QJER logo 2
[ Contents Vol 17, 2001 ] [ QJER Home ]

Emotional labour in educational research

Lucy Jarzabkowski
This article considers the role of the researcher in conducting an ethnographic case study of a primary school teaching community. It demonstrates how the research conducted within this organisation presented various ethical dilemmas for the researcher. Several issues appear to contribute to the creation of such dilemmas, particularly the topic of the research, the kind of participant group and the methodology used. Challenging the notion that educational research is a rational, linear, values-free process, this article recognises that the researcher's experience is emotion laden and that emotional labour may be performed as part of the research process. A proposition is put forward that emotional labour may be a significant feature of certain types of qualitative research and that its performance should be more openly acknowledged when conducting research within educational communities. The question of who benefits from the research takes on new meaning when considered from this perspective.

Qualitative research conducted in natural settings is common in the field of education. However, negotiation of access and the terms under which a researcher will be able to work with participants can be difficult to arrange. The research 'agreement' between participants and fieldworker is often a tenuous one. To a large extent it is formulated on a balance between perceived benefits and costs, although this is rarely made explicit by either party. We cannot talk about benefits of research for either party without thinking equally about the associated costs. This article seeks to address the issue of who benefits from research by investigating the emotional dimensions of the research process in action.

Ethnographic fieldwork in any setting can be an emotional challenge for a researcher (Kleinman & Copp, 1993). Research can be particularly emotion-laden in studies of small organisations where participants know one another well and a researcher's presence is obvious to everyone. In this article I describe my role as a researcher in a primary school and the ethical and emotional dilemmas that arose through my close association with participants. The emotional labour of research is clearly identified in this article, bringing into question who actually benefits from research and at what personal cost.


This article is based on a study conducted for a doctoral degree. At the time it was conducted I was a full-time PhD student with a particular interest in school organisation and culture. I wanted to examine the emotional dimensions of teacher collegiality in action. I decided that the best way to approach this topic would be to examine one site as a case study. Not wanting to be too ambitious with the research, I elected to use a primary school staff group rather than a secondary school group. This was because the latter, being much larger in size, is more likely to have what Hargreaves (1994) terms a 'balkanised' culture rather than a more ideal 'culture of collaboration'(Nias, Southworth & Yeomans, 1989) that I hoped to identify.

The case study site I eventually selected was a middle-sized, systemic Catholic school in an urban area. The school and its staff were completely unknown to me prior to my initiation of contact for research purposes. I was very much an outsider to the staff community on commencement of the research. How my position changed in this respect will be discussed below.

An important aspect of this particular study is that it was to be an investigation of a small community of people, specifically the working relationships of those people, 22 adults in all. However, not only did I wish to study how they interacted with one another, I wanted also to examine the emotional dimensions of the interactions from the perspective of those involved. In saying this, it is also important to explain that the research was being approached not from a psychological angle, but a sociological perspective. It was the emotional health of the collegial group that was important to me rather than the personal psychological foci of individual staff members.

A study focusing on organisational culture and interpersonal dynamics, and one with such a specifically emotional perspective, called for an in-depth, long-term association with members of the organisation. I therefore chose to use an ethnographic approach. I elected to be a participant observer at the school as well as to conduct interviews with any staff members who were willing to provide the time. In total I spent over one quarter of the school year on site, these days being spaced throughout the period of one academic year. I made copious fieldnotes during this time. I also formally interviewed over 80 per cent of the professional staff. The principal and assistant principal were interviewed together and separately on several occasions.

All interviews were transcribed by me and then cleared by participants. Teachers were granted anonymity within the group via the use of pseudonyms but it was impossible to do this for the principal and assistant principal, who allowed themselves to be identified by their positions in the final report, which I presented to them on completion of the research. Some teachers also, in granting clearance of their interview material, commented that they felt their colleagues would be able to recognise them by their stated points of view and/or speech patterns.

