Observation is used as a research method in two distinct ways – structured and unstructured (Pretzlik 1994). In positivistic research structured observation is a discrete activity whose purpose is to record physical and verbal behaviour. Observation schedules are predetermined using taxonomies developed from known theory. In contrast, unstructured observation is used to understand and interpret cultural behaviour. It is based within the interpretist/contructivist paradigm that acknowledges the importance of context and the coconstruction of knowledge between researcher and ‘researched’.
Structured observation is used extensively in psychology, and it is from this aspect that many nurse researchers have adopted the method for their own studies (see for example, Booth et al. 2001). The principle characteristics of structured observation and examples of its use in nursing are summarized by Mulhall (1998). 306 The purpose of this paper is to explore unstructured observation. Although unstructured observation is a key method in anthropological and sociological research, it is used less frequently in qualitative research in nursing where the principal data collecting method is the interview (Merrell & Williams 1994).
Of course ethnographic research in nursing has used observational methods, but there is a relative paucity of information about this technique. This is not to imply that the methodological debate has been totally neglected, but to highlight that the whole breadth of discussion has not taken place in the nursing literature. Some of the issues that cause controversy in observation are: • the role of the researcher; • informed consent/deception; • access; • ?eld notes. O 2003 Blackwell Publishing Ltd Methodological issues in nursing research Observation in qualitative research
The ?rst two of these have received attention in the nursing literature and will be described only brie?y here. In contrast, the issues surrounding ?eld notes and access have been neglected. This paper will therefore focus predominantly on access and ?eld notes. However, before progressing further it is important to discuss how unstructured observation differs from structured observation and to outline why the former technique is an important data collecting method for nurse researchers. The place and importance of unstructured observation
The place The ?rst point to emphasize is that the label ‘unstructured’ is misleading. Observation within the naturalistic paradigm is not unstructured in the sense that it is unsystematic or sloppy. It does not, however, follow the approach of strictly checking a list of predetermined behaviours such as would occur in structured observation. Instead, observers using unstructured methods usually enter ‘the ?eld’ with no predetermined notions as to the discrete behaviours that they might observe. What is already known about this topic Unstructured observation is underexploited in nursing research. • This research technique has not been afforded adequate critical discussion in the nursing literature. • Most of the debate has centred on the problems of the role of the researcher, covert research, and informed consent. • However, unstructured observation poses other practical, theoretical and ethical dilemmas for nurse researchers that require consideration. They may have some ideas as to what to observe, but these may change over time as they gather data and gain experience in the particular setting.
Moreover, in unstructured observation the researcher may adopt a number of roles from complete participant to complete observer, whereas in structured observation the intention is always to ‘stand apart’ from that which is being observed. These two different stances re?ect the two paradigms through which these methods arise. ‘Structured observers’ are attempting to remain objective and not contaminate the data with their own preconceptions, whereas ‘unstructured observers’ carrying with them the tenets of the naturalistic paradigm would contend that it is impossible to separate esearcher from ‘researched’. But should such a de?nite distinction be made between these two types of observation? Pretzlik (1994) claims that both may be used in the same study and cites as an example the work of Dunn (1988), who explored children’s understanding of feelings and behaviours within the family. Dunn acted as an unstructured observer within families, but also measured the children’s actions and words in a structured way. However, such triangulation of methods drawn from opposing paradigms must question the basis of the researcher’s ontological and epistemological assumptions.
Sometimes structured coding instruments are developed after less structured observations and discussion within the research team (Silverman 1987). However, such coding instruments are not equivalent to the tightly de?ned precoded schedules used in observational work in psychology. The importance of observation: ‘using our eyes as well as our ears’ Strangely, although nurses rely heavily on observation during clinical work, it is not a method that ?nds widespread appeal within their research. Although clinical observation is different in form and structure to research observation, this situation still seems rather anomalous.
As Silverman (1993, p. 42) pertinently notes ‘Unfortunately, we have all become a little reluctant to use our eyes as well as our ears during observational work’. The way people move, dress, interact and use space is very much a part of how particular social settings are constructed. Observation is the key method for collecting data about such matters. Often the primary reason for using observational methods is to check whether what people say they do is the same as what they actually do. But there are other reasons.
