Action Research: Comparison with the concepts of ‘The Reflective Practitioner’ and ‘Quality Assurance’.

© Regina Hatten, Donna Knapp and Ruth Salonga, 1997.


As a group of post graduate students involved in Action Research as an elective within a Master of Health Science Education degree, we were surprised by commonalities observed between AR and two other concepts with which we have had previous experience. Exposure to the concept of ‘The Reflective Practitioner’, as identified by Schon, occurred within the course work component of our Master in Health Science Education degree. While the concept of Quality Assurance/Total Quality Management was identified within our organisational settings.

Through group discussion, we subsequently decided that exploration of the three concepts identified would assist us to clarify and compare them. We decided that documentation of our findings in an electronic publication would provide a concise form of information regarding these issues for others who are new to this field.

 What is Action Research (AR)?

Action research (AR) is a non-traditional form of research which is often community-based and carried out by a practitioner in the field (Stringer, 1996, p.9). The linking of the terms ‘action' and ‘research' highlight the essential feature of this approach, which involves the testing out of ideas in practice as a means of improvement in social conditions and increasing knowledge (Kemmis & McTaggert, 1988, p.6).

AR, as described by Lewin, proceeds in a spiral of steps composed of planning, action and an evaluation of the result of the action. The AR process begins with a general idea that an improvement or change in the practitioner's area of work is desirable. A group then forms to clarify the mutual concern which has been identified. The group makes the decision to work together and focus its improvement strategies on the 'thematic concern' (Hart & Bond, 1995, p.54; Kemmis & McTaggert, 1988, p.8-9).

In the AR spiral, (refer to Figure 1) group members:

  1. develop a plan of critically informed action to improve current practice. The plan must be flexible to allow adaptation for unforeseen effects or constraints;
  2. The group members act to implement the plan which must be deliberate and controlled;
  3. This action is observed to collect evidence which allows thorough evaluation. The observation must be planned and a journal may be used for recording purposes. The action process and its effects within the context of the situation should be observed individually or collectively;


  4. Reflection of the action recorded during observation is usually aided by discussion among the group members. Group reflection can lead to a reconstruction of the meaning of the social situation and provides a basis for further planning of critically informed action, thereby continuing the cycle. These steps are carried out in a more careful, systematic and rigorous way than that which usually occurs in daily practice (Kemmis & McTaggert, 1988, pp.10-14; Zuber-Skerritt, 1992, p.16).
Figure 1 - The Action Research Spiral

(after Zuber-Skerrit, 1995, p.13)

The upward direction of the spiral IN Figure 1 indicates a continuous improvement of practice and an extension of personal and professional knowledge (Zuber-Skerrit, 1995, p.13).

 Characteristics of Action Research

There are a number of characteristics which distinguish AR from other forms of research. These include: collaboration between researcher and practitioner; solution of practical problems, change in practice; theory development and publicising the results of the inquiry (Holter & Schwartz-Barcott, 1993; Zuber-Skerritt, 1992, p.14).


The focus of collaboration involves interaction between a researcher or research team and a practitioner or group of practitioners. The ‘practitioners' are individuals who know the field or workplace from an internal perspective regarding the history of the workplace development, knowledge of how others in the setting expect things to be done and knowing how things are usually done. The ‘researcher' is an outsider who has expertise in theory and research but limited knowledge regarding the local setting. The collaboration between the two parties can vary from periodic to continuous collaboration throughout the study (Hart & Bond, 1995, p.55; Holter & Schwartz-Barcott, 1993).

The researcher may not be an outside expert and is often viewed as a ‘co-worker' doing research with and for the practitioners. This group may be broadened to directly involve as many people as possible who will be affected by the practices concerned (Kemmis & McTaggert, 1988, p.23; Zuber-Skerritt, 1992, p.13).


The primary purpose of AR is as a tool for solving practical problems experienced by people in their professional, community or private lives (Stringer, 1996, p.11). The problem is defined in relation to a specific situation and setting determined by the group, community or organisation. A variety of data collection methods can be used to identify the problem, eg. observation, interviews, questionnaires (Holter & Schwartz-Barcott, 1993; Stringer, 1996, p. 9).

Change in Practice

The results and insights gained from the AR should not only be of theoretical importance but also lead to practical improvements in the problem areas identified (Zuber-Skerritt, 1992, p. 12). The change in practice will depend upon the nature of the problem identified (Holter & Schwartz-Barcott, 1993).

