Nursing Power and Social Judgement

An Interpretive Ethnography of a Hospital Ward

Martin Johnson

• Publisher: Taylor & Francis
• Year: 2018
• Print ISBN: 9781138330757
• eBook ISBN: 9780429825378
• Edition: 1
• Topic: Social Sciences
• Subtopic: Social Work
• Index: Social Sciences

First published in 1997, this work makes a substantial reexamination of the social processes behind the labelling of patients in hospital care. Taking an interpretive perspective, the author analyzes the social construction of patient labels identifying strategies for and the consequences of giving and receipt of 'good' and 'bad' labels. He shows how the rich data of truly participant observation in the tradition of reflexive ethnography can powerfully illuminate the experiences and actions of both patients and their nurses. It is a critical analysis of key work in this field. Professor Johnson demonstrates the redundancy of trait theories of social judgment, offering a more complex and negotiated reality in which patient labels form a part of a rich web of unequal power relations between nurses and their clients.

1 Research perspective

A good deal has been written about research perspectives in qualitative research (Schwartz and Jacobs, 1979; Silverman, 1975; Denzin,1992). Various approaches have been identified, and much debate takes place about definition and emphasis of particular traditions. For example, the terms phenomenology, symbolic interactionism, ethnography, feminist methodology and even ethnomethodology seem to be used frequently and with a good degree of overlap. What can be said with some degree of confidence is that these approaches together constitute a reaction to, if not an attack upon, the cultural and scientific dominance of positivism, particularly in Europe, America and more recently Australasia.

Positivism

Positivism can be seen as the belief that both natural and social phenomena can, given the right developments in methods of classification and measurement, be recorded as facts. That is to say that there is assumed to be an 'objective reality' which can be identified, and which persists (Hammersley, 1992). Generally speaking positivist methods involve the production of 'hypotheses' which the method sets out to verify or refiite. These hypotheses are often (though this is not always stated) deduced from extant theory in the social, behavioural or biological sciences (see for example Boore, 1978). Every attempt is made within the positivist tradition to remain 'value free', or neutral, and any attempt to record mediating social processes or meanings may be seen as inappropriate. Writing as ethnographers but in order to give some flavour of what this can mean for the observation of the social world, Hammersley and Atkinson (1983:4) suggest that:

Scientific theories must be founded upon - tested by appeal to descriptions that simply correspond to the state of the world, involving no theoretical assumptions and thus being beyond doubt. This foundation could be sense data, as in traditional empiricism, or as with later versions, the realm of the 'publicly observable': the movement of physical objects, such as mercury in a thermometer, which can be easily agreed upon by all observers. Because observation in the social world is rarely as straightforward as reading a thermometer, this concern with a theoretically neutral observation language has led to great emphasis being given to the standardisation of procedures of observation. (Hammersley and Atkinson, 1983)

Among the many accounts of the arguments against such a position for social scientists, the most detailed exposition I have encountered is that of Denzin (1992). Drawing upon a rich historical account of the ideological and intellectual influences on qualitative researchers, Denzin explains some otherwise subtle differences between the various schools of interpretive sociology and their reaction to positivism. Hammersley and Atkinson (1983) explain the general thesis by arguing that non-positivists claim that the social world cannot be understood in terms of causal relationships or universal laws. This, they suggest is because human actions are based upon, or 'infused by', social meanings such as attitudes, motives, intentions and beliefs. They give as their example the assumption in much positivist social research that a stimulus-response model explains social behaviour without considering mediating processes. In the view of interactionists and other non-positivists:

...people interpret stimuli, and these interpretations, continually under revision as events unfold, shape their actions. (Hammersley and Atkinson, 1983:7)

Hammersley and Atkinson continue with the suggestion that stimuli can mean different things to different people and indeed to the same person at different times. This point is one of those crucial to the understanding of what I take to mean the interactionist perspective, and it turns out to be central in the understanding of the social process of evaluation or judgement of character in the context I describe.

Are qualitative and quantitative complementary?

There has been a good deal of argument concerning the merits of qualitative and quantitative research, and some views seem to be in favour of some sort of reconciliation (Leininger, 1985; Corner, 1991). For some time it was fair to say that feminist methods were broadly allied to the qualitative domain. Reinharz (1992) extensively documents this situation but concedes that in the post-modern era there may be strategic if not empirical benefit in multi-method research drawing upon both the main paradigms. Recent studies claiming a feminist perspective (see Smith, 1992) combine both approaches in some degree although sometimes the mixture can seem ill-matched. I am, however, left with a sense of unease that the two paradigms involve such substantial difference of beliefs about the nature of knowledge and the conduct of research that the nature of their compatibility requires a good deal of clarification if they can co-exist at all sensibly in the same study.

Symbolic interactionism

In this project, both in the early phase in which I produced discussion papers on method and some epistemological concerns (Johnson, 1990, 1992), and more recently during data collection and analysis, I have inevitably drawn guidance from a range of viewpoints on research perspective. The most influential of these is probably what I would broadly term symbolic interactionism and specifically the Straussian 'negotiated order' perspective.