Unstructured observation: • provides insight into interactions between dyads and groups; • illustrates the whole picture; • captures context/process; • informs about the in?uence of the physical environment. 307 What this paper adds • This paper discusses two methodological issues in unstructured observation – access into the ?eld and ?eld notes – and identi?es implications for research practice. • Nurse researchers need to recognise the hidden issues around gaining access, for example, power imbalance, self identity, and personal fears. • How the ‘?eld’ is conceptualized will in?uence researchers’ assumptions and practices within it. Careful attention to the form, meaning, use and construction of ?eld notes will enable researchers to clarify their particular theoretical stance and de?ne more closely when, where and how such notes should be recorded. O 2003 Blackwell Publishing Ltd, Journal of Advanced Nursing, 41(3), 306–313 A. Mulhall Clearly through observation it is possible to ascertain whether what people say they do and what they do in reality tally. However, it is important to note two things. First, both ‘accounts’ (what people perceive that they do and what they actually do) are valid in their own right and just represent different perspectives on the data.
Secondly, in my view observational data, rather more than interview data, are subject to interpretation by the researcher. I say this because observers have a great degree of freedom and autonomy regarding what they choose to observe, how they ?lter that information, and how it is analysed. Although the researcher conducting an interview has some freedom regarding the questions they ask, the informant probably plays a greater role in governing where the questioning leads. Moreover, many researchers return to the interviewee with the analysed data for member checking.
So interviewees may in?uence data analysis whereas those who are observed usually do not. Although focus groups are one strategy for ascertaining the views of dyads and groups, observation has the advantage of capturing data in more natural circumstances. The Hawthorne effect is an obvious drawback but my own ?eld experience has led me to believe that its effect in participant observation is overemphasized. Once the initial stages of entering the ?eld are past most professionals are too busy to maintain behaviour that is radically different from normal.
The situation seems analogous to a job interview – one can only perform for so long before reverting to type! As Frankenberg (1980, p. 51) notes, albeit in a wider cultural context, ‘I do not think that a single observer in, say, a village or tribe is going to change custom and practice built up over years or even centuries’. Observation also captures the whole social setting in which people function, by recording the context in which they work. The analogy of a jigsaw is useful here. Interviews with individuals provide the ieces of the jigsaw and these pieces are then ?tted into the ‘picture on the box’ which is gained through observation. Observation is also an ongoing dynamic activity that is more likely than interviews to provide evidence for process – something that is continually moving and evolving. Finally, observation is valuable because it informs about the in?uence of the physical environment. This aspect of observation is quite lacking in nursing research. Observations are made of people’s behaviour, but data about the physical environment seldom are collected.
Indeed Mays and Pope (1995) in their discussion of this method fail to mention recording information about the environment. The signi?cance of the physical environment may be illustrated by a quote from an ethnographic study of a teenage cancer unit (Kelly et al. 2000). The main ward had a large pool table taking up a 308 lot of space. This physical structure was highly symbolic. One mother said ‘the pool table is huge, and I don’t know, I don’t know if D has ever used it. I think in a way it makes it less like a hospital, I quite like it’.
Silverman (1993) makes a similar point about paying attention to the spatial organization of activities. He quotes the work of Perakyla (personal correspondence) who reports how the spatial arrangement of hospital wards differentiates groups of people. For example, patients do not enter nurse’s of?ces without invitation. Thus the physical environment creates different categories – professionals, patients, relatives – and in its turn the activities of participants reproduce this certain allocation of space. This aspect of observational data should not therefore be neglected.
Controversies in the use of observation Although observation is a unique and underused technique for gathering data in nursing research, it does pose problems. The role of the researcher and informed consent/deception have received attention in the nursing literature and will be discussed only brie?y here. The main focus of this paper will be access and ?eld notes. The role of the researcher and informed consent/deception The role to adopt during observation and the extent to which participants are fully informed are somewhat intertwined.
