Nurses in Action an Introduction to Action Research in Nursing

REVIEW ABSTRACT Action Research is one o f the new generation o f qualitative research methods in the social sciences which has special significance for nurses in South Africa. The collaborative, participative and reflective qualities o f A ction R esearch appeal to practitioners, and lend themselves to joint problem solving activities in local contexts. This paper sets out a rationale for Action Research, then describes its features, strengths, and limitations. Ways of overcoming the latter are suggested. The paper concludes that Action Research has potential application in the field o f nursing, not only for the purposes o f practical problem solving, but also fo r improving the personal and professional practice of nurses, arui for emancipating nurses from their subordinate position in the hierarchy of health science.

Participating in a creative and recreative process like this develops participants' sense o f p e rso n a l p o w er and s e lf w o rth , a fundamental principle of adult education (Brookfield, 1985;Freire, 1985;Knowles, 1980).The processes involved in Action Research -investigation, action and reflection -assume a commitment to adults participating actively in the world, deciding what they want to learn and the best way to learn i t For these reasons Action Research may be considered an educational process as well as a research method, and the scope of its application is extraordinarily wide (Davidoff et al, 1993;Brooks and Watkins, 1994;Carr and Kemmis, 1986).
WHEN IS ACTION RESEARCH USED? Cohen and Manion (1989) suggest the use of A ctio n R e se a rc h w h e n e v e r s p e c ific knowledge is required for a specific problem in a specific situation, or when a new approach is to be grafted on to an existing system.Contexts in which Action Research may be used effectively include teaching, learning, evaluating attitudes and values, in-service train in g , o rg a n iz a tio n a l d ev elo p m en t, m anagem ent and adm inistration.Action Research may be used for a number of purposes; • to spur individuals and groups to action; • to enhance interpersonal relations, morale and motivation within groups; • to improve professional functioning and efficiency; • to promote organizational change; • to in itiate p la n n in g , p o licy-m ak in g , im()lementation and evaluation; • to introduce innovation and change; and • to develop knowledge and skill for specific and general situations.
In a later section of this paper, the ways in which Action Research has been u s ^ in n u rs in g w ill be d e s c r ib e d , an d th e emancipatory potential of Action Research will be explored.

H O W D O ES A C T IO N R E S E A R C H D IF F E R F R O M T R A D IT IO N A L F O R M S O F R E S E A R C H ?
After the second world war, nursing is said to have copied the experimental approaches of medicine in research, in the same way that education borrowed positivist notions fh)m behaviourial psychology (Lacey and Lawton, in Meyer, 1993).The dom inant research paradigm of the past two centuries, positivism, is based on the belief that valid knowledge can be established only on the basis of what is founded in 'reality ', as perceived by the senses.
Positivist claim that science, and only science, offers an adequate base for research because it alone employs methods which ensure that knowledge is not contaminated by subjective preference and personal bias (Carr and Kemmis, 1986;Reason and Rowan, 1981).Scientific research aims to describe, analyze and explain phenomena, so as to predict and control events.Scientists, or researchers, develop theories about the external world, in which it is assumed that facts are readily o b serv ab le (T o rb ert, 1981).They then conduct pre-planned experim ents, under rigorously controlled conditions, in order to verify their theories.The role o f the researcher is detached, in order to minimize disturbance of the phenomena under investigation.It is assumed that the researcher has superior knowledge of the phenomena at the outset of th e in v e s tig a tio n , an d th a t s u b je c ts ' contribution to the research will be minimal.
Thus the researcher exercises external and unilateral control in gaining information from, and acting upon, others.
The past two decades have seen dramatic shifts in the ways that research is understood, valued and used.The trend in the social sciences has been away from em pirical research based on positivist paradigms and towards 'post-positivist' or 'new paradigm' research (Reason and Rowan, 1981;Carr and Kemmis, 1986).Reason and Rowan (1981) cite eighteen grounds for disapproval of positivist research, including a narrow model of the person, reductionism, a concern with variables rather than with wholes, an emphasis on testing, low actual utilization of results, a preoccupation with bigness and a tendency to serve conservative interests.But the main objection to positivism lies in its authoritarian nature and its rejection of personal and social values in knowledge creation.