In reality, the use of pseudonyms can itself be an ethical dilemma for a qualitative researcher in a small community. As Nespor (2000) notes, members know one another and events so intimately that for any written account to be persuasive it risks members being identified to one another. Indeed, attempts to anonymise people and places 'protects participants from identification and consequent harm or embarrassment only insofar as local people have no objection to what's written' (Nespor, p. 548). This point is particularly relevant to an article on the benefits of research, given that teacher participants in this study readily gave clearance for me to quote interview material that in some instances reflected their negative views of their principal or peers, a point to which I will return later.


For an ethnographer to gain a participant's perspective on the culture of a community, a significant amount of co-operation is essential. It is crucial, as noted by Adelman (1985) and Glesne (1999), that the ethnographic researcher makes an effort to be accepted by the community that he/she wishes to study. Good rapport with the community is essential if ethnographic research is to be successful. The researcher has to give serious consideration to his/her role within the chosen community.

At the case study school I found some staff members very welcoming, particularly the principal, with whom I developed a good working relationship. It did not take a great deal of time to overcome my 'outsider' status. There were several factors that made this easier. Foremost was the role that I chose for myself at the school, a role that centred on the principle of reciprocity, something that Glesne (1999) sees as crucial for qualitative researchers. From experience I was well aware that schools are very busy sites, and teachers and administrators who work in them are continually burdened with a variety of extracurricular duties. Therefore, it was important that some form of 'profit' or 'benefit' be derived from having a researcher in the school. I wanted my presence in the school to be seen as part of a collaborative agreement, for mutual benefit. While acknowledging to myself and others that my prime motivation for being in the school was to collect data for my thesis, there were also some benefits that the school could obtain from my presence.

Firstly, the research data themselves, and the final report, would provide the school with insight into its social operations, that is, its cultural dynamics, structures of support, and the 'emotional economy' as such, which all affect its ability to achieve its goals successfully and to manage change effectively. The information my research would provide could be used for the establishment and evaluation of development plans, mission statements and similar operating documents.

Secondly, since I had previously been a teacher, in studying a group of teachers I had a 'skill' I could offer. I proposed to become a volunteer, part-time staff member within the school. After initial discussion with the staff before my arrival, the principal suggested that I might like to work in the library on the days I visited. The library was the hub of teaching and learning activity, being home to a small computer suite as well as print and other media, and the principal felt that this would provide me with a prime position for viewing school activities during class time. It is important to note that the most important data collection times for me were before school, during break times and after school - the times when teachers congregated for professional or social reasons.

In the early period of visiting the school, I spent the majority of class time covering books in the library and observing the operations of teachers and library staff. I also told teachers at a staff meeting that I would be willing to assist them in any other way if they requested it. One teacher asked me to take an English as a Second Language student for special tuition, and this I did on a regular basis for some weeks. I also relieved various teachers of classroom duties for short periods at the request of the principal. However, it soon became obvious that what teachers really needed was some regular release time. I therefore offered to do a series of days of relief teaching, releasing in half-hour blocks those teachers with minimal non-contact time, thus allowing them to attend to other pressing work. I also spent several days working in the school office when the two general staff based there were required to attend in-service at the central education office. Later in the year I also undertook some paid relief work, taking classes for teachers who attended professional development courses.

Southworth (1987) notes that joining a school as a part-time staff member involves surviving a credibility test. The most obvious of these, he suggests, is being able to prove one's worth as a teacher. Performance at the 'chalk-face' obviously matters, and to gain acceptance one must be able to demonstrate capability as a classroom practitioner. I felt that this was one factor that helped me to be accepted by the staff. After teaching their classes, either in short blocks of time, or for a full day, I was able to relate better to individual teachers, and they to me. I was able to share their concerns about particular students and I gained knowledge about their class programs. Teachers saw me out in the playground doing supervision duty. They saw me managing classes during assemblies, accessing teaching resources and relating to students in a wide variety of ways. I became a much more credible character, someone who shared their experiences as a teacher, and several staff members were keen to let me know that I had 'become one of them'.

