Chapter 20. Action Research
Heather Waterman and Kevin Hope
▪ Introduction: history, definitions and typologies
▪ Characteristics of action research
▪ The practicalities of carrying out action research
▪ Example of action research
▪ Conclusion
Introduction: history, definitions and typologies
There is a complex history to the development of action research (Hart & Bond 1995). No single model can encapsulate the variety of thinking about action research (Elden & Chisholm 1993) and, recognising this diversity, Reason and Bradbury (2001) describe it as a ‘family’ of approaches to research made up of participative, experiential and action-oriented methods.
There are multiple definitions of action research depending on where the author sees action research to be located philosophically. The minority view locates action research within the constructivist interpretivist domain, seeing it as a method for gaining access to subjects’ understandings of their situations. A stronger case is made for locating action research within the domain of critical theory (Carr & Kemmis 1986), recognising its potential as a method for addressing ideological and power-related issues in the social situation as well as providing an impulse for action. Finally, a relatively recent addition sees it located within a participatory paradigm (Heron & Reason 1997) which places emphasis on the experiential knowledge of participants and the collaborative aspects of the method.
For some, this confusion is a source of irritation ranging from a straightforward expression of dissatisfaction with the situation (‘terminological anarchism’ according to Kalleberg (1990)) to a concern that the term might be over-applied in practice leading to a dilution in meaning (Hart & Bond 1995).
Waterman et al (2001, p. 11) define action research as:
… a period of inquiry, which describes, interprets and explains social situations while executing a change intervention aimed at improvement and involvement. It is problem-focused, context specific and future oriented. Action research is a group activity with an explicit critical value base and is founded on a partnership between action researchers and participants, all of whom are involved in the change process. The participatory process is educative and empowering, involving a dynamic approach in which problem identification, planning, action and evaluation are interlinked. Knowledge may be advanced through reflection and research, and qualitative and quantitative research methods may be employed to collect data. Different types of knowledge may be produced by action research, including practical and prepositional. Theory may be generated and refined, and its general application explored through the cycles of the action research process.
Because of the systematic methodology used, this definition encapsulates many of the facets that other authors refer to and is offered as a comprehensive and inclusive definition.
To make sense of the range, authors have identified typologies of action research (Table 20.1). These are offered as ‘ideal types’ (Hart & Bond 1995) which are not intended to be as prescriptive, but attempt to clarify and simplify the complex processes involved.
| Lewin (1948) | Kingsley (1985) | Holter and Schwartz-Barcott (1993) | Hart and Bond (1995) |
|---|---|---|---|
| Experimental Empirical Diagnostic Participative | Management model Autonomy model Underclass model | Technical-collaborative Mutual collaborative Enhancement | Experimental Organisational Professional Empowering |
Holter and Schwartz-Barcott (1993) identify three approaches: the technical-collaborative approach is where particular interventions are tested based on pre-specified theoretical frameworks; the mutual collaboration approach is when ‘the researcher and practitioners come together to identify potential problems, their underlying causes and possible interventions’ (p. 301) and plan change based on understanding of their practice; the enhancement approach aims to increase the closeness between actual problems and theory used to explain these and raise ‘collective consciousness’ to assist practitioners to identify and articulate problems (via critical reflection).
There are problems in trying to group action research in this way. Hart and Bond (1995) suggest that the type of action research engaged in might vary over the life course of a project and vary between or in different cycles. Additionally, a typology implies that action researchers will, at any time, be able to locate themselves within a given frame of reference and ignores complexities of the method. It seems more appropriate, therefore, to avoid falling into the trap of over-categorisation and focus on characteristics of action research. These are outlined in Box 20.1 and considered individually.
Box 20.1
Characteristics of action research
▪ Action research involves participation/collaboration
▪ Action research has a political, emancipatory and developmental role
▪ Action research involves change and improvement
▪ Action research focuses on the practical and is context specific
▪ Action research is cyclical
▪ Action research generates theory
▪ Action research involves reflection and reflexivity
Characteristics of action research
Involves collaboration and participation
Action research is a collaborative process between action researchers and participants or co-researchers. Co-researchers participate in all aspects of the action research; the emphasis is on researching ‘with’ rather than ‘on’ people. Participation has been described variously as a ‘defining characteristic’ (Rolfe 1996), ‘distinctive’ (Elden 1993 and Waterman 2001), ‘essential’ (Wallis 1998) or constitutes a minimal requirement or central position (Carr 1986 and Holter 1993). Brydon-Miller et al (2003) suggest that a shared commitment to democratic change is one of the unifying beliefs of action researchers. Arguably, participation is the key functional element through which democracy is achieved.
Holter and Schwartz-Barcott (1993) point out that participation can be continuous or periodic while Reed (2004) highlights some of the difficulties in achieving and maintaining collaboration. For others, the issue is about how participation and collaboration change over the course of a study with ownership of change shifting towards participants (Wallis 1998); collaboration is not a fixed concept.