Positivist approaches to research are limited when it comes to dealing with human beings in complex situations.Torbert (1981) remarks that the conditions in which knowledge is gained in scientific research -pre-defmed, co n tro lled, un in terru p ted -are not the co n d itio n s faced by p ra c titio n e rs.He observes: Practitioners are generally attempting to act well in situations which they do not fully comprehend, in pursuit of purposes which are not fully explicit and to which their commitment is initially ambivalent, and they are being interrupted all the while by other claims on their attention... what practitioners really require is the kind of knowledge they can apply to their own behaviour in the midst of ongoing events, in o rder to help them inquire more e ffe c tiv e ly w ith o th e rs ab o u t their common purposes, about how to produce outcomes congruent with such purposes, and about how to respond justly to interruptions.(Torbert, 1981: 143) Rooted in these objections to positivism, alternative approaches to research have e m e rg e d in r e c e n t y e a rs, in c lu d in g phenomenology, ethnography, naturalistic e n q u iry and sy m b o lic in te ra c tio n is m (Rosenwald and Ochberg, 1992;Carr and Kemmis, 1986).These approaches, which are based on the epistemologies of interpretive social science and critical theory, stress the importance of understanding, meaning and action on the part of individuals and groups.
The central proposition of such qualitative approaches to research is that knowledge of our social world is not something hard, real and tangible to be identified, manipulated and v alid ate as in a positivist model; it consists of a complex web of meanings created and sustained by individual members of society.These meanings can only be interpreted by reference to peoples' motives, intentions and purposes.Because knowledge is social, its meaning should be interpreted in social settings and traditions.

FIGURE 1: THE ACTION RESEARCH SPIRAL
The purpose of qualitative research is to allow researchers to enter into another person's perspective, or to gain new perspectives on th e ir ow n ex p erien ce.T he q u alita tiv e researcher begins with the individual, trying to understand the meaning she ascribes to certain actions or events.A theoretical account (or interpretation) of this action is developed and made available to others, revealing to them the rules and assumptions upon which they have been acting.This account 'enlightens' or 'illuminates' the significance of their actions (C arr and K em m is, 1986).By offering individuals and groups an opportunity to examine the beliefs and attitudes underlying their actions, they are invited to consider alternative ways of believing and acting, and to create new relationships and practices.
To achieve their aims, qualitative researchers docum ent episodes of human experience representing as closely as possible how people feel, what they know, and what their concerns, behefs, perceptions and understandings are.F avoured m ethods in clu d e in terv iew s, participant observation and unobtrusive measures, usually conducted in a natural setting with small, non-random samples.Data are usually analyzed inductively, resulting in explanatory themes and concepts.Results arc documented in natural language, often in the form of case studies.
Action Research involves a four-stage spiral of steps: strategic planning, acting, observing and reflecting.
It is a cyclical, recursive process whereby a plan of action is reformulated on the basis of the inform ation collected, then shared, discussed, recorded and evaluated (Cohen and Manion, 1989).Cunningham (1976) As its name suggests.Action Research aims to integrate research and action.It seeks theories which will inform -but not determinepractice.On the one hand.Action Research rejects scientific rigour and precision in favour of practical effectiveness; on the other hand it rejects mere activism by insisting on scrutiny of practical judgem ents.The reciprocal shaping of theory and practice central to Action Research is captured in the term 'praxis', which emphasizes the theoretically informed and committed nature of action, and the need for theory which i s "... open-ended, nondogmatic, informing, and grounded in the circumstances of everyday life" (Lather, 1986: 263).The integration of theoiy and practice is e ssen tial for the tw in g oals of A ction Research, empowerment and change, and has a strong bearing on the process of improving practice.
Action Research is highly regarded as a means of in-service training (Cohen and Manion, 1989) in which practitioners acquire new skills and methods, sharpen their analytic powers, and h eig h ten th e ir aw a re n ess o f th eir interventions in the context of everyday life experience.The development of such qualities should lead to improved practice, an important aim in nursing in South Africa.Action Research offers a number of opportunities for improvement of professional practice: • Participants acquire practical experience of research.They learn to formulate plans for action, examine their practice and reflect c r itic a lly on th e p o s s ib ilitie s and constraints of their situation; • The research process is demystified as group m em bers becom e involved in actively directing it rather than passively submitting to it; • C o m m u n ic a tio n s b e tw e e n n u rse s, researchers, support staff and patients participating in research are facilitated; • T he au th o rity ro les w hich separate researchers and research subjects, nurses and doctors, and nurses and patients are broken down.Participants are engaged as learners, and becom e committed to a process of self-education; • It makes good use of the resources in a group; and • Participants develop problem solving skills, as well as interpersonal and group interaction skills.
As in the field of education, there is currently an emphasis on reflection-in-action in nursing to produce different professional knowledge m ore a p p ro p ria te to p ra c tic e (M eyer, 1993:1067).A ction R esearch, with its o p p o rtu n itie s to ex p lo re and im prove professional practice, seems to offer several advantages in achieving this aim.Schon explains that "thinking-on-your-feet" (1983; 54-5), a process of reflecting on and in action, is the core of practice.As participants begin to make sense of the action, they reflect on the understandings which have been implicit in their action.New understandings surface, are criticized, restructured, and embodied in f u rth e r ac tio n .T he p ro c e ss feeds on continuous critical reflection, which may lead to a change in practice or raise new issues and standards for practice (Faulkner and Swann, 1993:246).