Southworth (1987) also notes the need to observe the social mores and personal values of the staff to gain acceptance. In a study of staff relationships this is crucial, and I was very attentive to the cultural environment. My occupation, gender and age were obvious factors that helped me gain acceptance. Also, at all times I dressed like a teacher. I was attentive to duties and responsibilities that arose as a teacher. I attended religious services and partook in other spiritual activities as part of my role. I joined in many professional and social events. I made every effort to do what I could to make my presence appear less of an intrusion within the community. In response, teachers and administrative staff made me feel very welcome and I was personally invited to all special events that occurred during the year.


One important factor in conducting any kind of research is the researcher's relationship with his/her participants. As many authors attest, the relationship is power based. In most cases it is the researcher who holds the power. This is particularly true for those who take a positivist approach to research, but may also be the case in ethnographic studies, particularly where the researcher is working with people from disadvantaged backgrounds (see Adams, 1998; Luttrell, 2000; McGraw, Zvonkovic & Walker, 2000). This was obviously not the case in my study, as the chosen site was one with which I could identify on very many levels. However, researchers hold power because of the choices that they make about what to research, how to gather the data and who is interviewed or observed, and most importantly because it is the researcher who interprets the data and ultimately decides what will be reported. In that regard I held power as a researcher.

However, ethnographic fieldworkers also know that the success of their work depends on their participants (Kleinman & Copp, 1993). Qualitative researchers know that to obtain information they need to build up rapport with the research participants. They need to be trusted and accepted by participants (see Mun Wong, 1998; Sword, 1999). As Adams (1998, p. 226) notes, 'in reality, the researched also hold considerable power [since] they can give or withhold information, access, or the right to observe'. For the researcher to understand truly the beliefs and principles underlying the group's behaviour, he/she must become the student, and group members become the teachers. Glesne (1999) sees that considering oneself a learner in the field is important because it helps create a specific type of interaction with others. The balance of power can therefore turn towards the research participants' favour.

In an ethnographic study, attachments must be formed with participants so as to be able to understand their actions and their perceptions of issues. It is here that ethical dilemmas may develop, particularly in studying a small group of people. There were 22 members of the professional staff (teachers and administrators) at my case study school. It is not unexpected that one would form closer relationships with some of them than others. As noted above, I felt welcomed into the staff community within a short period of time, but my relationships with each staff member differed. A degree of distance continued throughout the year with those members of staff who declined to be interviewed as part of the research. Initially, I assumed that these teachers would change their minds when they got to know me better. However, when each of them declined at a second invitation half way through the year, I accepted that they had other reasons for not wishing to be directly involved in the research.

I became quite friendly with several teachers very readily. I did not see it as a problem to develop collegial relationships with staff, as it is only what I would have expected to do as a new staff member anywhere. Interestingly, whereas most staff were warm and welcoming on my arrival, a few teachers were particularly kind and caring, readily including me in activities and conversations, joking and sharing with me, and generally engaging me in the social life of the community. If I was away from school for longer than a week, I felt a strong sense of welcome on my return from some staff. There were five teachers in particular, all female, about my age or older, with whom I felt I developed a close bond of friendship during the year of my study. This bond did not extend outside the school environment but was significant none the less.

Kleinman and Copp (1993) make the point that researchers feel comfortable when they like the research participants and know that they are liked by them in return. This was certainly the case for me. However, the authors also suggest that it is important that we analyse these mutual good feelings as they can be a useful source of data for the study. We need to consider what participants think we are offering them, rather than automatically assume the smooth relationships are a testimony to our skill at developing rapport with others. Kleinman and Copp (1993) point out that it may be that as researchers we give participants a chance to express opinions that they are otherwise unable to convey, or perhaps we provide legitimacy for their concerns. Thinking about the relationships developed in the field, and our feelings about those relationships, is valuable to researchers, and certainly assisted me in my attempt to position myself in the debate on who benefits from research.