Typically researchers have referred to a standard typology of research roles (Gold 1958): • the complete observer, who maintains some distance, does not interact and whose role is concealed; • the complete participant, who interacts within the social situation, but again whose role is concealed; • the observer as participant, who undertakes intermittent observation alongside interviewing, but whose role is known; • the participant as observer, who undertakes prolonged observation, is involved in all the central activities of the organization and whose role is known.
But how useful is this typology in modern research? Covert research requires considerable justi?cation, although Mays and Pope (1995) suggest that it may be necessary in sensitive areas such as football hooliganism. Clarke (1996) provides a thorough discussion of the merits of covert observation as illustrated by his study of a secure (forensic) unit. The argument against covert observation is that people have the right to privacy and to know they are part of a study whilst investigators have no special privileges.
The case for covert observation is that deception is justi?ed on grounds of utility (bene?ts for others later) or in instances where this approach might liberate individuals (Johnson 1992). O 2003 Blackwell Publishing Ltd, Journal of Advanced Nursing, 41(3), 306–313 Methodological issues in nursing research Observation in qualitative research There are other issues concerned with the adoption of particular roles. Many nurse researchers undertaking observational studies use their own workplace as the ?eld setting. This has obvious advantages regarding ease of access and prior understanding of the culture in the study site.
However, at least four dif?culties can be envisaged. First, how, when you are a part of a culture, can you become aware of things that are normally taken for granted? Secondly, as Kite (1999) describes observation may interfere with giving care through the sheer effort of trying to do two things at the same time. Thirdly, the researcher must consider whether they are exploiting their relationship with potentially vulnerable people. Finally, little is made of how patients, clients or their families may reconstitute their views of practitioners who suddenly become researchers.
How does this change the therapeutic relationship? Informed consent/deception are also constant problems during observation. How informed can participants really be? I faced this problem during ?eldwork in a residential home for the elderly. Although I tried to explain what I was doing, responses from participants ranged from misunderstanding by the matron who thought I was interested in infection control (because of my previous background) through to the incredulity of residents as to why I should be interested in studying them: ‘Why?
Do you want to live here? ’ one joked. Despite efforts on my part, I did not feel con?dent that the residents and staff were fully aware of who I was or what I was doing. There is also the practical problem of how, especially in large and busy social settings, say an accident and emergency department, to inform and obtain consent from everyone who might ‘enter’ into the ?eld of observation. What should be done with data that inevitably contain observations of some who have consented and others who by chance have been included but have not given consent?
Moreover, although participants may give their permission to be observed, are they always clear as to the extent of that observation? For example, if nurses agree to being observed whilst giving patient care, have they also by default given consent to be observed when talking to colleagues? These problems are compounded by the unpredictability of observational work: it is dif?cult to say in advance quite what it might be useful to observe except in very general terms.
The dilemmas involved in adopting particular roles and the choice between covert and overt studies are indicative of wider issues concerning validity and ethics in observational studies. Using their own semi-covert study as an example Turnock and Gibson (2001) explore validity in relation to the role of observers, participant/nonparticipant observation, covert/overt studies and structured/unstructured data collec- tion. As one example they highlight the threat to validity that may occur when people in covert studies become aware that they are being observed.
They suggest that individual observers need to consider whether continuing data collection in this situation will provide valid information. These authors re-emphasize the importance of documenting such decisions. Taking another perspective on validity Graneheim et al. (2001) used multiple data collectors with different perspectives (insider or outsider) to observe the same situation. This may not accord with the idea that every researcher may produce a unique account of a situation that is valid in its own right.
But with extensive mutual re?ection, as undertaken by Graneheim and colleagues, these combined observations may have consensual validity. However, from a practical standpoint few projects are afforded the luxury of multiple data collectors. Ironically, although most researchers who undertake unstructured observational studies are working in interpretative/critical/feminist paradigms and are deeply wedded to the idea of participants as coresearchers, ethical problems may be considerable. Some of the problems concerning covert research and informed consent have been mentioned already.
