How to...
Carry out action research

What is action research?

Definition

"Action research can be described as a family of research methodologies which pursue action (or change) and research (or understanding) at the same time. In most of its forms it does this by:

• using a cyclic or spiral process which alternates between action and critical reflection, and
• in the later cycles, continuously refining methods, data and interpretation in the light of the understanding developed in the earlier cycles.

It is thus an emergent process which takes shape as understanding increases; it is an iterative process which converges towards a better understanding of what happens.

In most of its forms it is also participative (among other reasons, change is usually easier to achieve when those affected by the change are involved) and qualitative."

From Dick, B. (1999), "What is action research?". Available online at http://www.scu.edu.au/schools/gcm/ar/whatisar.htm

The normal position of the researcher is detached, scientific, standing outside events and diligently recording them. A number of methods may be used – questionnaire, focus group, interviews, observation, etc. – but it is generally the researcher who controls data gathering for purposes that affect their research rather than the participants' agenda. In other words, the subjects are passive in research terms: they may either be unaware of being "watched" or unconcerned about the data used from their interview or survey.

In action research, however, people are not just subjects but partners in the research process. The research arises not out of a question from an external individual, but as a shared process of reflection between the researcher and the participants; the latter help gather data in relation to their own questions; research results are fed back to them directly to improve the situation that was the subject of the research.

The learning so acquired goes not only to answer a research question, but also to solve an organisational problem, or at least to take forward and deepen the organisation's understanding of itself.

However much confined to the organisation are the interests of the participants, the researcher will have broader concerns. He or she is seeking and creating knowledge which relates not just to the project itself, but which can be applied to other projects, in other contexts, and which can add to practice and theory itself. Good action research, like any good research, seeks a broad influence.

The context of action research is almost always the organisation: most action research studies are case studies. They are very often linked with a change agenda, via the creation of deeper knowledge and understanding about a particular social or organisational issue, in order to improve a particular situation.

Below are two examples of research projects that deliberately set out to change rather than just observe:

1. In "Operationalizing the concept of value – an action research-based model" Naslund et al. (2006) looked at the concept of value within an organisation, principally at a global corporation which leads in the packaging supply field. The data were collected in the organisation by means of document studies, informal surveys, observations and interviews. The object was to help people in the organisation look at their ways of working and overcome any resistance to better ways of doing things.
2. In "Performance measurement action research" Moss et al. (2007) describe how the Centre for Facilities Management worked with a provider of estates and facilities management to a major UK government department. It used action research to create an improved performance measurement system for the organisation.

In "conventional" research, the researcher either takes a snapshot of what is going on in an organisation at a particular time, or does a longitudinal study. In action research the participant as well as the researcher reflects on actions, while the researcher is equally concerned to see organisational change as to have research hypotheses substantiated.

In "The impact of e-resources at Bournemouth University 2004/2006" Beard et al. provide the following definition of action research:

"Action research … seeks to bring together action and reflection, theory and practice, in participation with others, in pursuit of practical solutions to issues of pressing concern to people (Reason, P. and Bradbury, H. (2001), Handbook of Action Research, Sage, London, p. 1)."

In "The quality of an action research thesis in the social sciences" Zuber-Skerrit and Fletcher (2007) quote the following definition of action research which was first given at a 1989 symposium in Brisbane:

"If yours is a situation in which:

• people reflect and improve (or develop) their own work and their own situations;
• by tightly interlinking their reflection and action; and
• also making their experience public not only to other participants but also to other persons interested in and concerned about the work and the situation, i.e. their public theories and practices of the work and the situation;

and if yours is a situation in which there is increasingly:

• data gathering by participants themselves (or with the help of others) in relation to their own questions;
• participation (in problem posing and in answering questions) in decision making;
• power sharing and the relative suspension of hierarchical ways of working, in a conscious move towards social and industrial democracy;
• collaboration among members of the group as a 'critical community';
• self-reflection, self-evaluation and self-management by autonomous and responsible persons and groups;
• progressive (and public) learning by doing and making mistakes in a 'self-reflective spiral' of planning, acting, observing, reflective planning, etc.; and
• reflection that supports the idea of the '(self-)reflective practitioner';

then yours is a situation in which action research is occurring."