As with the teaching staff, a positive relationship with the principal also developed over time. My previous experience in school administration and our shared interest in higher degree studies may well have contributed to that, although we were not close in age. Perhaps more significantly the relationships may have had links with the emotional labour involved in leadership and the need to 'unburden' oneself after emotional experiences (Jarzabkowski, 2000). Whatever its origins, this relationship contributed to a significant personal dilemma for me in conducting the study. It became clear to me that a problem might exist only after I started my interviews with teachers, about two months into the school year. I discovered that there were several teachers on staff who did not favour the leadership style of the principal, considering her to be too autocratic. Interestingly, through interviews I learned that the principal saw herself to be very democratic in her approach to decision-making.

At first I was quite concerned about this obvious disparity of views. I was also disturbed to uncover such a negative issue, fearing that it might 'explode' and ruin my research. My personal relationship with both the principal and the individual teachers presented an emotional conflict for me. Initially, I worried as to whom I should or would support in what appeared to be a silent 'war'. I personally saw no evidence of the autocratic practices, but on the other hand could also see how some teachers might have read this as a dominant leadership style.

It became apparent to me as time went on that some teachers were actually trying to use my research as a vehicle for airing their opinions about the leadership of the school (see also Danaher, 2001, in this issue). Those with negative opinions of the principal's leadership were almost insistent that their views be heard, recorded and published. My research report was to be the agent of expression for the views that some teachers felt could not be delivered face-to-face. Coming to the understanding that some teachers might want to take advantage of my research in this manner made it much easier to cope with the personal feelings that this experience generated.

Fortunately, my intention had never been to focus on leadership within the school and I had to remind myself strongly of this in the days following my discovery of the negative views. I tried to keep the matter in perspective and remind myself that many teachers take issue with their school's leadership. It is not unexpected or unusual within a school community to find such views held by teachers of their principals. Much to my relief, I also discovered in further interviews that teachers' opinions about the principal's leadership differed widely, and the initial negativity was not as widespread as I first feared. However, all relationships within the staff community were important to my research. Therefore, to investigate principal/teacher relationships I had to retain the trust and friendship of all staff members.

I thus found myself in a situation where I was not able to display openly any alignment with the principal when gathered with teachers in the staffroom. To do so would have inhibited the flow of conversation and cut off a valuable source of data for my research. I also could not be seen to take sides in any discussions that teachers had about distribution of workloads, decision-making processes or other organisational matters that reflected on the leadership of the school. Similarly, I could not enter into discussions on the work ethic of other teachers. This constraint was also evident during interviews with staff. I tend towards openness and honesty in interviewing, and support Oakley's belief (1981; cited in Fontana & Frey, 1994, p. 370) that there is 'no intimacy without reciprocity'. It was necessary, however, to be more guarded than normal in an effort to encourage the sharing of personal perspectives on staff relationships within the school.

It was evident that my position as researcher overlapped with my position as staff member. My role as researcher and interviewer provided me with personal insights into how participants viewed their relationships with one another and the emotional dimensions of those relationships, but it was critical that I guard those insights, regardless of the positive or negative light they might have put on group discussions. As well as the obvious ethical issues, to discuss other members of staff in their absence was personally unacceptable to me. I thus found myself in a paradoxical situation. On the one hand it was necessary for me to appear neutral about relationships and about many organisational and interpersonal issues because to voice a position different from the dominant one would have cast me as an outsider and prevented access to rich data. On the other hand I could not be a real insider because I could not indulge in sharing my true feelings about the work situation or my relationship with other members of staff, either positively or negatively. In reality, I developed what I term 'quasi-insider' status through my maintenance of neutrality (see also Harreveld, 2001, and McConachie, 2001, both in this issue).

This in itself is not without personal cost. I needed the acceptance of individuals and the group to continue the data supply for my thesis. To the extent that I never had a conversation stop when I entered the room, I believe I retained that acceptance throughout the school year. The trade off was my emotional experience. I had to hide my true feelings about people and issues during the study. I had to display at least an acceptable neutral position, but never a position counter to the dominant expression within the group. There were certain times on site where I could not 'be myself'. Admittedly there were equally numerous times where I was able to display true feelings and opinions, and enjoyed the experience of being 'part of the staff'. At some points, however, usually in the expression of negativity about organisational issues and personal relationships, I had to suppress my natural feelings and emotions.