Moore and Savage (2002) expand further on these highlighting the practical dif?culties in complying with a predetermined process of obtaining informed consent. They recount how the ethical committee’s requirement to ascertain informed consent on a day-to-day basis undermined the rapport between researchers and staff and appeared to question decisions that such staff had made already. Their suggestion is that researchers should respond to circumstances they encounter in the ?eld and that a prescriptive approach to ideal ethical practice will preclude researchers carefully considering social reality. Graneheim et al. 2001) make the point that where patients are unable to provide consent themselves, this should be sought from care givers and relatives. Furthermore, researchers need to be heedful of any violation of identity that may occur. In their study of a woman with dementia they made great efforts to pay attention to her reactions to their presence. Clearly the dilemmas involved in the conduct of observational research are considerable. It is important that researchers do not rely on ‘one-off’ considerations of these at the start of a project, but continue to interrogate the process of research throughout a study.
Access Access to the research site can be problematic. In health care settings access may involve a lengthy process of negotiation between several sets of professionals – managers, nurses, 309 O 2003 Blackwell Publishing Ltd, Journal of Advanced Nursing, 41(3), 306–313 A. Mulhall doctors, and social workers – quite apart from the formal approval of the ethics committee (Mays & Pope 1995). Similarly, Kennedy (1999) discusses the considerable effort required to liaise with managers to ensure that she was in the right place at the right time to observe home visits by district nurses.
Moreover, there is always the lurking consideration that the site has been chosen partly because of its accessibility to the researcher. My own Master’s study (Mulhall 1993) featured an element of this. I chose to study a residential home for older people, not only because I was interested in this social grouping, but also because the owner of such a home had informally granted me access. Similarly, many nurse researchers choose to undertake observational studies in their own workplace, citing their experience as an advantage.
However, it is possible that their decision may be in?uenced by the perceived ease of gaining access. There are also ethical dilemmas involved in access. The ?rst steps in negotiating access in health care settings often involve approaching the chief executive, consultant, director of nursing services or others in positions of authority and power. This is a pragmatic solution, but one that may override those further down the hierarchy. It might be argued that this problem is overcome by ensuring that informed consent is obtained from participants.
However, in practice this is not that simple. Moreover, if one’s manager has agreed to a study it is sometimes dif?cult to refuse to be involved. Negotiating access also involves a subtle but rarely acknowledged process of presenting oneself in the ‘correct’ way. Entrance may be denied if consciously or unconsciously the researcher does not meet the cultural expectations of gatekeepers. This process of access is most obvious in the way one dresses, speaks, or imparts particular knowledge.
For example, a researcher wishing to involve the patients of a hospital consultant in a study is likely to dress smartly, defer to the doctor’s wishes regarding the time and place of any meetings, and be ready to convince them that they have a good working knowledge of how things are run in ‘their’ unit. In short they will try and project an image close to that of the consultant themselves. Waddington (1994) discusses how gaining access involves a process of managing your identity; projecting an image; and convincing gatekeepers that you are nonthreatening.
Personal and psychological issues on the part of the researcher may also contribute to facilitate or impede access. It is not easy to enter unfamiliar social settings and a smooth entree is often related to the personal characteristics of the ? researcher, such as their ability to integrate quickly and mix with others. Most researchers entering the ?eld for the ?rst time express a sense of fear or hesitancy. Even experienced 310 ethnographers record how they have had to pluck up the courage to enter the ?eld, especially if it is unfamiliar territory.
Access, then, is not a straightforward process of speaking to the person in charge and obtaining the approval of the ethics committee. It usually involves considerable time and effort and a constant endeavour to strive for ‘cultural acceptability’ with the gatekeepers and participants in research sites. Field notes Much of what has been written in the nursing literature regarding participant observation has concerned relations in the ?eld, the role adopted by the researcher and ethical issues (Johnson 1992, Ashworth 1995, Baille 1995, Clarke 1996, Kennedy 1999, Kite 1999).