Theory Development

An important goal of AR is that the results assist the researcher to develop new theories or expand existing scientific theories ( Holter & Schwartz-Barcott, 1993). Through the process of AR, practitioners are able to develop a reasoned justification for their work. The evidence gathered and the critical reflection which occurs help create a ‘developed, tested and critically-examined rationale' for the practitioner's area of practice (Kemmis & McTaggart, 1988, p.25).

Public results

The theories and solutions which are produced from the AR should be made public to the other participants and also to the wider community who may have an interest in that work setting or situation (Zuber-Skerritt, 1992, p.14).

The characteristics of AR are neatly summarised in the CRASP model developed by Zuber-Skerritt.
Action research is:
Critical collaborative enquiry by
Reflective practitioners being
Accountable and making the results of their enquiry public,
Self-evaluating their practice and engaged in
Participative problem-solving and continuing professional development (Zuber-Skerritt 1992, p.15).

 Types of Action Research

There are three main types of AR - technical, practical and emancipatory.


The goal of this type of AR is the testing of an intervention based on a pre-specified theoretical framework. The researcher is questioning whether the selected intervention can be applied in a practical setting (Holter & Schwartz-Barcott, 1993). The researcher acts as an outside expert who aims to gain the practitioner's interest in the research, and agreement to assist in the implementation of the intervention (Holter & Schwartz-Barcott, 1993; Kemmis & McTaggart, 1988, p.12).


This type of AR involves the researcher and practitioner coming together in order to identify potential problems, underlying causes and possible solutions or interventions. The researcher encourages participation and self-reflection of the practitioner (Holter & Schwartz-Barcott, 1993; Kemmis & McTaggart, 1988, p.12).


This type of AR involves all participants equally with no hierarchy existing between the researcher and practitioner. The researcher aims to decrease the distance between the actual problems identified by the practitioner and the theory used to explain and resolve the problems. The researcher also facilitates reflective discussion with the practitioner to identify underlying problems and assumptions. This assists the researcher to become a collaborative member of the group (Holter & Schwartz-Barcott, 1993; Kemmis & McTaggart, 1988, p.12).

In conclusion, AR is an alternative social science research approach which aims to link theory and practice in solving practical problems for practitioners in the field.

What is ‘The Reflective Practitioner’?

Donald Schon has provided an individual, self-directed, experience-based professional learning and developmental process for the practitioner with the concept of the reflective practitioner. These practitioners have incorporated into their repertoire of skills, the art of transformative learning through reflection (Schon, 1991; Mezirow, 1991).

This concept represents Schon’s interpretation of the developmental path and characteristic of professional expertise, which had previously been defined by using the traditional 'technocratic model' as a paradigm (Bines, 1992, p.13). The use of the technocratic model developed from a belief that professional problem-solving can be mastered singularly through the habitualised and rigorous application of a proven discipline of knowledge, theories and techniques.

 To illustrate the processes described in The Reflective Practitioner several concepts have been reviewed. These concepts include Argyris and Schon’s (1974) ‘Single and double loop learning’; Montgomery’s (1993) ‘Meta model for learning’ and ‘Reflective learning process model’. The following diagram is an adaptation of these models/concepts developed by Hatten (1997).

Figure 2 - Processes within ‘The Reflective Practitioner’

(diagram devised by R. Hatten, 1997)


 The following information can be clarified by reference to Figure 2. Argyris and Schon (1974, p.18) refer to the ready reliance on a static frame of reference as 'single loop learning' (1). In a static society in which social systems remain constant, knowledge is relatively stable and dilemmas in life are mostly predictable. Solving problems is mostly patterned on previous experience and proven solutions. In a fast changing society in which the direction for change cannot be predicted, the ability to critically analyse, make informed judgements and direct actions, would be very much valued.

This ability is the result of the combination of experience, propositional knowledge, tacit knowledge or know how, critical thinking and other kinds of process and intuitive knowledge which have been developed through previous reflections (Boud, Cohen & Walker, 1993). An understanding of the nature of the reflective practitioner may help to illuminate the skills needed for transformative or double loop learning (Mezirow, 1991; Argyris & Schon, 1974).

