Systemic Action Research

A participatory approach to systemic change

Systemic Action Research (SAR) is a form of Participatory Action Research (PAR) which aims to create systemic change in both localised contexts, and  across much wider systems of relations – these may be socio-economic systems such as supply chains or social-ecological systems or territorial/geographic systems.

Typically, it involves multiple parallel action research processes, each acting on different aspects of the system dynamics it seeks to address, engaging different actors as agents of change. 

SAR was developed in response to critiques that PAR, while rich, very often impacts on very localised problems or environments and is implemented by and with specific and small groups of (usually marginalised) actors. When conceived and operationalised programmatically, it can inform a highly adaptive alternative to traditional development programme management. It can be used to stimulate change which is more akin to that of social movements which involve collective identity and action. Change which ripples out, and is adopted by new participants, because it is successful. It has been used to successfully engage with extremely marginalised communities and many groups who are not literate.

The methodology was articulated in detail in Burns, D. (2007) Systemic Action Research. It has evolved with each major project since then). Burns and Apgar  are working on a forthcoming open access book ‘The Practice of Systemic Action Research’ to articulate 20 years of learning about putting these processes into action. SAR has been extensively implemented, developed and evaluated by members of the Participation, Inclusion and Social Change team at IDS.

SAR can help answer the following questions:

• How to design participatory research at scale while ensuring deep participation
• How to stimulate systemic change through innovative action
• How to build evidence of the system dynamics which you want to change
• How to build ownership for participatory change

While learning about how to create systemic change processes can be adapted by students using Action Research (in particular how it engages with systemic change) most students will not be in a position to facilitate multiple parallel action research groups.

Furthermore, SAR is a process which takes time. It typically requires a minimum of 9 months, and the case studies outlined here are mostly 18 months to two years in duration.

This makes this method more appropriate for programmatic work, and for work which catalyses social movement oriented change. Students are recommended to explore co-operative inquiry methods.

Design and principles

Three key design principles of SAR are:

• It will have an orientation toward creating systemic change. By which we mean that it is focused on changing the dynamics that perpetuate oppressions rather than the ameliorating their impacts.
• It will lay foundations for the action research processes by generating a systemic causal analysis of the dynamics which are at play. This can be done through participatory processes such as systemic narrative analysis processes or through drawing metaphorical pictures, or even through a collective analysis of more conventional data such as interview or survey data etc.
• It will engage actors from across the system that it is trying to change, harnessing their knowledge and positionality in relation to the issues that need to change, and seeing them as agents of change.

Key assumptions underpinning SAR are:

Marginalised communities, children, people facing conflict have extraordinary capacity for analysis, and often produce research which is more nuanced in relationship to the specific issues at hand, than that of external professional researchers. Many of these people have lived with danger and risk all of their lives and their capacity to assess risk is usually far greater than those of outsiders.

SAR like other forms of PAR is rooted in an interplay between different forms of evidence (including experiential knowledge) and action. Emphasis is placed on developing robust causal evidence from across the system in order to both understand drivers within the system and to ensure that the actions have legitimacy beyond the lived experience of the people involved in the PAR group. Collective analysis is central to the iterative co-design of a SAR process. Where evidence is analysed collectively this:

• increases the robustness of the meaning making process as it is subject to critique from multiple perspectives – often referred to as analyst triangulation
• builds appreciation for the collective challenges and opportunities which, with ongoing facilitation, builds motivation and results in ownership for action.

Like other forms of PAR, taking some form of collective action through using the participatory evidence generated is a central mechanism for building confidence, ownership and agency through iterative cycles of action and reflection.

A sequence of phases which loosely follows this pattern is typical for SAR processes:

• Build relationships and trust with communities and stakeholders. This stage may take many months and without it the process is likely to fail.
• Generate evidence on the dynamics of the system that you want to change and collectively analyse them. This has been done in different ways in different projects but we have found the collection of life stories and a collective causal analysis of peers to be effective.
• Collective analysis of evidence leads to identification of system dynamics that the groups want to work on. Causal dynamics from the system map are identified for Action Research Groups  to engage with. Prioritisation focuses on causal dynamics which are priorities for action of specific selected actors, and domains where change is possible. The number of action research groups will depend on the resources available to support them.
• Set up ARGs paying careful attention to how to link from the prioritisation of themes by some participants to then expand the number of participants to include others with a stake in the dynamics and issues being explored. Action Research groups typically comprise 8-15 people, although they could have more or less people involved, they may be focused on one actor group (e.g. children in WFCL) or multi-actors (e.g. community based ARG including a diversity of social groups)
• Facilitated cycles of action and reflection for ARGs to generate more specific evidence on causal dynamics it is exploring. This often involves layering in the lived experience of participants with new research gathered using any research methods that might help build understanding.
• Facilitated generation of theories of change around possible actions after careful consideration of the evidence base.
• ARGs then plan and take actions, working with powerful others were necessary, engaging broader communities were necessary.
• Facilitated reflection and learning in order to assess effectiveness of the actions and to harness the learning from action. This may lead to another round of evidence generation and action planning or new actions, or course corrections.

Critical for the ARGs to remain part of a SAR process that aims to achieve systemic change, at various moments after action has been taken groups are supported to explore how the emergent work of their group might be enhanced by, or enhance, the work of other groups

Thus, typical characteristics of a SAR process would be:

Multiple parallel action research groups (This has been up to 30 in some of the processes that we have facilitated)

An emphasis on evidence – with a plural perspective on what evidence counts- to ensure that the best methods are used by the ARGs to answer their questions.

Deliberate spaces for these parallel processes to interact with each other, to both enable learning and to inspire change which can move beyond one location or theme to influence broader dynamics in systems.

In larger scale processes we might expect to see the development of a meta level action research process to guide the direction of the programme. This was achieved very effectively in the CLARISSA programme with a Participatory Adaptive Management approach and an evaluation design that used methodological bricolage.

Mini case study

Systemic Action Research has been used in work on slavery and bonded labour in India and Nepal, child labour in Bangladesh and Nepal, peace building in Myanmar and Mali, and in various urban social justice inquiries in the UK.

As an example, the CLARISSA programme focused on worst forms of child labour. The main participants in the SAR process were working children themselves, and small business owners in the informal economy who were employing them. Others within the system who were engaged were parents, teachers, and other local decision makers. In each of the two countries (Nepal which focused on the Adult Entertainment sector and Bangladesh which focused on children working in leather production) the programme supported 13 action research groups running in parallel. These focused on very different aspects of the system with a view to creating a wider systemic effect through shifting dynamics in the systems.

For example, some groups focused on different aspects of working conditions, others focused on the way in which family violence led to children being forced into WFCL, others still focused on chains of causalities such as the way in which family break up led to difficulties in getting formal identify cards, which in turn forced children to work in the informal economy.

Still others, in both countries, worked on persuading schools to support working children to get an education, and working with them to create a system which would enable this to work. Working across all of these domains both produced (a) innovation in relation to the specific issue of the groups, (b) innovation that resulted from working across all of these domains at the same time and a relatively unique depth of evidence from across the whole system which could be harnessed for policy change.

The CLARISSA programme have produced a reflection guide based on their experiences, helpful for anyone considering facilitating deep participation for systemic change using SAR.