In addition, there has been an extensive debate in the anthropological and sociological literature about ethnographic writing (Clifford & Marcus 1986, Geertz 1988, Van Maanen 1988) which has been touched on brie?y in nursing (Mulhall 1997). To extend the discussion concerning ethnographic writing would be pro?table but is not possible within the con?nes of this paper. Rather the focus here is on the stage before the writing of ethnography, that is ?eld notes. The lack of historical interest in ?eld notes is surprising considering their central place in observational studies in general and ethnographies in particular.
More recently, however, a number of authors have addressed the issue of ?eld notes, their form, meaning, use and construction (Van Maanen 1988, Sanjek 1990, Emerson et al. 1995). Before exploring the issues raised by these authors it is pertinent to consider what is meant by the ‘?eld’. For the way in which different researchers conceptualize the ?eld will in?uence their assumptions and practices within it. Realists consider that the ?eld represents a natural entity, out there, which needs to be objectively described by the observer, who acts as an impersonal channel through which information is conveyed to the reader.
In contrast, many ethnographers consider that the ?eld is ‘something we construct both through the practical transactions and activities of data collection and through the literary activities of writing ?eld notes, analytic memoranda and the like’ (Atkinson 1992, p. 5). This latter perspective suggests that the practices of researchers within the ?eld – the ways in which they present themselves, collect data, write notes, analyse – will be fashioned both by their particular disciplinary interests and by themselves as people.
This stance acknowledges that what may be considered as important and interesting to write into ?eld notes will be affected by the researcher’s professional and personal worldview. Thus the boundaries of the ?eld are O 2003 Blackwell Publishing Ltd, Journal of Advanced Nursing, 41(3), 306–313 Methodological issues in nursing research Observation in qualitative research ‘the outcome of what the ethnographer may encompass in his or her gaze…and what the ethnographer omits and overlooks as much as what the ethnographer writes’ (Atkinson 1992, p. ). How ethnographers approach the writing of ?eld notes is also dependent on the value they place on them. Some consider ?eld notes to be the very essence of a study. ‘Thus they emphasize writing detailed ?eld notes close to their ?eld observations, mining these notes systematically through qualitative coding techniques, and producing ‘‘grounded’’ analyses tied closely and speci?cally to the original ?eld note corpus’ (Emerson et al. 2001, p. 355).
In contrast, others claim that ?eld notes are secondary to becoming immersed in a culture. They contend that if too much time is devoted to writing detailed notes then the deeper, intuitive experience of being within a culture will be lost. The situation has similarities with the idea that at times it will be important to record speci?c events, to adopt a ‘participating-to-write’ approach (Emerson et al. 1995), and at others it is simply more productive to ‘go with the ?ow’ and become saturated in the experience.
Writing ?eld notes Ethnographers ‘…often experience deep ambivalence about whether, when, where and how to write jottings’ (Emerson et al. 2001, p. 357). The value of making ?eld notes at all has already been discussed. This ?nal section will consider when, where and how ?eld notes may be made. But ?rst it is worth making the observation that the nature of participant observation and the dif?culties in writing conspire to ensure that ?eld notes are messy, loose texts that make no claim to be ?nal or ?xed versions.
Moreover, many would concede that ?eld notes are only comprehensible to their author. This does not cause undue concern in sociological or anthropological quarters. However, it may have implications in the current era of evidence-based health care, for a central activity in this enterprise is the scrutiny of evidence for its validity and reliability. One of the mechanisms suggested in qualitative research to achieve this is the audit trail (Sandelowski 1986) whereby readers are able to follow the actions and decisions made by researchers.
If ?eld notes are incomprehensible to outsiders then assessing auditability may be problematic. However, the provision of an analysis of decisions made, as suggested by Clark (2000), does not seem unreasonable. As Waterman (1998) suggests this could extend to include ‘re?exive validity’ which articulates how the researcher as ‘the instrument’ has affected the direction and focus of data collection. Every ethnographer and every observer will have their own preferred strategies for recording the data from obser- vations.
