The Role of Collective Analysis in Generating Ownership and Action in Systemic Action Research


This is part of a series of chapter summaries of the Handbook of Participatory Research and Inquiry

In 2007 I published Systemic Action Research (SAR) which laid down some foundational principles as to how we could build on Participatory Action Research Approaches toward a more systemic approach to change. Since then, I have published widely on SAR. A detailed exposition of the key steps in the process (of which this is a summary) can be found in chapter 49 of the Sage Handbook of Participatory Research and Inquiry which I co-edited with Jo Howard and Sonia Ospina. The idea is to build methods that can take action research to scale without undermining (indeed strengthening) its participatory and inclusive nature. This is what I call scaling and deepening participatory research (Burns 2018).

By systemic I mean both that we are able to engage with the complex system dynamics within which the issues we wish to engage with our situated, and that our action research design mirrors that complexity. This means for example creating multiple action research groups which connect to the issue we are working on, in order to build in the different knowledges which represent the diverse perspectives and power relations across the system. This process has evolved over 20 years through multiple projects which I have co-facilitated with colleagues  and partners   They have each allowed an evolution of the methods across very different countries and sectors. More recent large scale projects have included work on slavery and bonded labour in India and Nepal, work on conflict transformation and peacebuilding in Myanmar and Mali, and work on Worst forms of Child Labour in Bangladesh and Nepal.

The chapter in the Sage Handbook of Participatory Research stresses the importance of collective analysis and shows how this can be done. It also stresses the ‘research’ in action research, highlighting the importance of grounding any action in knowledge gained from inquiry, as well as learning from the action itself.  These two things must go hand in hand. This is a version of evidence-based change.

Analysis is never neutral. It is enhanced by the understanding of context that the analysers bring to it. Collective analysis means that data analysis is subject to the scrutiny of being viewed through multiple lenses. This makes meaning open to critical contestation and dialogue. As a result, the analysis is likely be more robust.

Collective analysis is a process by which people explore the raw data themselves and with others. This means rather than being presented by analysis, they understand the nuances and texture of how that analysis was reached. This not only leads to better understanding but also to high levels of ownership of the analysis. Ownership leads to motivation for action, it also leads to championing of innovation enabling horizontal spread of ideas that work from community to community.

The process that I describe here is one way to do systemic action research. It combines two methodological elements. The first is collective evidence gathering and analysis in the form of life stories and the second is an action research process which is built from the analysis of the first process.

Life stories underpin this process. In a life story people respond to a prompt question which is open within broad boundaries such as “please tell me about your life and work here in this neighbourhood” These are not interviews (characterised by questions and answers). Open stories of this sort have the following advantages:

  • People tell the listener the things that they want to tell so we learn what is important by what the story-teller chooses to highlight.
  • Because people talk about what is important to them, we are likely to learn more, because unlike interviews we have not made prior assumptions (which underpin our questions)
  • People find the process of telling the stories and being listened two cathartic in its own right (although there is also a danger of re-traumatisation and support needs to be built into the process to meet any needs which manifest here)
  • Stories flow and it is much easier to see causes and consequences in a story than in an interview that is broken up into discrete questions. More generally it is much easier to see the inter-relationships between things.

Typically between 100 and 300 stories are collected in each focus hotspot. This gives a critical mass (some form of saturation) which allows us to feel confident that we have identified the most important patterns.

In the CLARISSA programme focussing on worst forms of child labour we have refined the key steps as follows:

  • Identification of story tellers and peer story collectors
  • Story tellers tell their stories and these are recorded and then written up
  • Story tellers and peer story collectors come together to  analyse the stories in an analysis workshop

The analysis workshop typically takes place over 4-5 days with 30-40 participants. Participants in pairs or trios will each analyse a proportion of the stories. We mix literate and non-literate participants to ensure that they can all engage.

The workshop has a number of phases.

We explain with lots of examples the idea of causes and consequences (direct and indirect, linear and non linear).

We give each pair or trio maybe 7-10 stories to analyse over a period of 24 hours.

We practise the process with one or two stories collectively so that everyone understands what to do.

The process involves (1) identifying the two key issues or relationships between things that are central to the story – put them each on a sticky note (2) drawing a small system map of all of the key factors in the story and linking them with arrows denoting causal relationships.

We analyse the sticky notes by placing them on a wall of paper and clustering them. This helps us to identify the main themes. We then have a collective discussion on what we see.

We analyse the wider pattern of causal relationships by transferring all of the factors and arrows which are on the small system maps onto a huge wall of paper (often a metre high by 6 metres long). Factors and arrows connecting them can only appear on the map once.

Once all of the relationships are depicted, we identify which lines are the strongest and check how many stories show that relationsip. Based on this we thicken the lines so that we can see which relationships are repeated and which are most dominant. We then have a collective discussion on what we see.

Based on this collective discussion we identify:

  • The major causal factors which are negatively impacting on the problem we are working on
  • Critical causal pathways where we think we might be able to catalyse some change
  • Areas which people are passionate to work on

Based on our discussion of these we identify core themes for action research groups which people want to work on. Action Research is a process for generating change by generating action from evidence, and then generate learning and evidence from action. Both the evidence gathering and the action are central to ensuring that the right actions are taken, that they are truly owned by participants. This in turn leads to the possibility that they might be sustainable and or taken to scale. We use a five-step cycle in our action research process:

Once the action research groups have been identified they follow the following steps over a period of 9 – 18 months.

Building on a kernel of enthusiastic participants at the workshop, Identify possible new participants and spend time building relationships and trust amongst all of the participants (this could take up to 3 months).

Develop a process for ensuring that all of the existing evidence (stories, maps, analysis workshop discussions) on the key issues for discussion are synthesised and communicatedto the action research group. Identify what evidence on that specific issue is still needed. Carry out an evidence gathering exercise on the specific causalities that the group is exploring.

Based on all of this evidence work with the group to develop ideas for action, and to think through the ground level theories of change, which will help to elaborate why participants think this course of action will work.

Plan Action, Take Action … and after an appropriate period of time assess what has worked and what has not worked.

Based on the learning from this exercise the groups will refine, reconstruct, or even abandon the action that was taken. If it is working well the group might then explore how to build on it.


The process described here is one in which action research groups are set up to explore and take action around issues of critical importance to participants. The choice of focus is built on a collective analysis of information. further evidence is gathered in order to collectively make sense of how the system works and to develop theories of change and the actions that flow from them. This enables action research groups to target their energies on what matters and as a result to ensure ownership, continued engagement and effectiveness.

Resources on Systemic Action Research

Other books articles and resources which will provide more detailed information on this approach:

Burns, D. (2021), The role of collective analysis in generating ownership and action in Systemic Action Research, in Burns, D.; Howard, J. and Ospina, S.M. (eds.) The SAGE Handbook of Participatory Research and Inquiry, London: SAGE Publications Ltd.

Burns, D. and Worsley (2015), Navigating Complexity in International Development: Facilitating sustainable change at scale.

Burns, D. (2015) ‘How change happens? The implications of complexity and systems thinking for action research’, Handbook of Action Research 3rd edition, pp 434-445 London: Sage 

Burns, D. (2014), ‘Systemic Action Research: Changing system dynamics to support sustainable change’, Action Research Journal, Vol. 12, issue 1. pp. 3-18, Sage  doi:10.1177/1476750313513910 

Burns, D. (2007) Systemic Action Research; A strategy for whole system change, Bristol: Policy Process