The hiding or changing of true feelings in an effort to display a more acceptable emotional front is called emotional labour. Hochschild (1983) was the first to highlight emotional labour as a form of work, describing it as 'the management of feelings to create a publicly observable facial and bodily display' (p. 7). Ashforth and Humphrey (1993, p. 90) define emotional labour in service roles as 'the act of displaying the appropriate emotion' by which they mean that there is conformity to expression norms or rules of emotional display.

The issue of emotional labour in research can be viewed from two perspectives - the participants' expectations of me and my own expectations of myself. First, as a researcher and a 'guest' in the school, I was expected by participants to behave in a certain way. I was expected to be courteous and pleasant and obliging because I had been privileged to use this particular group for my own personal benefit as a researcher. I was expected to manage my emotions to conform to these expectations. From a personal perspective, my emotional display was also limited by my own desire to 'fit in' and become an 'insider'. I needed to behave in a certain way to access data; that is, I needed to be 'one of the staff' to gain the acceptance necessary to obtain personal information from people. I had, in effect, two 'masters' for whom I needed to practise emotional labour: the participants and myself-as-researcher. Another interesting point is that, had I been a real insider, I would have had somewhat more emotional freedom of expression.

I posit that many researchers, particularly those working closely with participants, as is the case in ethnographic studies, experience emotional labour. Kleinman and Copp (1993) imply this in their personal experiences as researchers. Other personal accounts of research similarly imply that feelings about participants and their circumstances require emotion management (see, for example, Adams, 1998; Chesney, 2001). What complicated the usual demands of emotional labour for me in this particular case was the emotional tension among participants. In a larger organisation, the tension between superior and subordinates would not have evoked such personal difficulties for a researcher, but I clearly had 'a foot in both camps' and needed to display neutrality to encourage all participants to continue to share with me their personal experiences.

Stress is one of the most obvious costs associated with the performance of emotional labour. People can become estranged from their feelings and this can have a serious effect on mental, and ultimately physical, health. The performance of emotional labour causes, in effect, a separation of mind and body, which can in the long run be physically and mentally harmful for workers. Hochschild (1983, p. 90) called this 'emotive dissonance'. Despite the researcher's inability to eliminate 'tensions, contradictions, or power imbalances', as Luttrell (2000) suggests, it is important for the researcher to 'name' them. She supports, as do Kleinman and Copp (1993), identifying moods, doubts and worries within fieldnotes as a means of identifying and acknowledging the self as ethnographer.

We need to remember that '[a]t its core, ethnographic research is creative, inventive, emotionally charged, and uneasy' (Luttrell, 2000, p. 517). I suggest that this kind of research is not only emotion-laden but it also entails the performance of emotional labour, and it should likewise be acknowledged. Recognition and acceptance of emotional labour as part of the research process will ultimately assist researchers to analyse the research experience and their data better. More importantly, it will help them to cope with feelings of 'being compromised' in the field while working closely with participants.


This brings us back to the debate on who benefits from research. While externally it may look painfully clear that researchers benefit from research by gaining their higher degrees and/or publications, a closer look at the research process reveals that there may be considerable personal cost attached. Participants may quite unintentionally put emotional demands on researchers, forcing them to perform emotional labour. They might also be more overt in trying to use the researcher as an advocate, someone to expose their views in situations where they feel unable to do so, regardless of the initial purpose of the study. Benefits can obviously accrue to participants if they are successful in this endeavour. However, as researchers have ultimate control of how data are interpreted and what is written in the final report, in the long-term they are favoured as beneficiaries of research.

Another important consideration, however, is the dissemination of published research. In the case of the study described in this paper, while participants may benefit directly from an analysis of their own organisational relationships, the wider educational community can also profit if the research is published in academic and practitioner journals. An understanding of the emotional dimensions of collegial relationships in one organisation can be very valuable to others if they apply a similar critical analysis to their own organisations. Therefore, research that is published in the educational community can benefit a wider audience than the researcher and his/her participants. What might on the surface seem like a selfish act on the part of the researcher may, in fact, in the long-term, be advantageous to a much larger group.


