (2007: 90) noted that “feeling uncomfortable is part of doing this kind of work.” As time passed, my inability to establish a rapport or even basic relations with some of the staff members became evident: some of them refused to reply when I approached them and avoided making eye contact. Because the nurse’s aides were suspicious and apprehensive, the opportunity to invite them directly to participate in the study was limited.

The literature has documented the fact that populations from the developing world, minorities, and socially deprived communities are difficult to recruit (Sanghera and Thapar-Björkert, 2008; Sixsmith et al., 2003). Similarly, social groups of low social capital are more likely to refuse to participate in surveys (Groves et al., 2000). But the hurdles I faced trying to recruit participants made me reflect on my part in this challenge, that is, on my identity and position. Contrary to the similarities I shared with mid-level management and the paramedical staff, professionally and agewise, the dissimilarities between me and the nurse’s aides were blatant: I am white, Jewish, native Israeli, and belong to the privileged Ashkenazi ethno-national group in Israeli society. The ethno-nationalities of the nurse’s aides were FSU, African, and South Asian, and they spoke poor Hebrew. Often during participant observations I did not understand the interactions between staff members because they spoke Russian. Although workers switched to Hebrew at the presence of a non-FSU worker, they apparently did not regard me as an insider, and often I was ignored.

Because it was difficult to build a rapport during participant observations, I used the interview itself as a platform for building trust with interviewees. I employed two main mechanisms. First, I repeatedly expressed during interviews my appreciation of the care workers’ difficult work. Second, I offered interviewees empathy for their harsh experiences, and when they shared their difficulties, I offered a non-judgmental sympathetic ear, which some of them apparently needed. In some cases, the interviews became the starting point of a dialogue that continued in our subsequent encounters. These mechanisms altered the way in which some of the interviewees related to me: from complete ignorance to approaching me whenever we met and venting about the difficulties they experienced at work. The following excerpts exemplify how I applied these mechanisms.

One example was the relationship I formed with Jana, a nurse’s aide from FSU. Before her interview, Jana was one of the most unapproachable workers in the hospital ward. She had a constant expression of anger on her face, never answered my greetings, and if our eyes met, she never answered my smile or nod, retaining her poker face. The interview, however, completely changed her attitude toward me. At the interview, she was not only friendly but also talkative. Every time I met her afterwards, she had something to share with me, on her own initiative, despite a language barrier that existed between us. It seemed as if after the interview she perceived me as a person with whom she can share her emotions. Our relations led to the cooperation of two other nurse’s aides, with whom she was close. I became somewhat more of an insider, and participants became more approachable, answered questions, and initiated communication. This allowed me to be present at more unregulated situations in the wards, which were informative in themselves.

Another example was the relationship I formed with Tatiana, a nurse. Tatiana was perhaps too cooperative, and the experiences reported in her initial interview lacked richness. She seemed preoccupied with making a positive impression. Although Tatiana was laconic during the interview, I continued socializing with her and we formed a friendly relationship. Following the interview with Tatiana, there were two instances in which I almost felt that I was an insider. On one occasion, I was present at a discussion between nurse’s aides and nurses, which did not prevent her from venting her feelings about a demanding family member. The nurses and nurse’s aides accepted me as an integral participant in the interaction, and Tatiana was looking at the other workers and at me while expressing her anger. On that occasion, she expressed freely her frustration over the experience of her demanding role.

Conclusion and implications

Long-term care facilities for the elderly are becoming increasingly dominant in healthcare provision, as Western populations continue to age. This paper concerns the key issues that emerged in the process of gaining access to long-term care organizations, and offers my reflections about them as a researcher. The processes involved in the recruitment of low-tier participants in research are always contextual and situated, but researchers can derive some insights that relate to their own research sites from past experiences of fellow researchers.

I conclude by presenting the coping strategies and solutions I used in the course of my fieldwork, so that in the future researchers can benefit from my experiences. First, the findings have methodological implications for the design of research in hierarchical workplaces with tight supervision of subordinate tiers and suggest the necessity of adopting flexibility in the methodological choices undertaken during the process of data collection. Given the inaccessibility of potential participants, I became more reflexive and aware of my position and of the obstacles to developing a relationship with my informants. Facing hostility and uncooperative responses, I stopped to rely on collective social events and informal interactions to establish trust, and I broke through my isolation in the field by becoming skilled in building a relationship during the one on one interview itself, which helped me become more of an insider, and facilitated data collection.

Second, the findings indicate that researchers be highly suspicious of possible coercion by gatekeepers of their subordinates, and use an ongoing strategy of reassessment of participants’ consent. As a preventive measure, I stressed repeatedly in front of the gatekeepers the ethical norms of voluntary participation and confidentiality. I also reassessed the participants’ consent over and over before and during interviews.

