Work-Identity in Ethnographic Research: Developing Field Roles in a Demanding Workplace Setting

Background

The data in this article are drawn from a project about immigrant care workers' communicative practices in elderly care facilities in Sweden. One of the researchers had a special focus on care workers' practices of talk in their everyday communication with the elderly. The other researcher concentrated on care workers' literacy practices. The data collection methods we have used include participant observation, qualitative interviews, and audio/video recordings. The aim of our study was to identify, in collaboration with the community, the central dilemmas in care work (tensions between institutional frameworks and professional knowledge, among others), and to find out how these constrain the care workers' documentation practices and everyday communication with the elderly. The intention to work with rather than for the community required close interaction with its members (see Roberts & Sarangi, 1999). Attaining this aim required establishing close contact with the care workers and participating (even to a small extent) in their daily tasks. The population of the study consisted of approximately 50 care workers (assistant nurses and elderly carers) employed at ten different wards in three nursing homes. The nursing homes mainly host residents with dementia and therefore with great and immediate care needs. The majority of the care workers are immigrants from Middle Asia, Latin America, and Africa, and they have Swedish as a second language. Only a minority of the care workers are Swedish-born. Being a care worker with Swedish as a second language was one of our research topics in the project. Because elderly care in Sweden is a highly multilingual work context, we were interested in how the language of immigrant workers was shaped in the workplace. The questions we asked were how different kinds of work-identity are handled in communication and how institutional frameworks meet and contradict other perspectives within care work. Such research questions require an interdisciplinary approach that emphasizes microanalysis of situated interaction and linguistic text interpreted through an ethnographic understanding of the context of the workers' everyday tasks (Creese, 2008; Rampton, 2007; Tusting & Maybin, 2007). In line with recent developments within applied linguistics, we maintain that language and social life are mutually shaping (Rampton et al., 2004, p. 2). Our point is that the methods provided by ethnography generate insights into the language practices under study, which cannot be achieved in another way.

A central issue in ethnographic methodology is the reflexivity between the researcher and the research being produced. This stems from a tradition in which ethnographic research requires the researcher to be actively involved in social action. This involvement of the researcher in the field will have an unavoidable impact on the practices being studied. Where the study of language practices is concerned, reflexivity also raises another issue, namely the difference between how participants and analysts understand these practices (Creese, 2008, p. 579). The linguist's use of structured analytical tools runs the risk of deviating from the participants' perspective. In order to address this issue we have used participant observation and ethnographic interviews as data collection strategies. Our view is that the researcher's commitment in the field to come closer to the participants' perspective informs the analysis of language. We will return to this issue in the final section of the article in our discussion of how our position in the field impacted on the generation of data.

A central tool in ethnographic research for pursuing an emic perspective is the strategy of participant observation. Nevertheless, the nature of participant observation is contested and its foundations have been widely debated in recent studies (Allen, 2004; Angrosino, 2005; Angrosino & Mays de Perez, 2000; Davies, 2008; Savage, 2000). In particular, to date there has been no homogeneous view of the meaning of the term participation in participant observation. The several attempts that have been made to map out the various field roles that researchers may adopt in settings as participant observers (Adler & Adler, 1994; Gold, 1958) are nowadays considered of restricted value (see Hammersley & Atkinson, 2007). Ashworth (1995) highlights attunement to the emotions and motivations of the group's activities, for example attunement to the shared and taken-for-granted stock of knowledge that contributes to the forms of interaction of the group being studied, as crucial constituents of participation. Interpreted within Ashworth's theoretical construct, participation has the same meaning as to participate in the role of a practitioner. The implication of this reasoning is that the researcher, by virtue of his or her outsider status, will face difficulties when entering a new social group. Allen (2004), however, has argued that there are many other modes of participation in the field, and she warns against “an exclusive focus on the meaning of participation” (p. 17). Staying in the field for a longer time inevitably entails the researcher's involvement in the lives of the communities they study. As a consequence the researcher cannot remain an outsider throughout the process. In fact, many ethnographers account for the possible ways of negotiating a feasible membership role (Angrosino, 1997; Palmer, 2010; Perrone, 2010; Woodward, 2008).