Reflection (2) is the processing of the experience and re-evaluation of perceptions, which then become the basis of transformed or new knowledge, and decisions on further action (Boud and Walker, 1991). Informed, directed and committed action is often referred to as 'praxis' (3) (Brookfield, 1987, p.28; Kemmis, 1985, p.141). Kemmis (1985, p.141) argues that 'praxis' is 'the most eloquent and socially significant form of human action', which forms the basis of the social order. The truly ‘reflective practitioner’ (Schon, 1991) actively participates in this moulding of the social order through ‘praxis’, but not all professionals embrace the same level of reflective activity and commitment to action. The questioning and change in frames of reference used to learn has been named 'double loop learning' (4) by Argyris and Schon (1974, p.19).

Schon (1991) claims that this skill is essential to survival in a professional world in which both ends of the theory-practice gap are changing rapidly. Knowledge is evolving and exploding, and the context of practice is constantly being modified. Even expectations of society on the outcomes of professional expertise are continually being revised and the basis of this expertise is the ability to solve unique problems.

Schon’s (1991) basic argument is that problems do not present as neat packages of itemised elements to which the application of a series of logical yes/no questions is sufficient to produce a solution. Problems (5) are 'problematic situations which are puzzling, troubling and uncertain' and which can be described as dilemmas (Schon, 1991, p.40). They can be constantly transformed so that the means and the ends are always in shadows. The most important process to apply is 'Problem setting (6) ….. in which, interactively, we name the things to which we will attend and frame the context in which we will attend to them' (Schon, 1991, p.40).

This knowing what to name and frame is often implicit in our actions and often forms a large part of professional expertise. This expertise is composed of many pieces of information which, if explicit, can overwhelm the capacity of the conscious mind (Schon, 1991, p.49). Schon (1991, p.50) refers to tacit knowledge as knowing-in-action, which will remain implicit unless effort is expended to make it explicit.

This reflection-in-action is usually triggered by some ‘disorientating dilemma’ within professional actions which are habitually guided by tacit knowledge. The disorientation (7) is faced when these actions do not produce the usual expected results (ie. defined by previous experience) and problem setting and reflection are needed to bring about a paradigm shift (8) which then determines the next action (9) (Mezirow, 1991, p.56).

The effort made through reflection on this knowing-in-action , whether on the subject or his or her own actions or knowing, creates understandings which are made explicit, reprocessed and reinforced or modified. Schon claims that the 'art' of expertise has at its core, the reflection-in-action (10) during these moments (Schon, 1991, p.50).

The reflection-in-action is dependent on the 'action-present' when action is possible within the time frame of the reflection (Schon, 1991, p.62). Reflection-on-action (11) occurs when post mortems are carried out on previous actions at anytime after the experience has passed (Schon, 1991, p.276).

The objects of the reflection are multiple and varied, depending on the context and the stakeholders. In essence, the reflective practitioner is the researcher who is constructing a new theory, testing of which may help to find a solution for a unique case, but the theory construction is not separate to the action.

The initial inquiry is triggered by a problem which is initially set according to the observation at that moment. Comparison of this problem frame with knowing-in-action produces new phenomena or reframing of the initial problem. Awareness of feedback from the milieu stimulates reflection, which causes the individual to continue to reframe, experiment, transform knowledge schema and create new insights. This cycle is on-going, and can be indefinite thereby ‘praxis’ occurs through a paradigm shift (Schon, 1991, p.268).

From another perspective, the inquirer uses an existing repertoire of knowledge, reflects on similarities and differences, forms new hypotheses, tests shapes, as well as probing the situation. A 'generative metaphor' results from the processes, which is then used as the basis of the next cycle. Schon coined 'generative metaphor' to describe the identification of similarities in concepts which appear to be very different initially (Schon, 1991, p.183). These metaphors are then used to link generated ideas when similarities become obvious.

Throughout this process, some constants (12) are necessary to provide an overarching theory, as a stance for reflection-in-action which can become an ethic for inquiry (Schon, 1991, p.164). These constants give form to the process but also may cause limits to be set for reflective thinking (Schon, 1991, p.275). These are the frames of reference we all use to guide our lives. The circularity of this issue is not fully explored by Schon, and neither is the issue of whether actions are limited by the act of reflection itself.

Schon (1991) suggests that a paradigm shift is badly needed to turn mindsets away from the technical rationality of conventional professional education and practice, toward a system of reflective learning within all professional actions. Effectiveness of reflection is often dependent upon the generic skills related to an awareness of the moment, the ability to exclude other thoughts, and continuity of inquiry.