The history of action research

The concept of action research goes back to the German psychologist Kurt Lewin (1890-1947). As a Jew and a leftist, Lewin was forced in the 1930s to flee Hitler's Germany for the USA, where he did much to develop our contemporary ideas of change. He was particularly (and understandably) concerned with social justice and racial discrimination, seeing the latter as linked with problems of management and gatekeepers who determined what was and wasn't done. He believed that efforts for change should be focused on the group, which should challenge group norms and processes.

The concept of action research grew out of Lewin's strong social convictions – research would be linked with action to create a change process over organisational issues directly experienced by practitioners. Good research, Lewin claimed, should produce real change and not just end up in books (he was writing in an era well before the dominance of the top US journals with their "scientific" approach or the UK Research Assessment Exercise).

A particularly good account of Lewin's contribution to action research theory is given in Adams and McNicholas (2007).

In order to produce this change, the researcher needed to witness events at first hand as a partner in the process, to become truly involved in the system. His or her role is not only to document, but also to facilitate the process of reflection and enquiry.

A group of university researchers worked in partnership with the local high schools to help the latter determine what were the factors that led to pupils staying longer in school. They used a method of "appreciative inquiry", which is looking for the best and building on it. The positive environment thus created encourages and stimulates learning (Calabrese, 2006).

Action research has also been taken up in Latin America, by Paolo Friere and Orlando Fals Borda, and is currently a popular research methodology all over the world with a burgeoning literature, especially over the last two decades.

Applications of action research

According to Gapp and Fisher (2006) action research presents:

"a very effective alternative to social science research methods in that it is: practical, participative and collaborative, emancipatory, interpretive and critical. The process of action research is very effective in identifying creative solutions".

As described above, and because of its participatory and organisational nature, action research is much identified with change, whether as an actual change initiative, observation of change process or seeking to manage change in an organisation. Part of its strength here lies in its ability to take interaction with participants further than with normal research, yielding deeper understanding of individuals and drawing down more illuminating data. Equally it helps participants themselves to gain deeper understanding by reflecting on their experience.

Action research can also be strongly linked with action learning – learning from experience, particularly in an organisational context. As the researchers participate in the organisation, they create an impetus for learning as members of the organisation reflect on their experiences.

In Scandinavia, the "dialogue conference" has been in use for 30 years, and involves creating a public forum for participation and dialogue organised according to principles of respect and democracy. "Developmental magic? Two takes on a dialogue conference" (Philips and Huzzard, 2007) explores the use of this technique in action research to facilitate change within health organisations, taking on themes of visions, challenges, ideas and plans for future work.

Action research was used as a tool to try and understand the management of change in an organisation, the organisation being a university and the change being the wide scale adoption of electronic resources, as described by Beard et al. (2007).

Action research is common in research which looks at social issues, such as ethnicity, equality and the environment. Another application is in educational research, its popularity here being no doubt due to the obvious need to improve practice.

An EC funded project wanted to examine the question, would new migrants be more likely to work for the NHS if they had a better command of English and related material? This was researched through a number of focus groups comprising health service workers and non-working minority groups. Material was also created and trialled with a number of groups of learners.

Finally, there is an obvious attraction for action research for part-time management students looking for a way of integrating their studies with their work, many of whom use their project as a way of solving a workplace issue and carrying out some consultancy.

References

Adams, C.A and McNicholas, P. (2007) "Making a difference. Sustainability reporting, accountability and organisational change", Accounting, Auditing & Accountability Journal, Vol. 20 No. 3, pp. 382-402.

Beard, J., Dale, P. and Hutchins, J. (2007), "The impact of e-resources at Bournemouth University 2004/2006", Performance Measurement and Metrics, Vol. 8 No. 1, pp. 7-17.

Calabrese, R.L. (2006), "Building social capital through the use of an appreciative inquiry theoretical perspective in a school and university partnership", International Journal of Education Management, Vol. 20 No. 3, pp. 173-182.

Gapp, R. and Fisher, R. (2006), "Achieving excellence through innovative approaches to student involvement in course evaluation within the tertiary education sector", Quality Assurance in Education, Vol. 14 No. 2, pp. 156-166.

Moss, Q.Z., Alho, J. and Alexander, K. (2007), "Performance measurement action research", Journal of Facilities Management,
Vol. 5 No. 4, pp. 290-300.

Naslund, D., Olsson, A. and Karlsson, S. (2006), "Operationalizing the concept of value – an action research-based model", The Learning Organization, Vol. 13 No. 3, pp. 300-332.

Philips, M.E. and Huzzard, T. (2007), "Developmental magic? Two takes on a dialogue conference", Journal of Organizational Change Management, Vol. 20 No. 1, pp. 8-25.