The ethnography of nursing homes is a growing field of research (Diamond, 1992; Foner, 1994; Henderson & Vesperi, 1995; Higgins, 1998; Lee-Treweek, 2000; Reed-Danahay, 2001; Tinney, 2008). As has been described in this literature, the nursing home is a challenging research site and also a stigmatised location for those who live there. Several ethnographers (e.g., Backhaus, 2008; Grainger, 1995; Lee-Treweek, 2000) have focused on the horrors of nursing home life and described them as “total institutions” (Goffman, 1961) in Goffman's sense. Reed-Danahay (2001, p. 49), however, makes it clear that recent research has sought to understand nursing home life from the inside, in a way that is attentive to the stresses and strains on those who work there, for the most part women of colour in low-paid, low-status jobs. This endeavour to pay attention to the burdens of the staff was especially relevant in our study. Care workers in Sweden are under constant scrutiny by news media, where they are often exposed to public criticism and blamed for scandals. Elderly care work is associated with low prestige and, as in other Western countries, employs the highest proportion of immigrants, many of whom are women with few opportunities to advance in the labour market. Cuban (2009, pp. 178–179) has argued that the new demands placed on carers in the market-based industry help explain why migrant women are recruited to work in elderly care. Care employers seek out workers who are willing to comply with demanding requirements and stressful workplaces.

In previous publications, Nikolaidou and Karlsson (2012) have discussed the complex nature of care workers' knowledge and competence as the result of three different aspects of work-identities. Based on Roberts and Sarangi's (1999, 2003) analysis on discourses in medical care, we have argued that care workers' decisions and actions at work depend on three different expressions of selfhood: (a) the institutional work-identity, assigned through a hierarchical position in an institution; (b) the professional work-identity, constructed together with colleagues through shared responsibilities and the use of a common discourse; and (c) the individual work-identity, based on individual character and past-life experiences. These three aspects of work-identity are similar to Gee's (2000) categorisation of identity construction. He makes reference to an Institutional-identity that foregrounds membership in various institutions (similar to the institutional work-identity, described above) and an Affinity-identity adopted by people who share “allegiance to, access to, and participation in specific practices that provide each of the group's members the requisite experiences” (p. 105) (similar to the professional work-identity, described above). The third aspect Gee identifies is Nature-identity, which suggests an essentialist view of identity. This is similar to facets of individual work-identity (e.g., gender, ethnicity, and color of skin, etc.), even though our understanding of individual work-identity is broader. Finally, Gee identifies Discourse-identity, which is the way our selfhood is constructed by the way we are talked about by others and by the way we talk about others. This final aspect can be identified in all three expressions of work-identity.

In a recent publication on documentation practices in elderly care homes, we have shown that the institutional work-identity is dominant, especially in relation to the individual (Nikolaidou & Karlsson, 2012). Care workers are not allowed to include personal experiences and feelings about their interactions with the residents in documentation, but are instructed instead to follow rigid guidelines on the contents of records and to use a very restricted documentation language. Care workers promoted the professional work-identity in the documentation; for example when they included information they believed to be useful to their colleagues but not in line with the institutional guidelines. Documentation practices become more meaningful to care workers when they are related to their professional identity rather than to their institutional identity, whereas the individual aspect does not go down on paper and is instead expressed orally.

Entering the world of elderly care was not an easy task. This can be seen in relation to the complex nature of the research participants' professional roles, consisting of the various aspects of work-identities accounted for above. In this article, we therefore suggest an additional perspective to the understanding of membership roles, one that views them in relation to the three different expressions of identities. We examine our participation in the nursing home field by juxtaposing the care workers' work-identities with our own field roles as researchers. We argue that our choices and decisions in the field were also defined by the aspects of identity we chose to promote each time. Our actions were in a constant dialogue with those of the care workers, and together we helped each other shape new identities throughout the research process.

Although neither of us was a complete outsider in relation to the care workers, we had shaped our identities as researchers in different ways. One of us (Nikolaidou) was a young female in her late twenties, with a second-language background in Swedish and with limited experience of social acquaintances with elderly people in Sweden. The other (Jansson) was an older native Swedish female in her sixties who had worked in the field as a trainer in a previous project dealing with staff development and second-language use at work. Our aim in this article is to demonstrate how these different positions impacted on our field relationships and how we managed to negotiate our membership roles by adjusting our field identities to those of the research participants. In the following sections, we describe the activities through which we engaged in the field setting and how the participation we undertook positioned us in relation to the care workers.