This 'continuity of inquiry entails a continual interweaving of thinking and doing' and therefore must be simultaneously developed with the generic skills (Schon, 1991, p.280). Schon (1991) suggests these generic skills are often neglected. This process demands a repertoire of sophisticated skills which can always be improved with learning from experience through reflection.

What is Quality Assurance?


Through recent history, both in health care and industrial settings, the phrases of Total Quality Management (TQM), Continuous Quality Improvement (CQI) and Quality Assurance (QA) have been heard. There is a general misconception that these terms are synonymous. This is predominantly true for TQM & CQI, but is not so for QA (Al-Assaf and Schmele, 1993, p. 70).

Examples of Quality Assurance in Health Care can be identified in mid-nineteenth century England. Florence Nightingale served as a nurse during the Crimean War and she was able to make a positive correlation between adequate wound care and a lower mortality rate in soldiers (Al-Assaf and Schmele, 1993, p. 4). Further developments have occurred over time, driven primarily by the impacts of ‘reduced resources’ (ie. State Government cut-backs and fewer people joining private health funds); and the information needs of a more educated group of patients (Thornber, 1992, p. 56). These further developments have required health care organisations to expand previous concepts of quality assurance to include leadership, and the organisation’s culture as components which have a significant bearing on the outcomes of quality (Koch, 1991, p. 1). Subsequently the practices of TQM were adopted.


TQM can be defined as a 'management philosophy which seeks continuous improvement in performance of the processes, products and services. The emphasis is on understanding variation, measurement, the role of the customer and involvement of the employees at all levels of an organisation in pursuit of improvement' (The Australian Council of Health Care Standards, 1992, p. 6).

 Many people have provided definitions for QA. For the purposes of this publication it will be defined as the 'planned and systematic approach to monitoring and assessing the care provided, or the service being delivered, which identifies opportunities for improvement and provides a mechanism through which action is taken to make and maintain these improvements' (The Australian Council of Health Care Standards, 1988, p. 5).

Essentially the difference between TQM and QA is that TQM is considered a management philosophy and as such, has a broader focus. QA is more focused on the analysis and correction or remediation of an identified problem area. Subsequently, the outcome of QA is more concerned with problem definition and resolution, or the development of standards. TQM, however, is considered to be 'a way to manage the many processes which ensure these quality issues pervade and infiltrate every aspect of an organization to improve its effectiveness and competitiveness and ability to flexibly adapt to new conditions' (Koch, 1991, p. 2).

 Rationale for selection of comparison model

There are many models which describe QA, TQM & CQI. In order to enable sufficient detailed comparison, only one framework or model will be chosen.

In an effort to maintain equity within the cycles involved in AR, it was decided that the model of QA should be compared. The rationale for this being that QA, like AR, can be considered a tool in a broader context, eg. AR can be a tool in a research method, and QA can be used as a tool within CQI/TQM (Green, 1991). In addition, it was the similarity of the core processes which generated the authors’ perceptions that these concepts could be similar.

 Model of TQM

A brief illustration of how TQM needs to be a combination between management, systems and workers is given by Oakland (1989).

Figure 3. The TQM Model

(after Oakland, 1989, cited in Koch, 1991, p. 3)

According to Thornber (1991, p. 58), the ten key elements of the TQM model include:

  1. Quality being defined 'in terms of customer perceptions of both the content and delivery of the service;
  2. Analyses systems for errors and variations rather than putting the blame on the people;
  3. Develops long-term partnerships with external and internal suppliers and service partners;
  4. Uses accurate data to analyse processes and to measure system improvement;
  5. Involves the staff who do the work, in system analysis and improvement;
  6. Sets up effective, collaborative meetings as the basis of teamwork;
  7. Trains supervisors and managers in leading the on-going improvement process;
  8. Engages staff in setting targets and ensures that results are fed back;
  9. Highlights the need for senior executives to plan strategically for the implementation of improvement; and
  10. Achieves long-term improvement through small-step incremental improvement.'

 Models of QA

Dr Edward Demming is credited to be the ‘father’ of the contemporary TQM systems (Al-Assaf & Schmele, 1993, p. 6; Anderson, 1993, p. 18). Demming devised a simple framework which could be used in QA because he acknowledged that TQM could become unfocused. The cycle primarily incorporates Plan, Do, Check, and Act as its format (see Figure 4).