According to Denzin (1992), who has been an influential worker in this tradition for many years, interactionism grew largely from the influence of anthropology and phenomenology in the 1920's in the USA. Many of what later became known as the Chicago School of sociologists were, however, either European emigres or of that extraction. Denzin's detailed history gives a good deal of importance to those writers such as Goflman (1959) and Strauss (see Glaser and Strauss, 1967) who consistently set out to refute the verificationist (positivist) position which was dominant in sociology at that time. Goflman's dramaturgical model, taking many of its constructs explicitly from a conception of social life as acting or drama, threw a fascinating light on the otherwise mundane and commonplace. It is impossible, as can be seen from my account of my experiences in Howarth ward, not to draw considerable insight from his work. Goflman's notions of moral career and inmate have a special relevance to my own discoveries, even when analysed through contemporary eyes.

As to the late Anselm Strauss (he died in 1996), his influence has been immense. He and Barney Glaser developed the now widely used analytic approach of grounded theory which is based firmly in assumptions about the value of an interpretive approach to human action (Glaser and Strauss, 1967). Theory generated by this method tends to emphasise processes rather than causes or outcomes although these latter help to expand the theory. Denzin (1992) suggests that Strauss's conception of interactionism stresses indeterminacy and danger in everyday life. Here the concept of negotiation comes into its own as a frame of reference for the discussion of all social processes. It makes explicit the notion that no social phenomenon is permanent or wholly determined by specific traits or variables. This seems eminently plausible. Indeed despite awareness of many limitations of my own analysis I remain concerned at the degree to which investigators claim to have utilised interactionist insights and methods such as grounded theory, but have clearly failed to grasp this fundamental point of negotiated social order which is integral to the approach (examples are Lorber, 1975; Stockwell, 1972 and Wilkinson 1991 whose work I examine in some detail later).

Without wishing to exalt either individuals or their work, especially when they themselves deprecate that trait among their opponents the positivists (see Glaser and Strauss, 1967), the extent to which I draw upon the work of Strauss and a large number of his colleagues will become clear over the course of this thesis. To the extent that an interactionist perspective does then inform and guide my method and my analysis.

Phenomenology and other perspectives

Among many differences of interpretation, three clear positions upon what phenomenology is stand in relation to the rest. On the one hand some American writers such as Lynch-Sauer (1985) see phenomenology as a specific method of qualitative research which contrasts with grounded theory or ethnography. She argues that the phenomenological method attempts to describe the specific 'lived experiences' of individuals with a minimum of interpretation by the researcher. Benner (1984), whose work is immensely influential, suggests another version of phenomenology which allows of more interpretation by the researcher. She claims that her analysis of nurse's progression from novice to expert status was drawn out using an approach she says derives from Martin Heidegger, but recently these claims have drawn substantial criticism. Cash (1995) for example, argues that Benner's basic Dreyfus and Dreyfus (novice to expert) framework pre-existed the investigation and that Benner has merely fitted her data into it. Perhaps more serious is the sustained attack by Crotty (1996) in which he shows that Benner and her followers have failed to grasp the fundamental tenet of phenomenology: that it is phenomena rather than 'subjective lived experience' or 'narrative' which should be the main object of investigation.

It is not my wish to sustain confusion in these matters, but I think, however, that yet a third position has much to commend it. Among ethnographers such as Bogdan and Taylor (1975) and more recently Hammersley and Atkinson (1983) the most consistent position adopted is that phenomenology represents the paradigm, or large domain in which other non-positivist approaches lie. It is a convenient opposite to positivism, one which, broadly speaking, accepts multiple definitions of social reality with which social scientists in general and nurses in particular are mostly concerned. My study therefore, could be defined as phenomenological. This would, however, give no real indication of the types of concepts I am likely to find useful, or indeed of the method I am likely to use to collect and analyse my data.

Other perspectives with which I would claim to be associated would be the following. First, there is an important choice to be made in how data are not only to be collected but also written up. Amazingly, some accounts of this type of phenomenological or naturalistic research are still written in the 'objective' style of the third person singular. As Webb (1992) argues, this is inconsistent with any acknowledgement that the researcher is both present and has an unavoidable effect upon the setting. It is therefore more honest, Webb argues, if the presence of the researcher in the setting is explicit and when the appropriate mode of writing is used. This is not a minor point, since language (broadly defined) is a very powerful, if not the only, method through which meanings are shared in the social world. This point of Webb's is amply supported by Porter (1993). It is, as Webb (1992) observes, only in nursing literature that a pseudo-objectivity has been required in some of the academic journals and even then not consistently so. Indeed she argues further that even in positivistic studies to write as if the researcher had no human part to play in proceedings is misconceived and could even be seen as deceptive.

Reflexivity

Reflexivity is seen by Steier (1991) as "a turning back of one's experience upon oneself'. I attempt in this book to illustrate my reflexivity by discussing my perceptions of my participation in the setting. Of course the inclusion of myself as a participant makes the research semi-autobiographical. Some passages are highly personal, and whilst this can seem self-indulgent, with an appropriate balance it can represent researcher integrity and awareness. Such is for the reader to judge, but at least there is no attempt to conceal my effect on the setting even if I am unable to describe it accurately or in its entirety.

