Research Designs-

Action Research
Action research- Ranjit Kumar,David Gray (Ch 15)
Action Research
As the name suggests, action research comprises two components: action and
research.
Research is a means to action, either to improve your practice or to take action to
deal with a problem or an issue.
Since action research is guided by the desire to take action, strictly speaking it is not
a design per se.
Most action research is concerned with improving the quality of service. It is carried
out to identify areas of concern, develop and test alternatives, and experiment with
new approaches.
Action research seems to follow two traditions. The British tradition tends to view
action research as a means of improvement and advancement of practice, whereas
in the US tradition it is aimed at systematic collection of data that provides the basis
for social change.
Action research, in common with participatory research and collaborative enquiry, is
based upon a philosophy of community development that seeks the involvement of
community members.
Involvement and participation of a community, in the total process from problem
identification to implementation of solutions, are the two salient features of all
three approaches.
In all three, data is collected through a research process, and changes are achieved
through action. This action is taken either by officials of an institution or the
community itself in the case of action research, or by members of a community in
the case of collaborative or participatory research.
There are two focuses of action research:
1. An existing programme or intervention is studied in order to identify possible
areas of improvement in terms of enhanced efficacy and/or efficiency. The findings
become the basis of bringing about changes.
2. A professional identifies an unattended problem or unexplained issue in the
community or among a client group and research evidence is gathered to justify the
introduction of a new service or intervention. Research techniques establish the
prevalence of the problem or the importance of an issue so that appropriate action
can be taken to deal with it.
Features of AR
Research subjects are themselves researchers or involved in a democratic
partnership with a researcher.
Research is seen as an agent of change.
Data are generated from the direct experiences of research participants.
Types of AR

Participatory Action Research Insider Action Research External Action Research

Action Science Cooperative Inquiry

Participatory AR
Authentic participation means immersing people in the focus of the enquiry and the
research method, and involving them in data collection and analysis.
One of the primary aims of PAR is to transform situations or structures in an egalitarian
manner. Hence, it has been used to deal with issues such as inner-city and rural poverty,
education, mental health, disability and domestic violence.
In the 1990s, however, PAR has also been taken up as a legitimate research approach by
powerful agencies such as government departments, universities and multinational
companies.
In 1999, for example, the World Bank commissioned a ‘Consultation with the Poor’
involving over 20,000 people in 23 countries.
Gaventa and Cornwall argue that the key element in PAR is a process of reflection, social
learning and the development of ‘critical consciousness’. This is particularly so among
oppressed groups of people, where non-experts play a central role
Insider AR
‘Insider action research’, is one in which managers are engaged in action research projects
in their own organizations.
Often, these projects are undertaken as part of an academic programme of study such as an
executive MBA.
The kinds of issues addressed often include systems improvement, organizational learning
and the management of change.
One of the advantages of adopting insider action research is that managers have an
intimate knowledge of the organization being studied – they know its culture, its jargon and
its personal networks.
They can also participate freely in discussions or merely observe what is going on without
people necessarily being aware that they are being researched.
On the other hand, it may be difficult at times to maintain a sense of detachment and it may
sometimes prove difficult for an insider to cross departmental, hierarchical or network
boundaries.
External AR
External action research, where the researcher may be independent of the
professional context, but work within it and alongside professional practitioners (for
example, business leaders, managers, trainers or health professionals) to achieve
change.
Hence, action research is a process of collaboration for bringing about change.
The exact nature of this collaboration, however, may be problematic
Action Science
Action science attempts to integrate practical problem solving with theory building
and change.
Friedman (2001) but suggests that it involves a form of social practice which
integrates both the production and use of knowledge in order to promote learning
with and among individuals and systems.
The objective of action science is to help practitioners to ‘discover the tacit choices
they have made about their perceptions of reality, about their goals and about their
strategies for achieving them.
To achieve this, communities of practice are created in which both practitioners and
researchers make explicit their interpretations, which can then be made subject to
rigorous testing for their validity.
Gummesson (2000) divides action science into societal action science and
management action science.
The former is concerned with the kinds of macro social, political and economic
issues that arise, say, when a company is threatened with closure. This could
involve, for example, a participatory study by groups of workers who are directly
threatened by the closure. Such an approach stems from a belief that research
should not lie in the hands of ‘professional experts’, who will have their own
agendas and subjective biases.
Management action science is focused on a company as a business. Here, the action
researcher has the difficult task of tackling issues and producing results that are of
value to both science and to business.Thus, from a theoretical perspective, the
action researcher will seek to contribute to knowledge, understanding and
theoretical perspectives. But this must also be knowledge that can be applied and
‘validated in action’ (Gummesson, 2000: 119). This means that the life of the action
scientist is often prone to role conflict and ambiguity.Another aspect of
management action science is that it is interactive, that is, it requires close
collaboration between the researcher and the company client.Again, this may pose
problems for the researcher who may be pressurized to change original research
designs in the interests of producing short-term actionable results
Cooperative Inquiry
Finally, cooperative inquiry is related to action research in that both focus on
research with people rather than research on people. Where cooperative inquiry
differs is in the way collaboration between researchers and participants takes place.
Heron and Reason describe how co-subjects become ‘immersed in and engaged
with their action and experience’.
They develop a degree of co-openness to what is happening through deep
experiential engagement, often generated through music, drawing, drama and
dance.
Areas where AR is used
AR Process

Planning Acting Observing Reflecting

 Planning For
planning
Why activities are required.

purposes
Choosing a focus for the action the focus What actions need to be taken.
research project + having a becomes:
sense of commitment to
improvement. How tasks are to be accomplished.

