Thinking About Action Research and Time

In researching pedagogy, one of the methodological approaches I favour is action research. In some quarters action research is out of favour. It is seen as being too ‘local’, too ‘small scale’, and has little to say in the rush towards finding pedagogic ‘solutions’, the bread and butter of the ‘what works’ movement. But perhaps one of the main perceived problems with action research is its potential for bias, and a lack of structured evidence. These criticisms may be correct – in some instances, but why? I wonder to what degree time and the perception of time in education might be responsible for this.

Action research was initially popularised on the back of the work of Kurt Lewin. As a result of his work, the often summarised ‘process’ of action research became identified by the diagram below:


This diagram can give those wanting a quick fix of research an oversimplified notion of what is involved in action research. It has the potential for a very powerful methodology, but if taken at face value can lead to a seriously atrophied version of the process. Here, the cycle can be read as one focusing on a teacher, or teachers, reflecting on their practice, deciding what deficit might exist, planning for change and then enacting that change before reflecting on how successful they thought it was. But there is no explicit consideration here of the data capture which might be used, the degree of claim to be made at the far end of a cycle of research, or whether the focus chosen was based on personal bias or wider evidence. If it is taken as a personal or group-based reflective process, it becomes a ‘quick’ process. Reflection, planning, acting and observing can become a quick process and can give the impression of moving forward at a rapid rate. This is the illusion of action research as ‘rapid innovation’. However, it also panders to the current vogue in education for making rapid shifts, showing accelerated change with equally certain proclamations of success. But I would argue that true transformation and change is paradoxically a slow, measured process, but one which is also to a great extent contextualised.

I think action research has a huge potential utility in bringing positive change and for acting as a basis for informed discussion of pedagogic practice and change. However, to act as a useful and nuanced tool it is important that action research is approached in the same way as any other research methodology – with care and time. All too often it is seen as a ‘soft’ and ‘easy’ option, something that can be utilised as long as Lewin’s cycle above is followed at face value. I think one useful step we could take in moving action research forward is to abandon Lewin’s cycle within popular accounts and discussion, and replace it with the cycle developed by Andy Townsend (2010). The diagram below is a summary graphic of my interpretation of his framework:


This cycle begins with a consideration of an area for work, with further discussion to refine that idea to one that can become the focus of a piece of research. Importantly, a reconnaissance stage is included. This stage is intended to explore the chosen issue further, do others see the same problem as the person/people who are conducting the research? Does some form of baseline data help characterise the issue in the context in which it is being explored? In discussion with some Chinese ELT tutors, we even discussed the idea of looking at larger-scale quantitative data from within and beyond the organisation involved. This stage helps us to begin to gain a more in-depth and critical understanding of the context we wish to explore and intervene in – but it takes time. Having reflected on the initial focus in relation to this reconnaissance data, we can develop a more focused and meaningful action, or if we find that our initial ideas were misplaced, we might go back to redefining an initial focus, starting the process over. The cycle then involves an intervention, and from this to a reflection and evaluation of the change involved. The evaluation is important as it emphasises the need to include in the planning for action phase a coherent and meaningful framework for data collection. If a coherent data collection framework is developed, this also leads to the need for a coherent data analysis/interpretation framework from which reflection and evaluation has more critical meaning and offers more well-founded insights for future work.

This alternative way of understanding action research makes the need for a deeper and more critical approach much more explicit. But the essential feature for me is the need for a greater amount of time. It is reflective throughout, critical and considered. It makes explicit the need for a data collection framework which extends well beyond ‘reflecting on practice’, and a proper consideration of data interrogation. This model of action research is a slower process; in recently submitted projects as part of a PGCert in action research, students only completed one or two cycles of action research over the course of an academic year. However, the insights they gained were based explicitly in the data they had collected, and also recognised the contextual and nuanced messages their research could offer. By making the complexity of action research more explicit, and by repositioning it as a slower, data-based process the insights we gain may actually lead to more rapid and meaningful change and innovation.


Townsend, A., 2010. Action Research. In: Hartas, D, ed., Educational Research and Inquiry: Qualitative and Quantitative Approaches. Continuum. 131-145.

Also, read the following as a great introduction to action research:

Townsend, A, 2013. Action research: the challenges of understanding and changing practice Maidenhead : Open University Press.