Entering the Field: Juggling Between Roles

In ethnographic research, it is not possible to predict the way the study will progress or what kind of relationships will be established with the research participants. Due to our lack of nursing background and the nature of the study, it was evident from the beginning that we could be given only the status of peripheral members. Adler and Adler (1994) argue that peripheral membership can be intentionally sought when the researchers do not wish, for a number of reasons, to participate in all the activities of the group they are studying.

Negotiating Field Identities

The benefits of enacting a field identity are frequently discussed in the ethnographic literature (e.g., Dwyer & Buckle, 2009; Hammersley & Atkinson, 2007). Many ethnographers explain how a membership role allows them more rapid and more complete acceptance by their participants. In the following section, we show how we gradually, and often unconsciously, modified our actions and behaviour in the field in order to come closer to the care workers. We illustrate how we negotiated our field identities, not only to be accepted by them but also to attain a better understanding of their working knowledge and experiences and the reasons for the choices they make concerning communication.

Discussion

In this article, we present a reflective account of the way we as ethnographic linguists approached a research field that was almost unknown to us. Our aim has been to show how we managed to approach the research participants by examining their knowledge and work-identities and by making an often unconscious effort to align our identities to theirs. We discuss the care workers' work-identity as a complicated phenomenon composed of three different aspects: institutional, professional, and individual; and we show how they came to be relevant in our efforts to create a membership role for us in the field.

We first entered the field with the initial dichotomy between insider and outsider researcher in mind. We viewed ourselves as complete outsiders to the field of elderly care and wished to gradually negotiate an insider role. We soon came to realise that this dichotomy was too barren and did not accurately describe the situation we were in. Even though neither of us had ever worked in a nursing home before, we felt we could not be called complete outsiders, because we both had some kind of previous relation to the field and to the research participants. According to Christensen and Dahl (1997), the insider-outsider dialectical roles constitute a continuum rather than a dichotomy “with the extreme points in the continuum existing in conceptualization rather than fact” (p. 274). The authors have questioned the extent to which outsiders can really be outsiders in relation to what they are researching and suggest that all researchers are close to their research participants in one way or another. Similarly, Allen (2004) has argued that the insider-outsider status in the field “is something to be discovered and renegotiated as part of the research process” (p. 22). We found that this role negotiation was performed through a constant shaping and reshaping of our identities, in relation to those of the research participants.

Early in the research, we realised that the fastest way to come closer to the care workers was to make use of the shared features in our personalities and backgrounds, or in our Nature-identities (Gee, 2000). For example, Jansson's ease in communicating with the residents made the care workers appreciate her. Nikolaidou had a shared immigrant background with the research participants, and this enabled them to open up to her in many different ways. Interestingly, neither of us had initially thought that these individual features would play an important role in the research. Rather, they gradually became prominent during our efforts to establish good relations with the participants and an insider role in the field.

The professional aspect of the care workers' identity was harder to interpret and potentially imitate. As academics, we had different professional backgrounds and different aims from the care workers. We wanted to collect good and informative data, whereas the research participants focused on the residents and also on protecting their own professional integrity. Some of the examples we have discussed in this article indicate this incongruence of professional identities; for instance, the care workers initially refused to give Jansson access to the residents' everyday lives and some of them resisted the idea of participating in the project, while others asked Nikolaidou to help with tasks for which she was not prepared.

The more the research progressed, the easier it became for us to decode the care workers' professional identity, and the more we tried to align our identities to theirs. In other words, we made an effort to prioritize their goals in the activity over our own and, in this way, to be accepted as something more than just observers. The care workers gradually started placing trust in us. They allowed, and sometimes encouraged, Jansson to be involved in certain forms of care, while they talked more freely to Nikolaidou about their problems with documentation at work. Jansson was no longer merely a trainer and Nikolaidou merely a researcher. We had embraced their own priorities and this was made clear to them. They could now talk in front of us freely at lunchtime and they could share their opinions about the manager and the residents. We would argue that we had come as close to being colleagues as is possible for someone who is actually in the field as a researcher.