A c tio n R e se a rc h ta k e s p la c e in th e complicated climate of an ongoing simation.
T his m akes it highly unlikely that the researcher will know in advance the exact pattern of the enquiry which will develop.The definition of the problem, the questions to be asked and answered, the methods to be employed, are likely to be modified during the course of the research.As findings are validated or invalidated in practice, new questions or solutions are suggested by the developing situation.Participants in Action Research need to investigate their problem and its context carefully, and repeatedly, in order to gain the theoretical insights which will guide and correct their practice.This way of conducting a social enquiry is a mutually educative enterprise.
In order for the Action Research Process to be innovative and developmental, rather than merely repetitive, critical reflection must feature in the process.Critical reflection may be interpreted in at least two ways.It may refer to a se lf-e v a lu a tiv e p ro c e ss am o n g st participants (which might be considered 'learning from experience' in a practical context).It may also be considered part of a 'critical theoretical' process (Brookfield, 1985;Freire, 1985) which aims not only to understand, but to change practices by clarifying the values and understandings of those involved in activities, and those who benefit from their continuation.Critical theory urges participants to study the nature and consequences o f actio n ; the nature o f institutions and structures providing the framework for action; and the nature of the action itself.This 'critical' knowledge is the base for transform ing or reconstructing practice, and indeed society as a whole.
Action research has been accused of being a form of 'naive enquiry', prone to error (Cohen and Manion, 1989;Lather, 1986).The charge of naivety apparently stems from Action Research's rejection of positivist assumptions about the nature of reality, the authority and control exercised by the researcher, and the appropriateness of experimental methods in uncovering 'the truth'.'Errors', on the other hand, may stem from the researcher's biases, prejudices and anxieties, or from pressures for group conformity.One implication of this for Action Research is that the design of the research project will have to go beyond the predisposition of group members (or the host organization) to establish a trustworthy basis for problem-solving, decision making and the creation of new ideas.

What validity criteria would serve Action
Research best?Lather believes that "we must form ulate self-corrective techniques that check the credibility of data and minimize the distorting effect of personal bias upon the logic of evidence" (1986: 270).She suggests a "vigorous self-reflexivity", incorporating concepts of validity taken from traditional research (such as convergent and contextual validity; catalytic validity and internal validity) but in a revised and expanded form.
Reason and Rowan (1981) make several practical suggestions for increasing validity in qualitative research, including: • In v o lv in g p e o p le w ith d iffe re n t backgrounds and perspectives in every aspect o f the project, increasing the likelihood that the researcher will not always be supported, but also challenged and confront«l; • Using a variety of methods for collecting data (for example interviews, observation, nominal group technique, diary notes, c r itic a l in c id e n t te c h n iq u e ) to get information from a range of primary and secondary sources; • Employing 'feedback loops', a process of feeding back tentative conclusions to participants for them to mull over and redefine; and • Developing high quality awareness on the pan of co-researchers through systematic methods of personal and interpersonal development.
Several writers (Walker, 1993;Winter, 1987) h av e q u e s An additional but related problem is that the results of Action Research tend to be kept local; they are not shared or com m unicat^ with others.This sometimes happens when the research focuses too much on outcomes (so lu tio n s to problem s) rather than on exploring and reporting the process that leads to them.The scope o f the research then becomes too local and specific; there is little in it to interest or benefit the wider society.
In Action Research problems are analyzed in a specific context and solutions are sought in that context.The emphasis is on gaining knowledge for a particular situation and purpose rather than on obtaining generalizable scientific knowledge.This specificity has led to charges that Action Research is too local and contextualized, and that it promotes activities of an incidental and 'ad hoc' nature (Cohen and Manion, 1989).However, the process of action research, based on principles of consultation, cooperation and participation, should be relevant to practitioners and researchers more widely, particularly in South Africa with its current climate of rapid and unpredictable social change, dissatisfaction with existing m odels and services, and emphasis on negotiated solutions to political and other problems at all levels.
The point has already been made that Action Research has the potential to produce mutual learning, new knowledge and solutions to specific problems.Control of the research design and p rocess is sh a re d am ongst participants, implying that research goals and procedures must be understood and approved by all involved.Critics of Action Research have suggested that such principles are in h e re n tly flaw ed.T h ey m a in ta in that participants' capacity to diagnose problems, implement procedures and evaluate their own actions is often limited.Participants may also lack the authority, prestige or power to secure changes in their situations.Meyer (1993) points out that in the field o f nursing, participants, who may be patients and their families, cannot give truly informed consent w h en th e n a tu re o f th e p ro c e s s is undeterm ined.The changes proposed in Action Research may be threatening, as may be the nature of the collaborative relationship.