My own personal schema includes the following types of ?eld note: • Structural and organizational features – what the actual buildings and environment look like and how they are used • People – how they behave, interact, dress, move • The daily process of activities • Special events – in a hospital ward this might be the consultant’s round or the multidisciplinary team meeting • Dialogue • An everyday diary of events as they occur chronologically – both in the ?eld and before entering the ?eld • A personal/re?ective diary – this includes both my thoughts about going into the ?eld and being there, and re?ections on my own life experiences that might in?uence the way in which I ?lter what I observe. When and where Most researchers would agree that it is important to record ?eld notes as closely as possible in time to when events were observed. For some this means writing an account at the end of each day, whilst others attempt to record events as they occur in situ, or by retiring to a discrete location directly following the observation period.
Still others delay the writing of a comprehensive ?eld note until they have left the study site completely. There are merits to all these approaches. Recording events as they happen or shortly afterwards ensures that details, and indeed the entire event, are not lost to memory. On the other hand, shortor long-term re?ection on observations may provide a different gloss on the events. Moreover, speci?cally detailed accounts of events may not be required, and capturing broad patterns may better be achieved at a longer distance from the ?eld. There is also the question of when analysis should begin. Some ethnographers attempt to avoid any explicit analysis during the ?eldwork stage.
However, Spradley (1980) suggests that observations that are only descriptive are both time-consuming and ineffective. He promotes a process whereby data collection must be followed by a period of analysis that leads to more focused ?eldwork. Moreover, it is now recognized that any writing, both in the ?eld andhereafter, is a representation or a construction of events by the author. Thus however, hard we try to be objective such descriptive accounts are theorized, and ethnographers choose to focus on certain activities, key events, and their reactions to them. In this way unconscious analysis of events is constantly occurring as ?eld notes are written.
When notes are made tends to govern where they are constructed. Sometimes it is possible to move around the ?eld with a notebook or computer and record information as it 311 O 2003 Blackwell Publishing Ltd, Journal of Advanced Nursing, 41(3), 306–313 A. Mulhall occurs. However, although participants may accept their actions being recorded directly in one instance, at another time they may object. Additionally, people may develop expectations about recordings and become upset when an event that they consider important is omitted. To a large extent these problems are created and or resolved by the relationship that the researcher has with those in the ?eld.
An ‘insider’ such as a nurse undertaking observation in their own place of work is at an advantage here. They will intuitively recognize situations that are sensitive, embarrassing, or generally ‘off limits’ and thus best avoided, and they are more likely to be trusted than an outsider. Many ethnographers merely jot down phrases or notes to remind them of key events and dialogue that are then written up in more detail in a private space – the garden, sluice, or elsewhere. Although this technique relies on an accurate memory and recall of events, it does avoid some of the problems above and also the issue of con?dentiality, as onlookers may ask to look at notes taken contemporaneously.
This record and notes taken at the time of the event may be further elaborated on at a later date. How Van Maanen (1988) has described three classic genres of ethnographies. Realist ?eldwork accounts are most common and represent the ethnographer as an impersonal channel through which information about the ?eld is conveyed to the reader, interpretation being exercised by the ethnographer alone. In contrast, confessional tales include the researcher’s personal experiences and methodological confessions alongside, but separate from, the descriptive ?eldwork account. However, as with realist accounts, confessional tales still maintain the ethnographer’s authority as interpreter.
The ?nal genre – impressionist tales – attempt to provide such evocative accounts that the reader is quite simply ‘pulled into’ the story to interpret it themselves. Although ?eld notes are not as polished or structured as ?nal ethnographies, they may encompass one or more of the above approaches to writing. These different ways of approaching ?eld notes and ethnographies also raise the question of how far the self should be used within the construction of accounts. Whilst some would not advocate such an approach (Geertz 1988), others such as Ellis (1995) have used their own emotional experiences of ?eldwork to enrich and enlighten their notes and subsequent ethnographies. Ellis projects herself as both the narrator and the main character in the events she observes.