Figure 4 - The PDCA Cycle

(Adapted from Murdock, 1991, p. 73)

The stages include the following components:


~ formulating a vision/mission,

~ predicting how people and systems will perform,

~ determining what data is necessary to describe the performance of a system or a process;


~ testing ideas and proposed changes on a trial basis;


~ evaluation of the results from the ‘Do’ stage;


~ incorporate information learnt from the trials within the ‘Do’ stage.

(Murdock, 1991, p. 73)

Since the development of Demming’s PDCA model, the need for development in QA has required further models to be developed. The current QA models are consistent with seven key elements (as highlighted by italics in the previously given definition).

Organisational Requirements

For QA to work within any organisation, there are two basic requirements. Firstly, QA projects need to be Planned. This planning should encourage the workers who are involved in the tasks to be active in the quality cycle. Preferably QA projects should also be organised in a planned manner across the organisation (Koch, 1991).

Secondly, the organisation ensures that there is a system which assists in maintaining the continuity of QA. Systematic refers to both the process and time-frames used. Systems which are set up to ensure that QA is regularly undertaken and the plans to carry out QA are ongoing, would be related to time-frame. A system which is set up to ensure that QA goes through a logical step-by-step process, involves people who are core to the issue being assessed. This requires that communication occurs in a set manner for it to be effective and would be one where the process is systematic.

 Phases of the QA cycle


Monitoring - This term refers to 'any systematic, ongoing process of collecting information on clinical and non-clinical performance. It is in this phase of the QA cycle that topics for review are identified, and subsequently monitored' (Anderson, 1993, p. 5). This phase is also known as the ‘data collection’ phase because of the emphasis on collection of information.

There are two categories for topics being reviewed. These are:

a) an area of concern which has already been identified; and
b) an aspect of service delivery or care which is important. This second area encompasses the monitoring of an issue for maintenance of a standard, eg. adherence to staff annual education on fire safety.

Information which is collected in the Monitoring phases is analysed and interpreted. This phase of the QA cycle is known as Assessing. It is important to identify trends from data collected which indicate areas of deviation from an acceptable standard. Organisations develop and implement systems to collect the data, but fail to identify the trends, which suggest exploration of the associated area is required. Anderson (1993, p. 5) identifies this phase as the area where many organisations ‘fail’.

From the analysis of data and information, plans to modify the practice or system are designed. These plans are then implemented in the Action phase.


Opportunities for improvement - It is important to realise that in QA it is not always ‘problems’ which are the focus of the program. QA also intends that an organisation can be pro-active. This means that QA can be carried out on an area which is identified as an ‘opportunity for improvement’ (The Australian Council of Health Care Standards, 1988, p. 5), eg. refining an existing system or identifying a potential problem. This would be done in order to improve the systems which are already in place.


Follow up or Evaluation - According to Anderson (1993, p. 5) this phase of the QA cycle represents the Assurance component within QA.

The aim of ‘follow up’ is to identify whether a change occurred after implementation of the Action and in the expected manner, ie. did the resultant modification occur toward the trend that is desired, rather than away from it. For validity, it is important to ascertain that the change which occurred after implementation of the Action was as a result of the Action rather than of any other variable.

The Evaluation phase is intended to check the quality of the outcome of the actions involved. Evaluation, although part of the phase of follow up, has a distinctive meaning as it 'involves a review of the actions taken to ensure that they were appropriate. The desired result of this evaluation is to show that the problem or concern was resolved or reduced to an acceptable level' (Anderson, 1993, p. 5). Where it is deemed that problems/concerns are resolved or reduced to an acceptable level, the outcome would be stated as satisfactory.

 Figure 5 - The QA Cycle

(after Anderson, 1993. p. 6)


Feedback - should be occurring throughout the entire cycle. Communication between the people involved and affected is necessary both to maintain drive within QA, and to review the effects of activities and plans (The Australian Council on Healthcare Standards, 1988, p. 5). Additionally, feedback regarding the proposed and implemented changes, is needed for system changes or concerns to be expressed. The proposal may cause potential difficulties which are not identified by the QA team. But where effective feedback is occurring, others who will have to implement the changes may be able to identify these difficulties. This is so that they can be addressed prior to implementation.