Hammersley (1990) identifies several modes of writing which appear in varying balance in qualitative research. Naturalism is that in which pure description occurs with relevant theory. Legitimatory mode is like this section, in which purpose and methods are explained. Confessional mode is the 'inside story' in which an inappropriate balance can seem too journalistic or indulgent. Such matters may also be aesthetic and so experimentation may be necessary. I believe that I utilise at least these three modes of writing, and that the humanistic orientation of my research perspective should be evident from the text. Reflexivity involves an awareness by an investigator of the degree to which they are making their own interpretation of events which might, indeed inevitably, would be portrayed otherwise by other writers. It involves reflecting constantly upon events as they unfold, and responding by following up notions which emerge as important in fieldwork rather than from preliminary reading.

Ethnography

I define ethnography, following Hammersley and Atkinson (1983), very broadly indeed. It consists of study of the social life, social world, or other aspects of the culture of a particular group or social setting. The methods and perspectives are commonly qualitative and phenomenological but not exclusively so. The word does however commonly assume a 'participant observation' approach, although as I outline later some investigators (although I am disinclined to agree with them) are prepared to classify interviews as a form of participant observation. Ethnography then, constitutes both a method (mainly involving fieldwork) and could also be said to be a perspective. In using the term ethnographic perspective, the aim would be to convey, in a relatively shorthand form, a sense of acceptance of a number of assumptions about research activity and the derivation of knowledge. These would include the use of naturalistic or fieldwork methods, interactionist approaches to analysis of data such as analytic induction or grounded theory, and an attempt to share in the social world or culture of the informants in order to elucidate the meaning of the actions themselves. It would also perhaps mean a perspective in which epistemological subtleties are seen to be less important than the pursuit of naturalistic data from the field. Both Hammersley and Atkinson (1983) and more recently Layder (1993) could be said to represent this tradition. The point about epistemological subtlety is not misplaced when the view of Anne Williams (1989) is considered. She amply illustrates the contextual nature of method and ethics whatever perspective is assumed in her description of the 'everyday messy life of the ethnographer' (p329).

Summary

I have provided an overview of the influences and specific perspective which have given me guidance in this work. To return briefly to the notion of epistemological pluralism, I would argue that many events in the natural sciences are amenable to investigation by means of rigorous measurement and the analysis of nominal data by means of the wide range of inferential statistics available. I would even accept that in the social and behavioural sciences, a good deal can be learned from the considered use of inferential statistics where concepts are well defined, samples are rigorously derived and of appropriate size.

I remain sceptical, however, that much that counts as quantitative nursing research meets these criteria and rather, statistics are used to add, just as with the third person singular style, a spurious objectivity to studies which attempt too often to quantify the unquantifiable. In these circumstances therefore, I have little confidence in epistemological pluralism if it renders studies incoherent as a result of the necessary reduction in size and energy used for either the qualitative or the quantitative analysis. As for this study, I claim to follow broadly a symbolic interactionist perspective. As may be evident, however, the influences of conflict positions and the social philosophy of Foucault (1991) have had an increasing part to play in my development over the course of this work.

2 Analytic approach

I have suggested that I will be working from a broadly symbolic interactionist perspective in the collection and interpretation of data. Within this tradition there remains a good deal of debate about the merits of various approaches to analysis and collection of data. An important point is that early anthropologists and ethnographers almost certainly gave little thought to such procedural matters. They made detailed notes of their observations, often in the form of a diary, and then wrote a description of events and phenomena as they saw things. Sometimes and where it was possible to do so, they would interview informants informally to check out their interpretation. The emphasis was on description of the social context they studied, and little or no claim was made to the generation or testing of formal theory. William Foote Whyte's (1955) Street Corner Society used this approach. Dingwall (1977) notes that accounts such as this were always open to the criticism, notably from positivistic social scientists, that such descriptions were 'mere journalism' and lacked credibility in any scientific sense.

Probably in order to defend qualitative research from such criticism, substantial energy was put by some workers into developing procedures for analysis {and collection) of data which would at least demonstrate how interpretations were derived. Probably because research reports were commonly being evaluated by the criteria accepted for scientific or positivistic research, attempts were often made to satisfy aspects of these criteria. Very large sample sizes were sometimes used even by noted Chicago School interactionists in an attempt to give the appearance of representativeness and therefore generalisability (e.g. Hughes', 1958 Twenty Thousand Nurses Tell Their Story).

A number of systematic approaches develope...

Table of contents

• Cover
• Dedication
• Title
• Copyright
• Contents
• Tables
• Introduction
• 1 Research perspective
• 2 Analytic approach
• 3 Participant observation
• 4 Access and ethics
• 5 Good and bad patients: a review in depth
• 6 Judging people and assessing
• 7 The climate of social judgement
• 8 Negotiating
• 9 Struggling
• 10 Acquiescing
• 11 Aspects of caring and coping
• 12 Conclusion: a theory of social judgement
• Bibliography
• Index