Third, the findings have implications for social science researchers’ ethical training, which is usually standardized and generalized, and fails to recognize the complexity of the research process. Moreover, students are trained to comply with ethical standards, but often have no fieldwork experience. Therefore, ethical training should address the grey areas of ethical conduct, the politics and micro-practices of access, and specifically issues of subordination and coercion. Training should also address reflexivity and peer consultation, and encourage understanding the process of data collection as data in itself.

Footnotes

Declaration of conflicting interests

The author declares that they have no competing interest.

Funding

The author received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Neta Roitenberg

Author biography

Neta Roitenberg, PhD, is a doctoral student in the Sociology Department at Bar-Ilan University. Her research interests include work identity construction processes of health practitioners who provide treatment to contested illnesses, and of health practitioners at long-term geriatric care facilities. She uses both qualitative methodological approaches and quantitative measures to investigate the perceptions and attitudes of health practitioners.

References

Abbott

(1988) The System of Professions: An Essay on the Division of Expert Labor. Chicago: University of Chicago Press.

Acker

(2006) Inequality regimes: Gender, class, and race in organizations. Gender & Society 20(4): 441–464.

Ackroyd

Kirkpatrick

Walker

(2007) Public management reform in the UK and its consequences for professional organization: A comparative analysis. Public Administration 85(1): 9–26

Ajzenstadt

Rosenhek

(2000) Privatisation and new modes of state intervention: The long-term care programme in Israel. Journal of Social Policy 29(2): 247–262.

Allen

(2004) Ethnomethodological insights into insider–outsider relationships in nursing ethnographies of healthcare settings. Nursing Inquiry 11(1): 14–24.

Ashforth

Kreiner

(1999) ‘How can you do it?’: Dirty work and the challenge of constructing a positive identity. Academy of Management Review 24(3): 413–434.

Beck

Doan

Cody

(2002) Nursing assistants as providers of mental health care in nursing homes. Generations 26(1): 66.

Benjamin

(2011) Gender outcomes of labor market policy in Israel. Equality, Diversity and Inclusion: An International Journal 30(5):394–408.

Bogdan

Biklen

(2007) Qualitative Research for Education: An Introduction to Theories and Methods, 5th edn. Boston, MA: Allyn & Bacon.

10.

Boulton

Parker

(2007) Informed consent in a changing environment. Social Science & Medicine 65(11): 2187–2198.

11.

Brodesky

Snor

Be’er

, et al. (2018) The Elderly in Israel, 2017. Jerusalem: Myers-JDC-Brookdale.

12.

Cunliffe

Alcadipani

(2016) The politics of access in fieldwork: Immersion, backstage dramas, and deception. Organizational Research Methods 19(4): 535–561.

13.

Davies

(1994) The tensions between process time and clock time in care-work: The example of day nurseries. Time & Society 3(3): 277–303.

14.

Deery

(2008) The tyranny of time: Tensions between relational and clock time in community-based midwifery. Social Theory & Health 6(4): 342–363.

15.

Duffy

(2005) Reproducing labor inequalities: Challenges for feminists conceptualizing care at the intersections of gender, race, and class. Gender & Society 19(1): 66–82.

16.

Dutton

Roberts

Bednar

(2010) Pathways for positive identity construction at work: Four types of positive identity and the building of social resources. Academy of Management Review 35(2): 265–293.

17.

Dyer

McDowell

Batnitzky

(2008) Emotional labour/body work: The caring labours of migrants in the UK’s National Health Service. Geoforum 39(6): 2030–2038.

18.

Ehrenreich

Hochschild

Kay

(eds.) (2003) Global Woman: Nannies, Maids, and Sex Workers in the New Economy. New York, NY: Metropolitan Press.

19.

Fielding

(2001) Ethnography. In: Gilbert

(ed.) Researching Social Life, 2nd edn. London: Sage Publications, pp.145–164.

20.

Franklin

Rowland

Fox

, et al. (2012) Research ethics in accessing hospital staff and securing informed consent. Qualitative Health Research 22(12): 1727–1738.

21.

Freidson

(1970) Profession of Medicine. New York, NY: Dodd, Mead & Co.

22.

Geertz

(1973) The Interpretation of Cultures. New York, NY: Basic Books.

23.

Glaser

Strauss

(2017) Discovery of Grounded Theory: Strategies for Qualitative Research. London: Routledge.

24.

Groger

Mayberry

Straker

(1999) What we didn’t learn because of who would not talk to us. Qualitative Health Research 9(6): 829–835.

25.

Groves

Singer

Corning

(2000) Leverage-saliency theory of survey participation: Description and an illustration. Public Opinion Quarterly 64(3): 299–308.

26.

Hammersley

Atkinson

(2007) Ethnography: Principles in Practice, 3rd edn. London: Routledge.

27.

Haron

Levy

Albagli

, et al. (2013) Why do nursing students not want to work in geriatric care? A national questionnaire survey. International Journal of Nursing Studies 50(11): 1558–1565.