Zuber-Skerrit, O. and Fletcher, M. (2007), "The quality of an action research thesis in the social sciences", Quality Assurance in Education, Vol. 15 No. 4, pp. 413-436.

Research design and data collection

Like any research approach, action research has its own development cycle, its links with theory, what may rather grandly be termed its epistemological stance (i.e. whether it aims to build on or build theory) and its preferred data collection methods. This is all part of research design, and this section will look at how to design an action research project in a thorough and effective way.

Link with participants

As we saw in the previous section, what really distinguishes action research from other forms of research is that the participants have a much more direct role. This will influence the design from start to finish. For example participants will:

• need to be involved in the planning process, in order to ensure that the research asks questions that are relevant to their concern,
• possibly have a hand in data gathering, and definitely an interest in the results,
• look for practice related, as opposed to research, outcomes, and a change in their own practice.

Action research has been compared with consulting in that it is concerned with development of practice; the difference is that it is cyclical rather than linear, the cyclical element being concerned with reflection, designed not only to improve practice but also to generate theory. The time-scale is often longer, and the budget larger.

The theoretical connection

For all that action research is concerned with improvements to practice, the theoretical angle is also important as all research seeks to build theory.

Action research is generally inductive rather than deductive – that is, the data collection builds rather than tests theory. The researcher may start with a particular theoretical position, but should be open to what the research yields, and to open reflection.

This openness, however, does not negate the importance of immersing oneself in the relevant literature, both content and methodology, in order to give a context to the problem.

The unique nature of action research interventions, and their non repeatability, means that they are not good for rigorous theory testing. On the other hand, they are good at testing theoretical frameworks, or theories as related to other theories. Hence their usefulness in organisation studies, where such frameworks apply.

The important thing is for the particular research to have results which extend beyond the confines of the particular project and have a general application, whether as a test of theoretical frameworks as described above, as theory which informs more robust practice, or a tool, model, or method which can be used in a range of situations.

In "Making a difference. Sustainability reporting, accountability and organisational change" Adams and McNicholas (2007) conclude their study of sustainability reporting thus:

"Our study has shown that, through action research, academics can assist organisations in bringing about improvements to their sustainability reporting processes, accountability and sustainability performance. Action research might also contribute to academic literature and theorising by improving our understanding of: what drives organisations to provide an account of their sustainability performance; what determines the level of accountability attained; the complex nature of the interactions and relationships between organisations and their stakeholders on sustainability issues; and the manner in which sustainability reporting processes impact on organisational change towards improved sustainability performance."

In "Action research as culture change tool" Marcinkoniene and Kekäle describe a programme of change in post-communist Baltic schools. They use the literature to describe the prevalent culture, and conclude that it would be possible to apply a similar action research intervention in other post-communist states, where schools are likely to have a similar culture.

Action research takes the researcher and the organisation through a cyclical process of deepening understanding, both of the organisational problem and the research question, in which reflection happens not just at the beginning and the end, but throughout the process. This reflection yields a greater sense of empowerment for participants, who feel themselves more in charge of decision-making processes and more able to see problems clearly.

The iterative nature of the reflection process makes action research similar to grounded theory, which involves going back into the field after looking at data with a clearer understanding of the key issues, and developing further understanding through more research.

In "The quality of an action research thesis in the social sciences" Zuber-Skerrit quotes her own theoretical framework of action research – the CRASP model. Action research is:

• Critical (and self-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.

Collecting data – multiple sources of evidence

The great advantage of the case study (and most action research, as we have seen, takes place in a case study environment) is that it can yield data from many different sources. Many action research studies use a combination of artefacts, document studies, surveys, interviews, focus groups, discussions, participant observation, group work, performance measurement. In addition, hard data may be available as in the following example.

Example

Adams and McNicholas (2007) report on the data collection methods, which supplemented discussion, observations in meetings and interviews, with hard data such as operational statistics, financial accounts, marketing reports, sustainability reports, as well as detailed study of the website, annual reports etc.

The most striking thing about action research, however, is that much of the data is supplied by the individuals themselves as part of feedback on how the object of data collection helps them in their work and lives.

In "Prioritizing tactical quality improvement: An action research study" Hales et al. (2006) describe the testing of a method for improving decision making at the shop-floor level. The method was tried out on the shop-floor with interviews being conducted with users and observation of people using it.