Such experimental ethnographers contend that the subjective aspect of the lived experience is an integral part of that experience and as such it cannot be set aside in a personal diary. Adopting a similar but less radical stance, Rosaldo (1984) demonstrates 312 how life events may reposition the ethnographer. He explains that until the violent death of his wife he was unable to imagine the grief and rage associated with bereavement. Following his own bereavement he was able to gain a different perspective on his data about head-hunters who seek victims following the death of a family member. Experimental ethnographers naturally use the ?rst person in writing their ?eld notes.
Using the ?rst person projects a certain perspective that emphasizes the personal experience of the ethnographer and their particular concerns. Emerson et al. (2001) suggests that writing in the ?rst person is particularly effective where the ethnographer is a member of the group they are studying. This is often the case in ethnographies undertaken by nurses. In contrast, writing in the third person presents a more objective account that concentrates more on others, rather than self. However, as with realist accounts, there is the danger that the researcher assumes an ‘omniscient point of view, assuming privileged access to the character’s thoughts and feelings and motives’ (Emerson et al. 2001, p. 360).
Other practical issues concern how data should be recalled and whether dialogue or action should form the focus. Two strategies are possible for recalling data. One uses the simple chronological ordering of events – this occurred, and that led to another event. Alternatively, during the writing up of notes speci?c critical incidents or exchanges may be related to other similar or contrasting events. Moreover, researchers may either write up events as they happen in real time, remaining ignorant of the future unfolding of events, or they may write from some known endpoint that may construct the earlier events in the light of a subsequent more complete understanding.
Concerning the focus of ?eld notes, Emerson et al. 1995) distinguish between descriptions (or inscriptions) that portray the physical environment, people, actions and smells which make up a setting, and dialogue (or transcriptions) which is the written representation of something that was said. Focusing on both actions and dialogue in a social situation is practically quite dif?cult, but will enrich the subsequent account. As with many aspects of writing ?eld notes, observers can and do combine a variety of these strategies. Although these two matters – how to recall and what to focus on – present practical problems, they also represent particular theoretical stances that the observer may adopt. For example, the intensive use of dialogue may represent a desire by the observer for people within a setting to invoke it in their own words, this being akin to the impressionist tales discussed above.
Dialogue is a very powerful way of presenting data and its use may make a more lasting impression on the reader than descriptive accounts by the researcher. O 2003 Blackwell Publishing Ltd, Journal of Advanced Nursing, 41(3), 306–313 Methodological issues in nursing research Observation in qualitative research Graneheim U. H. , Norberg A. & Jansson L. (2001) Interaction relating to privacy, identity, autonomy and security. An observational study focusing on a woman with dementia and ‘behavioural disturbances’, and on her care providers. Journal of Advanced Nursing 36, 256–265. Johnson M. (1992) A silent conspiracy? Some ethical issues of participant observation in nursing research.
International Journal of Nursing Studies 29, 213–223. Kelly D. , Mulhall A. & Pearce S. 2000) A Good Place to Be If You’re Having a Bad Time: an Ethnographic Evaluation of a Teenage Cancer Unit. Report to the Teenage Cancer Trust, London. Kennedy C. (1999) Participant observation as a research tool in a practice based profession. Nurse Researcher 7(1), 56–65. Kite K. (1999) Participant observation, peripheral observation or apart-icipant observation? Nurse Researcher 7(1), 44–45. Mays N. & Pope C. (1995) Qualitative research: observational methods in health care settings. British Medical Journal 311, 182–184. Merrell J. & Williams A. (1994) Participant observation and informed consent: relationships and tactical decision making in nursing research.
Nursing Ethics 1, 163–172. Moore L. & Savage J. (2002) Participant observation and ethical approval. Nurse Researcher 9(4), 58–69. Mulhall A. (1993) Farewell House, Farewell Home. An ethnography of a residential home for the elderly. Unpublished MSc Thesis, Brunel University, London. Mulhall A. (1997) Nursing research: our world not theirs? Journal of Advanced Nursing 25, 969–976. Mulhall A. (1998) Methods of data collection for quantitative research. In Research and Development in Clinical Nursing Practice (Roe B. & Webb C. eds), Whurr Publishers, London, pp. 135–169. Pretzlik U. (1994) Observational methods and strategies. Nurse Researcher 2(2), 13–21. Rosaldo R. 1984) Grief and a headhunter’s rage: on the cultural force of emotions. In Text, Play and Story: the Construction and Reconstruction of Self and Society (Bruner E. M. ed. ), American Ethnological Society, Washington, DC, pp. 178–195. Sandelowski M. (1986) The problem of rigor in qualitative research. Advances in Nursing Science 8, 27–37. Sanjek R. (ed. ) (1990) Fieldnotes: the Making of Anthropology.