Issues such as equality, being in receipt of confidential information (with resultant power and vulnerability), difficulties of maintaining confidentiality and anonymity and so on, may all impact on the capacity o f group members to participate fully in problem-solving and strategic planning.In addition, the pressure of e x te rn a l e v e n ts, and th e to le ra n c e o f participants, may affect the research process.
The existence of such factors needs to be probed at the outset of an Action Research e n q u iry , an d th e ir p o te n tia l im p a c t acknowledged in the research design.
Finally, Cunningham (1976) and W inter (1 987) have o b serv ed th at the A ction Research process has no terminus.The cycles of plaiming, acting, reflecting and replanning continue as the problems to which they are directed change.Action Research can begin anywhere, and once begun, is without a principle for completion.In this respect, it may be decided to set an artificial limit to activities.
It is important to remember, however, that the enquiry may not be finished when the project ends and the report is written.
ACTION RESEARCH AND NURSING Titchen and Binnie (1993) suggest that Action R esearch could be used to fu rth e r the following interconnected aims in nursing; • To help practitioners research and improve their own practice by generating and testing theory; • To in tro d u c e c h a n g e an d fa c ilita te innovation in professional practice; • To facilitate professional learning and reflective practice; and • To democratize health care through the emancipation of nurses from the nursing hierarchy and the traditional role of doctor's handmaiden.
According to Titchen and Binnie (1993)  The main aim of nursing is to effect positive change in the health status of people.Meeting this aim requires appropriate action theories, whether or not the practitioner is aware of them.As Yura and Walsh observe.
T he n u rsin g process is an orderly, systematic manner of determining the client's problems, making plans to solve them, initiating the plan or assigning others to implement it and evaluating the extent to which the plan was effective in resolving the problems identified.(In Burgess and Ragland, 1983: 45) This process sounds remarkably like the A ction Research cycle described above.
Nurses have followed these steps to address areas of pain, sleeplessness, wound healing and urinary tract infection (Tolley, 1995).
They have used Action Research to evaluate painful emergency procedures, orthopaedic procedures and bums (Robinson, 1995).In general nursing care.Action Research has been used to promote more patient-centered n ursing, increasing the involvem ent of patients' families and friends in patient care prior to discharge (Titchen and Binnie, 1993).Action research has also been used to address indiv id u al p atien ts' problem s from the p atien t's perspective (M cG arvey, 1993;Meyer, 1993).Critical Action Research has the potential to emancipate nurses from their entrenched subordination.As Lather describes it, ... analysing ideas about the causes of powerlessness, recognizing systematic o p p ressiv e fo rces, and acting both individually and collectively to change the conditions of our liv es... Empowerment is a process one undertakes for oneself; it is not something done 'to ' or 'for ' someone. (1992; 4) T he se lf-reflectio n and self-ev alu atio n u n d erp in A ction R esearch process as practitioners leam to look beyond established behaviours and traditions to develop deeper insights into the nature of nursing, and into the structures, processes and politics which control and shape the field.Stark (1994) describes how self-reflection gave her the courage and skills to confront other issues w h ich a ffe c te d h e r p ra c tic e as h er s e lf-a w a re n e s s in c re a s e d .S tark and G re e n w o o d (1 9 9 4 ) cla im th a t the self-evaluative nanire of the Action Research process increases flexibility and adaptabihty, and develops nurses' practical reasoning, analytical and decision-making skills.

CONCLUSION
The last two decades have seen dramatic shifts in the ways in which research is understood, used and valued.The trend in the social sciences has been away from positivist research based on a natural scientific model, and tow ards a 'p o stp o sitiv ist' or 'n e w paradigm research', characterized by a more qualitative, human and flexible approach to knowledge creation.This paper has advanced the cause of Action Research in nursing, on the grounds that it is interactive, in the sense that theory and practice interact; contextualized, in the dual physical and social worlds of nursing; and compelling, in th e sense that it in v ites participation in the research process, and in the sense that it helps participants understand and change their worlds.
T he w rite rs ack n o w led g e th a t A ction Research is not a flawless or unrivalled route to the creation of knowledge.What Action Research has to offer is a fram ew ork for challenging and changing nursing practices, and for understanding the process of doing this.It is the writers' contention that such a framework is appropriate and useful in view of the changing context of nursing in South Africa.