Andersen et al. (2006) describe how the researchers were involved in bringing about changes in a bank, and deployed a number of methods to identify shortcomings and suggest improvements, including:

• interviews with all employees in the bank;
• observation of employees' interaction with customers in customer meetings and other situations;
• interviews with customers, both those observed in interactions with service personnel and those who had not been in contact with the bank in a while;
• performance measurement of some key factors, e.g. customer flow in the bank office;
• group work among the bank employees to analyse problem areas and develop improvements; and
• the use of tools like business process analysis, root cause analysis, and similar techniques to shed light on problem areas.

Beard et al. (2007) used action research because it was participatory, encouraging researchers to seek views which they did through online questionnaires and subsequent interviews with a number of participants, and which they triangulated with data on online use.

Gapp and Fisher (2006) describe how the traditional course evaluation questionnaires were supplemented by focus groups as an additional way of finding out how to improve the course.

References

Adams, C.A and McNicholas, P. (2007) "Making a difference. Sustainability reporting, accountability and organisational change", Accounting, Auditing & Accountability Journal, Vol. 20 No. 3, pp. 382-402.

Andersen, B., Henriksen, B. and Aarseth, W. (2006), "Holistic performance management: an integrated framework", International Journal of Productivity and Performance Management, Vol. 55 No. 1, pp. 61-78.

Beard, J., Dale, P. and Hutchins, J. (2007), "The impact of e-resources at Bournemouth University 2004/2006", Performance Measurement and Metrics, Vol. 8 No. 1, pp. 7-17.

Gapp, R. and Fisher, R. (2006), "Achieving excellence through innovative approaches to student involvement in course evaluation within the tertiary education sector", Quality Assurance in Education, Vol. 14 No. 2, pp. 156-166.

Hales, D.N., Siha, S.M. and McKnew, J.I. (2006), "Prioritizing tactical quality improvement: An action research study", International Journal of Operations and Production Management, Vol. 26 No. 8, pp. 866-881.

Marcinkoniene, R. and Kekäle, T. (2007), "Action research as culture change tool", Baltic Journal of Management, Vol. 2 No. 1, pp. 97-109.

Zuber-Skerrit, O. and Fletcher, M. (2007), "The quality of an action research thesis in the social sciences", Quality Assurance in Education, Vol. 15 No. 4, pp. 413-436.

Exposing action research

The advantages of action research will have become clear. It allows the researcher to work on a problem, not only yielding answers to the problem but also informing theory. It empowers participants, enables change, and creates opportunities for organisational learning. It yields rich data from multiple sources. It creates solid theory about practice, and hence leads to improvements.

Despite these benefits, action research is not without its limitations and critics. The latter point to the lack of validity and generalisability, on the ground that the intervention is a one off and not repeatable. Perhaps more seriously, however, is the fact that it is not perceived as mainstream, and will therefore not be welcomed by the gatekeepers of the all important mainstream international journals.

The advantages of action research will have become clear. It allows the researcher to work on a problem, not only yielding answers to the problem but also informing theory. It empowers participants, enables change, and creates opportunities for organisational learning. It yields rich data from multiple sources. It creates solid theory about practice, and hence leads to improvements.

Despite these benefits, action research is not without its limitations and critics. The latter point to the lack of validity and generalisability, on the ground that the intervention is a one off and not repeatable. Perhaps more seriously, however, is the fact that it is not perceived as mainstream, and will therefore not be welcomed by the gate keepers of the all important mainstream international journals.

How to improve the validity of action research

Validity can be ensured by applying rigour in the research methodology and design. In particular:

• Using the relevant literature, both method and content, to situate the problem, and then to reflect on findings.
• Using measures and methods that are well respected in the literature.
• Using standard tools to increase reliability – for example a written protocol for interviews.
• Writing up the research in a well argued and well documented manner.

The case study approach, and focus on the particular (organisation, problem, etc.) encourages multiple sources of data. This in itself increases validity, as evidence is coming from a range of sources, and allows for triangulation – checking of one source against another.

At the end of their article, "The impact of e-resources at Bournemouth University 2004/2006", Beard et al. comment:

"Action research has proved to be an appropriate methodology for this type of research, as it allows quantitative and qualitative data to be used and learning to occur through action and reflection. Triangulation enables the views of a diverse community of academics and students to be considered alongside data from management information systems".