28.

Harper

Zubida

(2010) Making room at the table: Incorporation of foreign workers in Israel. Policy and Society 29(4): 371–383.

29.

Hemmings

(2006) Great ethical divides: Bridging the gap between institutional review boards and researchers. Educational Researcher 35(4): 12–18.

30.

Herdman

(2000) Reflections on “making somebody angry.” Qualitative Health Research 10(5): 691–702.

31.

Jervis

(2002) Working in and around the “chain of command”: Power relations among nursing staff in an urban nursing home. Nursing Inquiry 9(1): 12–23.

32.

Kane

(1990) Everyday life in nursing homes: ‘The way things are’. In: Kane

Caplan

(eds.) Everyday ethics: Resolving dilemmas in nursing home life. New York: Springer, pp. 3–20.

33.

Lee

(1993) Doing Research on Sensitive Topics. Newbury Park, CA: Sage.

34.

Lincoln

Guba

Pilotta

(1985) Naturalistic inquiry. International Journal of Intercultural Relations 9(4): 438–439.

35.

Meyer

(1993) New paradigm research in practice: The trials and tribulations of action research. Journal of Advanced Nursing 18(7): 1066–1072.

36.

Mikecz

(2012) Interviewing elites: Addressing methodological issues. Qualitative inquiry 18(6): 482–493.

37.

Miller

Bell

(2002) Consenting to what? Issues of access, gate-keeping and “informed” consent. In: Mauthner

Birch

Jessop

Miller

(eds.) Ethics in Qualitative Research. London: Sage, pp.53–69.

38.

Montayre

(2017) Nursing work in long-term care: An integrative review. Journal of Gerontological Nursing 43(11): 41–49.

39.

Morgan

Semchuk

Stewart

, et al. (2002) Job strain among staff of rural nursing homes: A comparison of nurses, aides, and activity workers. Journal of Nursing Administration 32: 152–161.

40.

Noy

(2008) Sampling knowledge: The hermeneutics of snowball sampling in qualitative research. International Journal of Social Research Methodology 11(4): 327–344.

41.

Ostaszkiewicz

O’Connell

Dunning

(2016) “We just do the dirty work”: Dealing with incontinence, courtesy stigma and the low occupational status of carework in long-term aged care facilities. Journal of Clinical Nursing 25(17–18): 2528–2541.

42.

Patton

(1990) Qualitative Evaluation and Research Methods. Thousand Oaks, CA: Sage Publications, Inc.

43.

Patton

(1999) Enhancing the quality and credibility of qualitative analysis. Health Services Research 34(5 Pt 2): 1189.

44.

Perez

(2019) In support of situated ethics: Ways of building trust with stigmatised “waste pickers” in Cape Town. Qualitative Research 19(2): 148–163.

45.

Pitfield

Shahriyarmolki

Livingston

(2011) A systematic review of stress in staff caring for people with dementia living in 24-hour care settings. International Psychogeriatrics, 23(1): 4–9.

46.

Pope

(2005) Conducting ethnography in medical settings. Medical Education 39(12): 1180–1187.

47.

Remennick

(2001) “My life is one big nursing home.” Russian immigrant women in Israel speak about double caregiver stress. Women’s Studies International Forum 24(6): 685–700.

48.

Sanghera

Thapar-Björkert

(2008) Methodological dilemmas: Gatekeepers and positionality in Bradford. Ethnic & Racial Studies 31(3): 543–562.

49.

Shenton

Hayter

(2004) Strategies for gaining access to organisations and informants in qualitative studies. Education for Information 22(3–4): 223–231.

50.

Simonds

(2002) Watching the clock: Keeping time during pregnancy, birth, and postpartum experiences. Social Science & Medicine 55(4): 559–570.

51.

Sixsmith

Boneham

Goldring

(2003) Accessing the community: Gaining insider perspectives from the outside. Qualitative Health Research 13(4): 578–589.

52.

Toffoli

Rudge

(2006) Organizational predicaments: Ethical conditions for nursing research. Journal of Advanced Nursing 56(6): 600–606.

53.

Tyldum

(2012) Ethics or access? Balancing informed consent against the application of institutional, economic or emotional pressures in recruiting respondents for research. International Journal of Social Research Methodology 15(3): 199–210.

54.

Van Maanen

(2011) Tales of the Field: On Writing Ethnography. Chicago: University of Chicago Press.

55.

Webb

(2002) Using focus groups as a research method: A personal experience. Journal of Nursing Management 10: 27–35.

56.

Yeatts

Seward

(2000) Reducing turnover and improving health care in nursing homes: The potential effects of self-managed work teams. The Gerontologist 40(3): 358–363.

57.

Zaman

(2008) Native among the natives: Physician anthropologist doing hospital ethnography at home. Journal of Contemporary Ethnography 37(2): 135–154.