An Introduction to Action Research

Implementing an Action Research Process

An Action Research process is usually prompted by the belief among senior management that their organization needs to change. This could be in response to a new challenge, a specific objective or some other business need.

The process then continues in a four-stage cycle, as outlined in the diagram below. At the completion of the fourth stage the process begins again, with each cycle improving the effectiveness of the organization.

Observe

The first stage begins with the gathering of information from a broad range of stakeholders. This might include managers, employees, customers, suppliers or shareholders.

Information can be collected by:

• conducting interviews, either face-to-face or by telephone
• hosting discussion groups
• seeking written answers, usually in the form of a story or anecdote
• sourcing objective data, like sales figures or retention statistics
• inviting input via a website or intranet

Importantly, Lewin advised against poll-taking or any kind of short survey. In his view, this kind of inquiry lacks detail, can be skewed by people’s perceptions and limits the ability of the researcher to ask further questions. (This is not a universal view; see footnotes for further details. [5][6])

During the observation period, every participant performs the role of researcher. This encourages them to speak openly and honestly, and can increase support for the change process later on.

Reflect

In the second stage, participants reflect on what they have learned. They do this by analyzing data, comparing answers, and discussing further those areas where there are conflicting viewpoints or information.

The objective at this stage is to identify whether change is necessary and, if so, what needs to change.

It is also important to record an objectively defined ‘base level’, often a set of figures, so that participants can judge the success of their efforts. For example, if participants were investigating their organization’s high turnover rate, the base level might be ‘the number of employees to leave the organization in the year prior to the change’. If they were investigating their organization’s health and safety record, the base level might be ‘the number of reported accidents in the year prior to the change’. The success of both change initiatives would be signaled by a reduction in this number.

Plan

In the Planning stage, participants decide how to change. They discuss the various options available to them and, where necessary, consult with other stakeholders from throughout the organization. This might include internal communications staff, project managers and the organization’s finance director.

This helps participants:

• gain access to alternative points of view (What is the best way to implement change?)
• invite constructive criticism (How realistic is our plan for change?)
• build more robust plans (What are the potential dangers of the change?)

In addition, the Planning stage is used to:

• encourage support for change throughout the organization by involving stakeholders in the decision-making process
• assign responsibility for specific actions

At this stage, participants might decide that it is unrealistic to make sweeping changes all at once. In this case, they could agree on a small-scale change as part of an ‘overall plan’ to be implemented over the course of several Action Research cycles. In this way, small changes can be made and their success measured before further actions are taken, without the risk and investment of a large-scale change process.

Act

In the final stage of the cycle, the change is implemented. Because stakeholders from throughout the organization took part in the Action Research process, and expertise was sought during the Planning stage, change can be communicated quickly to all necessary parties and engagement is likely to be high.

Starting Over

Following the Act stage, the cycle begins all over again with a return to the Observation stage. This time, participants are tasked with:

• evaluating how their action performed against expectations
• learning the strengths and weaknesses of their action
• collecting information that will help them plan the next step
• modifying the overall plan in light of the results

The duration of this second Observation stage will vary depending on the change that was implemented. For example, a plan to reduce energy waste might result in a visible reduction in bills over the course of a month, but an attempt to tackle high employee turnover might take a year before success can be measured.

The Action Research can then continue, with each cycle bringing participants closer to completion of their ‘overall plan’. [7]

Variations on the Action Research Model

Over the past 70 years, Action Research has continued to evolve, such that there are now a number of variations available to change leaders.

Appreciative Inquiry was developed by Professor David Cooperrider and seeks to instigate change by asking unconditionally positive questions. Rather than tackle problems, participants focus only on what their organization does well to establish a foundation that can be built upon. [8]

Action Science is a technique pioneered by Harvard’s Chris Argyris. It introduced the idea of 'single’ and 'double-loop learning’. While single-loop learning involves testing a number of actions to achieve a desired outcome, double-loop learning re-examines why that outcome is desirable in the first place. [9]

Action Learning is a technique developed by Professor Reginald Revans, who posited that individuals learn best by doing. It places the emphasis on participants to discuss a problem among themselves, with minimal input from a consultant or change leader. [10]

Soft Systems Methodology was developed by Peter Checkland and his colleagues at the University of Lancaster, and views organizations as a mental construct. The Action Researcher in this method would develop a conceptual design of what the system could be and compare it to what the system is, so as to identify areas to improve. [11]

Common Themes

Variations of Action Research are nuanced. Some are led by individuals, others by groups. Some deal with specific problems, others tackle wider social issues. But common themes among them include: [12]

• Real problems in context – researchers investigate the perceived problems from within the organization.
• Collaborative engagement – researchers work with individuals from throughout the organization to make positive changes.
• Iterative cycles – understanding of the organization and its problems develop as solutions are tried.
• Dual outcome – the process solves a problem and grows knowledge within the organization.

Criticisms

Despite its popularity - and perhaps because of it - Action Research has been the subject of some criticism. Each new variation attempts to overcome a perceived fault in the original method, suggesting that there is no 'right way' to conduct Action Research and that each inquiry will by nature be unique. [13]

This complexity is not reflected in diagrams like the one above, and the change leader should be aware that Action Research is rarely as simple as it is portrayed. It requires a great deal of time for participants to immerse themselves in the organization and its problems.

Action Research has also been criticized in academic circles for its lack of scientific basis. [14] Like many social sciences, it is based on the perceptions and ideas of those involved and objective data is difficult to come by. For example, participants may agree that sales figures are poor, but have very different views on how to change them.

Finally, the method’s collaborative nature has been criticized because some participants may not have the skills necessary to carry out a thorough investigation. [15]

Conclusion

Despite these criticisms, Action Research cannot be written off. Time and again it has proven effective as a process for managing change, when conducted as an in-depth and honest investigation. Around the world, different organizations have used different variations, but the fundamental principle of all Action Research methods is the same: that understanding an organization helps generate ideas to change it, and changing an organization helps improve people’s understanding of it.