Has a political, emancipatory and developmental role
There appears to be consensus that action research is about empowering others because there is analysis of power and its distribution within a given social setting. However, there is variation as to expectations regarding the nature and degree of this, with changes at a structural or organisational level seen by some as a primary aim whereas others tend to focus on the impact on the individual.
An example of the organisational perspective in nursing is demonstrated by Robinson (1995). His starting point is a critical examination of the manner in which power is played out, often in ways unnoticed in daily work, which reinforces inequalities. In contrast, we can turn attention to Lewin’s (1948) original view that raising self-esteem of participants is a goal and Elden and Chisholm’s (1993) position that change should be a self-generating and self-maintaining process. Arguably, change that has an impact on power differentials within a given environment both requires and determines increased self-confidence and esteem of participants.
Is about change and improvement
Important aims of action research are to change and improve. Some argue that change is a prerequisite of action research. For example, Elden and Chisholm (1993) are dogmatic about this, indicating that change-based data is one of five elements that need to be evident for any research to be classified as action research. Others argue that change is not always achievable (Webb 1989) nor indeed an indicator of the success of a project (Waterman et al 1995). Van Mannen (1990) challenges the change assumption and suggests that ‘maybe more significantly action research must learn to deal with what we should have done’ (p. 154). The intention to change or improve needs to be present.
Webb (1989) suggests that action research builds on existing motivation in a working environment and brings with it the authority to change. In addition, nurses operate in a professional context within guidelines and parameters laid down by professional bodies. Rules, regulations, guidelines and codes of practice can serve as orientating or motivational forces which point action research groups in the right direction or which can serve as supportive evidence for defending the route being taken.
Focuses on the practical and is context specific
For nursing, the focus on practice is an attraction of the approach and is reflected in Holter and Schwartz-Barcott’s (1993) assertion that the solution of practical problems is one of its central characteristics. A closely related characteristic is the view that action research is context specific (e.g. Hart 1996, Hart 1995, Hendry 1996 and Lathlean 1994). For Lyon (1998) contextual awareness means that change is more likely to occur because judgement about options available is made with reference to salient aspects including relevant contextual factors. Conversely, critics maintain that, due to its specificity, the generalisability of the research is low and that the subjective influence of participants contaminates the research environment.
Action research is cyclical
The notion of a cycle including planning, acting, observing and reflection was first articulated by Lewin, the central principle being that the floor of subsequent cycles is determined by the ceiling of previous ones. For Hart and Bond (1995), the cycle is seen as a means of interlinking research, action and evaluation. In reality several cycles of activity may occur simultaneously which could influence the direction of the major change. Heron (1996) suggests that this ‘research cycling’ can be conceptualised as having a pruning effect. As a consequence, needless vagueness and ambiguity is reduced but amplification and deepening of the research focus is enhanced. He sees the cycle as having a positive influence on the subsequent validity of a piece of action research.
Theory generation
Another recognisable characteristic is that action research should generate theory (Holter 1993 and Susman 1978). This aspect is seen as something that differentiates action research from other forms of change strategies, audit and evidence-based practice (Holter 1993, Wallis 1998 and Waterman 1996).
Rolfe (1996) argues that the outcome of action research is a highly personal form of knowledge which he terms ‘grounded practice’. For Rolfe (1996):
The researcher-practitioner evaluates a situation, develops a theory to account for that situation, tests the theory by constructing and implementing a clinical intervention, evaluates the new, transformed situation, modifies the theory accordingly and so on… (Rolfe 1996, p. 1317)
He points out that such theory is particular to the clinical setting being studied and, as such, cannot be separated from, or generalised beyond, that setting.
Greenwood (1994) aligns with the perspective that there are generalisable aspects suggesting what should happen in a given situation, all things being equal. Greenwood (1994) appears to be arguing that there is the possibility of generating theory about how change is or is not brought about.
Involves reflection and reflexivity
Robinson (1995) argues that action research is ‘essentially an interactive reflexive process’ (p. 67) which, according to Koch and Harrington (1998), is characterised by ‘ongoing self critique and self appraisal’ (p. 887). There appears to be a potential for the terms reflection and reflexivity to be used interchangeably. With regard to the latter, Waterman (1994) has traced different interpretations ranging from a stated need to understand, reflect and analyse the effects that the researcher has on the research process to one which emphasises the nature of the context specificity of understanding. For Waterman (1994), subjectivity is an inescapable aspect of the research process and reflexivity is the means by which we acknowledge, monitor and understand it.
The practicalities of carrying out action research
Box 20.2 contains the key activities involved in undertaking an action research project.
Box 20.2
Get Clinical Tree app for offline access
Key activities in action research
▪ Selecting participants
▪ Setting up the project
▪ Managing meetings
▪ Clarifying the problem
▪ Fact-finding or reconnaissance
▪ Gaining ethics approval
▪ Research
▪ Critical reflection
▪ Describing and explaining the problem
▪ Planning
▪ Action