 Benchmarking and Standards

The expanded cycle of QA identifies how the QA model can be used to establish standards and/or benchmarks. Figure 6 visually represents the possible options available, where the outcome of a QA cycle is satisfactory. When a follow up or evaluation phase within a cycle of QA identifies that an acceptable level of resolution has been achieved, a second modified cycle can be commenced. The purpose of this second cycle is to establish a set of criteria in order to define or set a benchmark or standard. Once these criteria have been established, the modified cycle continues in a similar manner to the original QA cycle. A Monitoring phase occurs, where data is gathered to identify that the benchmark is being met, then assessment of the data occurs. If analysis of the data identifies that the defined benchmark is not being met (or is unsatisfactory), then the cycle returns to the initial QA cycle. The purpose of the return to the initial cycle is to establish where the problems are occurring and allow for resolution. In the event that analysis of the data shows that the benchmark is being met the modified cycle continues.

Figure 6 - The Expanded QA Cycle

(after Anderson & Noyce, 1991, p. 32)

Similarities and Differences in the Characteristics of Action Research (AR) and ‘The Reflective Practitioner’ (RP)



AR: Since it involves a group of people, only requires that some of these people have this ability. The others can learn during the process.


RP: Requires that a practitioner has the skills and abilities to engage in reflection on their own practice.

Trigger Point


AR: The trigger point for AR involves a group of people who have identified an area of common concern.


RP: This involves only one individual identifying a disorientating dilemma.



AR: Differing levels of tacit knowledge in the group members dictate the nature of the negotiation process which occurs in the development of the plan of action.


RP: This does not need to occur within the RP cycle.


Both: Tacit knowledge impacts on the outcome of the negotiation process..

Example: An analogy to describe the differences in negotiation patterns can be drawn from the differences between decision making as a single person and as a married couple. The single person is able to make personal decisions without conferring with another person, however in the marriage relationship partners need to confer to determine the impact and negotiate possible prior to selecting a solution.


AR: In the AR team, it is not essential that all participants have a well developed level of professional tacit knowledge, however some members need to have this for the group to function. Also all members bring tacit knowledge developed from personal life experiences to the AR process.


RP: Requires the individual to have a well developed level of professional tacit knowledge (although this may not be at a conscious level) to engage in a reflective process.



AR: Feedback occurs in all phases of the AR spiral, however is most predominate in the reflection component.


RP: Feedback is most explicit during critical analysis of the individual’s frame of reference. ie. during double-loop learning


Both: AR and RP use both reflection-in-action and reflection-on-action to enhance the level of feedback generated either within the group or by an individual practitioner.

Time frames


AR: Requires increased time for consensus to be reached within the group and for the process/plan of action to be made explicit.


RP: The process of developing a plan of action can occur in a very short period of time as it occurs internal to the practitioner, and as such consensus is not an issue. Additionally the RP process may never become explicit.

Benefits / Outcomes


AR: This process improves an external condition (ie work practice/conditions etc.) however may not contribute to the development of the individual practitioner.


RP: Engagement in the double-loop learning cycle assists in developing the professional’s expertise (eg assist to refine/develop clinical reasoning etc.).

General Comments


AR: The process of AR operates on a collaborative group basis.


RP: Operates on an individual basis.


AR: The focus may be on or around any issue and tends to be broad due to group input.


RP: The focus in RP is primarily practice based and narrow, due to the individual basis of the process.

 Similarities and Differences of Action Research (AR) and Quality Assurance (QA)

Trigger Point


AR: The trigger point is a mutual concern identified by a social group. The concern can be fuzzy and indefinite and may be modified throughout the cycle so that a beginning problem can become a very different problem at the end.


QA: The problem is always clearly defined and identified as a specific issue and remains the focus throughout the process.

Problem Identification


AR: The identification of the problem can either occur from within the group or a group may be persuaded by an external source to participate in an AR project.


QA: A problem can are externally imposed or internally identified. Participation in a QA project may not be voluntary and consensus is not a prerequisite.



AR: The next step is planning critically informed action.


QA: The next step involves monitoring and collecting data.



QA: Feedback is an integral part of the whole concept as it must occur during each phase of the QA cycle. The primary focus of reflection during QA is on the action which has already occurred. The feedback mechanisms provide input into the process of reflection-in-action.


AR: Feedback may also occur in all phases of the AR spiral however this feedback tends to be implicit. Feedback can become explicit in the reflect component of the AR spiral.



QA: The outcomes from a QA project can either demonstrate an improvement in benchmarking of standards or an improvement in the quality of a service. QA tends to result in improvement in standard operating procedures within an organisation.


AR: The results of an AR project contribute to the development of new theories or can expand existing theories. Outcomes can provide the practitioner with a justified foundation for their methodologies. Generally outcomes from AR result predominantly in improvements in working and social conditions for the group members.