Adams, J. (1998). The wrongs of reciprocity: Fieldwork among Chilean working-class women. Journal of Contemporary Ethnography, 27(2), 219-241.

Adelman, C. (1985). Who are you? Some problems of ethnographic culture shock. In R.G. Burgess (Ed.), Field methods in the study of education (pp. 37-50). Lewes, UK: Falmer Press.

Ashforth, B.E. & Humphrey, R.H. (1993). Emotional labor in service roles: The influence of identity. Academy of Management Review, 18(1), 88-115.

Chesney, M. (2001). Dilemmas of self in the method. Qualitative Health Research, 11(1), 127-135.

Danaher, M. (2001). Three dilemmas in researching Japanese environmental lobbyists. Queensland Journal of Educational Research, 17(2), 178-192. http://education.curtin.edu.au/iier/qjer/qjer17/danaher.html

Fontana, A. & Frey, J.H. (1994). Interviewing: The art of science. In N.K. Denzin & Y.S. Lincoln (Eds.), Handbook of qualitative research (pp. 361-376). Thousand Oaks, California: Sage Publications.

Glesne, C. (1999). Becoming qualitative researchers: An introduction (2nd ed.). New York: Longman.

Hargreaves, A. (1994). Changing teachers, changing times: Teachers' work and culture in the postmodern age. London: Cassell.

Harreveld, B. (2001). Discourse co-ordinations within adult literacy teaching. Queensland Journal of Educational Research, 17(2), 138-151. http://education.curtin.edu.au/iier/qjer/qjer17/harreveld.html

Hochschild, A.R. (1983). The managed heart: Commercialization of human feeling. Berkeley, California: University of California Press.

Jarzabkowski, L.M. (2000). The principal as an emotion manager. Paper presented at the annual conference of the Australian Council for Educational Administration, Hobart.

Kleinman, S. & Copp, M.A. (1993). Emotions and fieldwork. Newbury Park, California: Sage.

Luttrell, W. (2000). 'Good enough' methods for ethnographic research. Harvard Educational Review, 70(4), 499-523.

McConachie, J. (2001). Who benefits from exploratory business research? The effect of sub-cultures on the implementation of an enterprise system: An Australian regional university perspective. Queensland Journal of Educational Research, 17(2), 193-208. http://education.curtin.edu.au/iier/qjer/qjer17/mcconachie.html

McGraw, L.A., Zvonkovic, A.M., & Walker, A.J. (2000). Studying postmodern families: A feminist analysis of ethical tensions in work and family research. Journal of Marriage and the Family, 62(1), 68-77.

Mun Wong, L. (1998). The ethics of rapport: Institutional safeguards, resistance, and betrayal. Qualitative Inquiry, 4(2), 178-199.

Nespor, J. (2000). Anonymity and place in qualitative inquiry. Qualitative Inquiry, 6(4), 546-569.

Nias, J., Southworth, G., & Yeomans, R. (1989). Staff relationships in the primary school. London: Cassell Educational.

Southworth, G. (1987). The experience of fieldwork; or insider dealings, who profits? Cambridge Journal of Education, 17(2), 86-88.

Sword, W. (1999). Accounting for presence of self: Reflections on doing qualitative research. Qualitative Health Research, 9(2), 270-278.

Author details: Lucy Jarzabkowski is a lecturer in educational leadership and management at Murdoch University.

Address for correspondence: Dr Lucy Jarzabkowski, School of Education, Murdoch University, South Street, Murdoch, WA 6150. email: l.jarzabkowski@murdoch.edu.au

Please cite as: Jarzabkowski, L. (2001). Emotional labour in educational research. Queensland Journal of Educational Research, 17(2), 123-137. http://education.curtin.edu.au/iier/qjer/qjer17/jarzabkowski.html

[ Contents Vol 17, 2001 ] [ QJER Home ]
Created 12 Dec 2004. Last revision: 19 Dec 2004.
URL: http://education.curtin.edu.au/iier/qjer/qjer17/jarzabkowski.html