Triangulation in action research, however, is not merely a matter of being able to check one set of results against another; it is also possible to check differences in data from observation, personal accounts, etc. over the passage of time. However, the researcher here is looking for difference, not similarity, and for the opportunity to reflect on the different perspectives the former reveals.

Validity is also increased by having several researchers on the project, which gives greater objectivity. Hales et al. (2006) describe how they tested a method one of them had developed for improving tactical decision making. They used multiple sources of evidence – including open-ended interviews with senior managers, supervisors and workers, and observation – as a way of validating data, and four researchers to increase objectivity. The first author developed the method, the second and fourth authors conducted the interviews, which were analysed by the third, who also reviewed research logs. Three of the authors carried on participant observation.

Dissemination

Dissemination is obviously an excellent way of generalising research. According to Eden and Huxham (2002), action research calls for book length, owing to the incremental nature of theory development, the need to provide background and the complexity of the data itself. Doherty and Manfredi (2006) believe that despite the growing popularity of action research, it is disdained by "mainstream" management journals which favour more traditional methodologies.

Failure to benefit from the top publication outlets (i.e. top American journals) hits both UK and US based scholars. The former are in hock to the Research Assessment Exercise, which provides top ratings for international (and mostly American) journals which promote a conventional, positivist, research ethos, and in any case treats with suspicion any research that is not mainstream. The latter can only receive tenure if they publish in the aforementioned top journals. For a young scholar, doing action research can be just too much of a risk.

That notwithstanding, much action research does get published in journals – for example, a fulltext search for the phrase "action research" in Emerald's database yielded over 1,300 entries (search as of January 2008).

Students using action research as a method for a doctoral thesis may also experience problems, taking longer to complete, and having difficulty finding a supervisor who understands (Zuber-Skerrit and Fletcher, 2007). According to the latter, a good action research thesis should solve a real and complex problem, benefit from top level commitment in the organisation the student is researching in, and the reflection should be to find out what is the gap in knowledge that will earn the doctorate.

There is quite a bit of advice on writing an action research thesis, as will be explained in Section 4 "Action research resources".

Zuber-Skerrit and Fletcher (2007) advise that a quality action research thesis is one which:

• solves a real, complex problem – it is necessary to get the cooperation of top level management,
• contributes to both practical and theoretical knowledge –- above all, identifying a gap.

This quality is achieved by:

• a carefully designed, explained and justified methodology,
• an original contribution which provides relevant support and validation and which is well argued,
• the use of relevant literature, both methodological and content, aligned to the topic,
• a clear, concise and accurate piece of writing, free from errors.

References

Beard, J., Dale, P. and Hutchins, J. (2007), "The impact of e-resources at Bournemouth University 2004/2006", Performance Measurement and Metrics, Vol. 8 No. 1, pp. 7-17.

Eden, C. and Huxham, C. (2002), "Action research" in Partington, D (Ed.), Essential Skills for Management Research, Sage Publications, London.

Doherty, L. and Manfredi, S. (2006), 'Action research to develop work-life balance in a UK university', Women in Management Review, Vol. 21 No. 3, pp. 241-259.

Hales, D.N., Siha, S.M. and McKnew, J.I. (2006), "Prioritizing tactical quality improvement: An action research study", International Journal of Operations and Production Management, Vol. 26 No. 8, pp. 866-881.

Zuber-Skerrit, O. and Fletcher, M. (2007), "The quality of an action research thesis in the social sciences", Quality Assurance in Education, Vol. 15 No. 4, pp. 413-436.

Action research resources

There are a number of web resources on action research. Undoubtedly the best is that from Southern Cross University.

A substantial collection of action research resources presented by Bob Dick of Southern Cross University. It contains AREOL, an online course in action research, many useful papers on topics such as evaluation, using particular techniques and processes etc. There is also a section on writing an action research thesis, as well as abstracts of completed AR theses.

Has a lot of useful publications, including doctoral and masters' theses, as well as some useful papers including an introductory chapter from the Handbook of Action Research.

This site rates low on usability and is a hotchpotch of assorted items, but it does provide links to theses as well as a paper aimed at new action researchers: http://www.jeanmcniff.com/booklet1.html.

A professional network of action researchers, based at Manchester Metropolitan University, with publications and a discussion list.

A series of links which relates to action research in an educational context.

• Southern Cross University – Bob Dick

• Centre for Action Research in Professional Practice (University of Bath)

• Jack Whitehead's pages at the University of Bath – ActionResearch.net

• The Collaborative Action Research Network

• Action research at Queen's University