Overall comment


AR: The AR spiral implies that action and knowledge development must occur otherwise it cannot be considered a successful action research project.


QA: The cycles can remain static over time and not be acted on. They can be suspended at any stage and rejuvenated at a later time.


Both: In general, QA has a much more rigid structure for project completion as opposed to AR which is flexible and allows for continuous revision.

Use of Reflective Practitioner and Quality Assurance within Action Research

Initially our thoughts regarding the three concepts were that if you combined QA and RP, the resultant process would be AR. From combined experience, we considered that in comparison to AR, the element lacking in QA was reflection or a formalised reflective component. RP however, appeared to lack the participatory and active components available in QA. Through combining elements of the QA and RP cycles, the missing elements would be filled and it was felt that this would produce a similar process to AR.

Further exploration and reflection have led us to conclude that, although our initial thoughts had merit, there were other issues which had not been fully understood or included in our considerations.

We tried to equate the various steps of the processes within each of the cycles, only to discover (as our comparisons suggest) that they are different. This was particularly apparent in the areas of ‘depth of processing’ and ‘context of use’. On several occasions, the group explored how the different cycles could fit into each other in order to enhance the final outcome, without changing the core intent of each of the processes.

The group found that RP could be used at the reflection stage of AR to enhance the possible outcome. Considering that not all practitioners have developed the skills necessary to engage in RP, participants in the AR process may be at differing levels of this skill development. This could invariably lead to variations of the depth and influence that RP has on the AR process.

QA also did not fit snugly into the AR framework. QA could be used to supplement the ‘plan’ stage as an external quality control measure and would feed back into the ‘act’ stage. QA would not enhance the overall AR process if it was not present. QA was perceived to be more of a stand-alone mechanism which could contribute to the quality of the process. This was achieved by maintaining the quality control when combined with any individual stage. It was felt by the group that this may become more of a barrier to the flow of discovery and interpretation in the AR process than an assistance.


Summary / Conclusion

Discovering the boundaries of QA and RP has allowed the authors to clarify boundaries in the AR process, and thus become more familiar with what the true nature of Action Research could be.

The major similarities between the three cycles of AR, RP and QA were:

the cyclical nature of the processes;

the basic functions of the stages of the cycles;

all cycles involved a combination of action and review/reflection on this action to clarify the initial concern/dilemma identified;

each concept was used to address areas of concern/problems identified by either an individual or group;

the overarching principle is exploration and solution of the initial concern/problem

Major differences between the three cycles of AR, RP and QA were:

Context of the processes: Overall there are differences in the nature of the context in which each concept is situated.

AR being socially oriented intends that the outcomes will be evidenced explicitly ie. through changes in social situations, systems and social conditions.

RP is internal to the practitioner and evidence for the outcome may be implied or explicit (ie. Seen through changes in learning, practice, or systems).

QA while being situation specific intends that the outcomes will be related to the identified problem, they may be implicit but are more likely to be explicit. ie. changes to the causative agent, system or educative to contributing members of the process.

Isolation of the RP process as opposed to the necessity of collaboration and interaction in the AR process.

Interaction level: Differences between working individually or as a group are highlighted when this aspect of the three concepts are considered.

AR: must be collaborative in orientation. Negotiation to determine both the area of concern and the plan of action is unavoidable.

RP: due to the individual nature of the process negotiation is unnecessary.

QA: can be group or individually based. This impacts on the level of negotiation required.

Degree of complexity involved in the process: The complexity in implementation of each concept is influenced by the content and/or the process

AR: The stages in the cycle are defined, however as the content of AR can tend to be ‘fuzzy’ and abstract in nature, the cycle can become complex.

RP: The depth of internal involvement in the process by necessity is complex due to the double-loop learning requirement.

QA: Generally the cycle is less complex due to the process being clearly defined and the content concrete.

Overall AR, RP and QA present as cyclic processes with differing levels of involvement and complexity within the stages. Each would be selected when scenarios require a particular outcome. AR would be selected in preference to RP or QA when a social question or problem needs to be solved. RP would be used to refine the individual’s expertise in practice. And QA would be selected when the issue of concern was to the quality of outcome.

 Questions for Further Inquiry

It would be interesting to:

  1. Explore the outcomes of applying each concept to the same problem to determine if the outcomes would be different given the concept used.
  2. Use RP and QA within the AR spiral, and determine if the outcomes of AR are significantly improved.
  3. Explore the level of conscious and subconscious engagement of participants at various stages of each concept, and how these interactions affect the outcome of each concept, especially for AR.