The institutional aspect of the care workers' identity was of great interest for our study and in some cases easier to adapt to. Some aspects of care workers' institutional behaviour were imposed on us from the very beginning; for example, we had to wear the same clothes they did for sanitary reasons. Similarly, we were committed to treating the residents with respect and discretion. But this is where the similarities ended. As researchers, we were not obliged to provide care to any resident or to participate in the documentation processes. This often placed the care workers in an awkward position because they did not know where to position us in the institutional structure. This is made clear in Jansson's example when she found a resident on the floor. The care workers were not sure if they were supposed to include her in the documentation practice because she did not have a formal role. Nikolaidou often encountered mistrust when she asked to make copies of the residents' files, and she had to reassure the care workers more than once that she had permission for this. Our institutional identities could hardly be aligned, therefore, with that of the care workers because we belonged to a different institution and were subject to different rules and frameworks.

Despite the strategies we invented in order to acquire a membership role, issues of power were always at stake in the research field. Power relationships between researchers and research participants are a constant concern in many ethnographic studies (e.g., Beoku-Bets, 1994; Labaree, 2002; Tinney, 2008). Unequal distribution of power in our study was manifest in many different ways, often experienced as an obstacle in the negotiation process between the care workers and us. Because of her age and professional status, the participants often saw Jansson as a language expert who was there to help them and even correct them. In addition, she was associated with an earlier project where she had been working as a communication trainer. Jansson found it difficult to convince the care workers that she was there only to observe and learn and that, in most cases, they were better judges than her about what should be written. Nikolaidou, however, struggled with presenting herself as an immigrant who is also an academic. Even though she managed to make successful use of her identity as an immigrant who has difficulties with Swedish, her professional status contradicted the similarities she claimed to have with the care workers.

Finally, the field roles we negotiated for ourselves had a great impact on the kind of data we gathered. Our involvement in the field puts into focus the way in which the researcher's involvement alters the language practices under study, and hence the data collected (Tusting & Maybin, 2007). Because we came so close to decoding and understanding the components of the care workers' work-identity, it helped us get a complete picture of the communication and documentation practices in the nursing home. Our analysis demonstrates how we moved from a horrified perception of the nursing home to a more nuanced picture that is more sensitive to the care workers' dilemmas. This tension between perspectives is reflected in earlier research within the field of nursing home ethnography. Recent researchers are more sensitive to the strains and stresses of those who work there (Reed-Danahay, 2001), while others (e.g., Grainger, 1995) use Goffman's description of homes as “total institutions” (Goffman, 1961). What we have done in this article is to describe the strategies through which we revised our understanding of life in the nursing home and the language practices under study.

Jansson's ease with the elderly gained her the trust of both employees and residents, and she managed to get permission to video record the interactions between them. In this way, she captured the subtleties that characterized everyday staff-resident interactions in the nursing home, which made the dilemmas described by the care workers more understandable. Had it not been for her close field relationships with the care workers, the carers would probably not have shared their negative feelings about being forced to implement good communication practices in problematic and sometimes hostile environments. Nikolaidou's identification with the immigrant and second language status of the employees made it possible for her to ignore their claimed language difficulties and locate problems in the documentation structure of the nursing home. In other words, it was our negotiated research roles in the field and the continuous interaction with the research participants that determined the quality of the data we collected.

By “making a role” for ourselves we were able to negotiate access and finally collect useful data. It has been claimed that such information can be more easily gained by insiders (Haniff, 1985; Merton, 1972), but we tend to agree with more recent studies that describe field roles as contingent and shifting between contexts (Allen, 2004; Christensen & Dahl, 1997; Dwyer & Buckle, 2009; Labaree, 2002; Ledger, 2010). Even when not participating fully, the ethnographer is likely to align herself with the identities of the research participants. While we were influenced by our own identities, institutions, and research aims, the research process brought us closer to the care workers' professional world and their identities in the workplace. It was through unravelling the institutional, professional, and individual aspects of their identities, but also through opening up our own selves and our own identities that we gained their acceptance. In other words, it can be argued that ethnographic work at its best took place only when we started negotiating who we were in relation to the research participants and vice versa. Negotiation of this kind seems to represent ethnographic reality more realistically than arguing for, or aiming to adopt, fixed and robust roles in the field.