Meeting with stakeholders to

obtain a consensus on the Who is to be responsible for each activity.
actions that are planned.
It is trying to identify the kinds Where the tasks are going to be
of actions that can lead to performed.
positive change.
When the activities are going to commence
and when they are to be completed?
 Acting
Having identified the focus of your research, the next step is deciding what sort of actions to
initiate and then what data to gather.
It is usually best to focus on the kinds of performance indicators that show whether you, or
others who are the focus of your research, are being effective in initiating change or not.
Data collection should be as comprehensive as possible, because important insights may only
emerge once the data are being analysed.
This means that you may have to use a wide range of data gathering tools, such as interviews
(individual and focus groups), participant or non-participant observation, informal meetings and
document analysis.
For every piece of data, ensure that you record the date, time, place and the people who were
present. Transcripts of conversations and records of meetings should be authenticated by
getting them signed by a relevant participant.
The main problem is knowing how much data to gather without the process becoming unwieldy
and unmanageable. As usual, the key is aiming to achieve a representative selection from the
possible range of data.
Frameworks for Data gathering (Stringer,
1999)
Interpretative questions.
• Participants might be encouraged to work through these in order to extend their understanding of the problem. These
questions might include: what are the key elements of the problem? How is the problem affecting us? Who is being
affected?
Organizational review.
• Participants should focus on analysing various features of their organization, including: the general mission or purpose of
the organization; its goals and objectives; the structure of the organization, including roles and responsibilities and the
efficiency or otherwise with which they are conducted; the factors that inhibit the enactment of these responsibilities.

Problem analysis.
• This is similar to concept mapping, only here participants are asked to identify the problem itself, the antecedents that led
up to it and the major consequences that have ensued.

Concept mapping.
• This is used by stakeholders to understand how different key elements in the problem relate to each other. The facilitator
begins by drawing a word that sums up the central problem. Participants then add new labels to the chart that represent
other elements associated with the problem. They then decide how the issues are linked.
Observing
Since action research is about taking action and often involves experimentation,
action researchers have to take note of the impact of their actions. This might
include providing authentic descriptions of what has been achieved.
These may be either factual (for example, transcripts of conversations), subjective
(such as, diaries and personal reflections) or fictionalized accounts that preserve the
anonymity of participants but are used to highlight issues explicitly.
Reflecting
The individual action researcher is not in a position to say whether their actions
have had an impact – it is for participants in the project to judge for themselves.
But action research is not just about fostering change in organizations, it is about
generating learning amongst the action research participants. This too needs to be a
focus of reflection – what was learned, what is its value, can it be applied
elsewhere?
Coghlan (2001) suggests that the outputs of action research should include evidence
of:
• How researchers engaged in the steps of action research, how they recorded
their data and how they were a true reflection of what was studied.
• How they challenged and tested their own assumptions and interpretations of
what was happening on a continual basis.
• How they accessed different views of what was happening showing both
confirming and contradictory interpretations.
• How these interpretations and analysis were grounded in academic theory and
how this theory both confirmed and challenged the analysis.
Methods of Data Gathering
Observations,
Dairies Interviews,
Questionnaires

Photography,
audio and video Memory work
recordings
Validating AR
According to McNiff et al. (1996), the purpose of validation in action research is to:
• Test out key arguments with a critical audience to identify where there is a lack
of clarity or focus.
• Sharpen claims to new knowledge and ensure that the data match these claims.
• Develop new ideas.

The starting place for establishing claims for the validity of action research is with the
researcher. McNiff (1988) suggests that the researcher needs to demonstrate
publicly that he or she has followed a system of disciplined inquiry. This includes
checking that any judgements made about the data are reasonably fair and accurate.
Validation can be quite an informal process, but may also involve the use of formal groups,
especially selected to scrutinize the outcomes of an action research project.
These could include critical colleagues, advisers or mentors, or fellow action research
colleagues working on the project.
If formal groups are used, it is important to ensure that participants both understand and
can empathize with the context of the research.
Waterman (1998) argues that the process of validation in action research is strengthened
by the ‘to-ing and fro-ing’ between the elements in the action research spiral (planning,
acting, observing and reflection).
Typically, action researchers are not satisfied with one turn of the action research spiral but
will repeat the process several times, allowing for the refinement of ideas and practices.
Ethics
Negotiating access with/within organizations or communities, with individuals and
with parents, guardians or supervisors.

Confidentiality

Right to withdraw

Communication

Good faith
Limitations
Significant amounts of time and effort have to be expended on maintaining
collaborative networks
Since action research studies, typically, take longer to complete than other
approaches, staff turnover and people leaving the project can be disruptive.
Also, while new knowledge generated through studies may lead to practical results,
these may not be widely reported in the academic literature – hence, they do not
reach the public domain, and their application to other situations may be limited.
Another problem is that generalizability