Argyris, C., & Schon, D.A. (1974). Theory in Practice: Increasing professional effectiveness. San Francisco: Jossey-Bass Inc.

Al-Assaf, A.F., & Schmele, J.A. (1993). The textbook of Total Quality in Healthcare. Florida: St. Lucie Press, Incorporated.

Anderson, B. (Ed.) (1993). Quality Assurance for Occupational Therapists: A Resource Kit. Zetland, Australia: The Australian Council of Healthcare Standards.

Anderson, B., & Noyce, J.A. (1991). Clinical Indicators and their role in Quality Management. In B. Anderson (Ed.) (1993). Quality Assurance for Occupational Therapists: A Resource Kit. (pp. 32 - 37). Zetland, Australia: The Australian Council of Healthcare Standards.

Bines, H. (1992). Issues in Course Design. In H. Bines & D. Watson. (Eds.) Developing Professional Education. (pp. 11 - 27). Buckingham: SRHE & Open University Press.

Boud, D., Cohen, R., & Walker, D. (Eds.) (1993). Using Experience for Learning. Buckingham: The Society for Research into Higher Education and Open University Press.

Boud, D., Keogh, R.., & Walker, D. (Eds). (1985). Reflection: Turning Experience into Learning. London: Kogan Page.

Boud, D., & Walker, D. (1991). Experience and Learning Reflection at Work. Melbourne: Deakin University.

Brookfield, S. (1987). Developing Critical Thinkers: Challenging Adults to Learn. Milton Keynes: Open University Press

Green, D. (1991). 'Quality Improvement versus Quality Assurance?' Topics in Health Records Management, 11, 58 - 70.

Hart, E. & Bond, M. (1995). Action Research for Health & Social Care: A Guide to Practice. Philadelphia: Open University Press.

Hatten, R.A. (1997). Guidelines for Effectiveness in Continuing Professional Education. Unpublished master’s thesis, University of Sydney, New South Wales, Australia.

Holter, I.M., & Schwartz-Barcott, D. (1993). Action research: What is it? How has it been used and how can it be used in nursing? Journal of Advanced Nursing, 18, 298 - 304.

Kemmis, S. (1985). Action Research and the Politics of Reflection. In D. Boud, R. Keogh & D. Walker. (Eds). Reflection: Turning Experience into Learning.(pp. 139 -163). London: Kogan Page.

Kemmis, S., & McTaggart, R. (Eds.) (1988). The Action Research Planner. Melbourne: Deakin University.

Koch, H.C.H. (1991). Total Quality Management in Health Care. Essex: Longman Group.

Mezirow, J. (1991). Transformative Dimensions of Adult Learning. USA: Jossey-Bass.

Montgomery, J.R. (1993, August). Reflection, A Meta-Model for Learning, and a Proposal to Improve the Quality of University Teaching. Paper presented at the faculty seminar on Reflection and Improving the Quality of University Teaching and Learning.

Oakland, J. (1989). TQM. Oxford: Heinemann. Cited in H.C.H. Koch, (1991). Total Quality Management in Health Care. (p.3) Essex: Longman Group.

Schon, D. (1991). The Reflective Practitioner: How Professionals Think in Action. Avebury: Ashgate Publishing Ltd.

Stringer, E.T. (1996). Action Research - A Handbook for Practitioners. London: SAGE publications.

The Australian Council of Health Care Standards. (1992) Position Paper. Zetland, Australia: The Australian Council of Health Care Standards.

The Australian Council of Health Care Standards. (1988) Quality Assurance in Practice: An overview of the concept and examples of activities. Zetland, Australia: The Australian Council of Health Care Standards.

Thornber, M. (1992). A model of Continuous Quality Improvement for health service organisations. Australian Health Review, 15, (1), 56 - 69.

Zuber-Skerrit, O. (1992). Action Research in Higher Education: Examples & Reflections. London: Kogan Page Ltd.

Zuber-Skerrit, O. (1995). Models for Action Research. In S. Pinchen & R. Passfield. (Eds.) Moving On: Creative applications of action learning and action research. (pp. 3 - 29). Queensland, Australia: Action Research, Action Learning and Process Management.

Back to

Action Research Electronic Reader

Send mail to with questions or comments.
Copyright © 1998 Ian Hughes, The University of Sydney
Last updated:
24 August, 1999