Cornell University Press, Ithaca, NY. Silverman D. (1987) Communication and Medical Practice. Sage, London. Silverman D. (1993) Interpreting Qualitative Data. Sage, London. Spradley J. P. (1980) Participant Observation. Harcourt Brace Jovanovich, London. Turnock C. & Gibson V. 2001) Validity in action research: a discussion on theoretical and practice issues encountered whilst using observation to collect data. Journal of Advanced Nursing 36, 471–477. Van Maanen J. (1988) Tales of the Field: Writing Ethnography. University of Chicago Press, Chicago, IL. Waddington D. (1994) Participant observation. In Qualitative Methodology in Organizational Research: a Practical Guide (Cassell C. & Symon G. eds), Sage, London, pp. 107–122. Waterman H. (1998) Embracing ambiguities and valuing ourselves: issues of validity in action research. Journal of Advanced Nursing 28, 101–105. Thus the when, where and how of constructing ?eld notes have many theoretical as well as practical implications.
These need to be considered both when initiating an observational study and throughout the course of that study. Conclusions To date observation as a method of collecting data in nursing research has been somewhat undervalued, many nurse researchers preferring to use interviews. Part of the explanation for this may be the insuf?cient discussion in the literature of the problems and dilemmas associated with this technique. In particular, little has been written about access to the ?eld and the issues surrounding ?eld notes. Both of these have practical considerations and theoretical concerns that require the attention of the researcher. This paper has discussed these issues.
Ashworth P. (1995) The meaning of ‘participation’ in participation observation. Qualitative Health Research 5, 366–387. Atkinson P. (1992) Understanding Ethnographic Texts. Sage, Newbury Park, CA. Ballie L. (1995) Ethnography and nursing research: a critical appraisal. Nurse Researcher 3(2), 5–21. Booth J., Davidson I., Winstanley J. & Waters K. (2001) Observing washing and dressing of stroke patients: nursing interventions compared with occupational therapists. Journal of Advanced Nursing 33, 98–105. Clark J.E. (2000) Action research. In The Research Process in Nursing, 4th edn (Cormack D. ed.), Blackwell Science, Oxford, pp. 183–198. Clarke L. (1996) Participant observation in a secure unit: care, conflict and control. NT Research 1, 431–440. Clifford J. & Marcus G. (1986) Writing Culture: the Poetics and Politics of Ethnography. University of California Press, Berkeley. Dunn J. (1988)
The Beginnings of Social Understanding. Basil, Oxford. Ellis C. (1995) Final Negotiations: a Story of Love, Loss and Chronic Illness. Temple University Press, Philadelphia. Emerson R.M., Fretz R.I. & Shaw L.L. (1995) Writing Ethnographic Fieldnotes. University of Chicago Press, Chicago, IL. Emerson R.M., Fretz R.I. & Shaw L.L. (2001) Participant observation and fieldnotes. In Handbook of Ethnography (Atkinson P., Coffey A., Delamont S., Loﬂand J. & Loﬂand L. eds), Sage, London, pp. 352–368. Frankenberg R. (1980) Participant observers. In Field Research: a Sourcebook and Field Manual (Burgess R. ed.), Allen & Unwin, London, pp. 50–52. Geertz C. (1988) Works and Lives: the Anthropologist as Author. Stanford University Press, Stanford, CA. Gold R.L. (1958) Roles in sociological field observations. Social Forces 36, 217–223. Ó 2003 Blackwell Publishing Ltd, Journal of Advanced Nursing, 41(3), 306–313