What’s the connection between Agile and action research principles?

This blog post focuses on the relatively unknown relationship between principles of Agile for team-based project development and the long and dense tradition of action research. It’s based on a series of readings, practices and observations during my PhD about Agile and co-design for collaborative research processes (with support from Dimmons and a CECAN Fellowship), where I try to describe in parallel to case studies and paper writing some of the relevant findings and reflections I find.

Action research implies a highly participatory and iterative approach to the concept of research, mainly in the field of social sciences, which combines theory with practice, favouring a vision of pluralistic and instrumental knowledge. Strongly oriented to action and change, among the different formulations of action research this can be described as the collaborative analysis of problems by the affected people or communities, followed by a change in their situation considering the results. It is also characterized by taking place in local and private practice contexts, where it constructs descriptions and theories based on the perceptions of the participants. A broader definition of the basic characteristics of action research can be found in this chapter by Reason and Bradbury (2008):

• A set of practices that respond to the desire of people to act creatively in the face of practical and even pressing problems in their lives, organizations or communities;
• Calls for a commitment to people in collaborative relationships, opening new “communicative spaces” in which dialogue and development can flourish;
• Based on many forms of knowledge, both in the evidence that is generated in the research and in its expression in various forms of presentation as the learning is shared with broader audiences;
• Oriented to values, trying to address important issues related to the flowering of people, their communities and the wider ecology in which we participate;
• As a living and emerging process that can not be predetermined, but changes and develops as those involved deepen their understanding of the problems that must be addressed and develop their co-investigator capacity both individually and collectively.

Another useful definition by Altrichter et al (2002), reinforces the aspects of self-reflection and pragmatic approach of action research: “a form of collective self-reflective enquiry undertaken by participants in social situations in order to improve the rationality and justice of their own social or educational practices”. In this sense, action research represents a shift towards a collectivist orientation of research, in harmony with the constructionist perspective of knowledge formation, and involves dialogical processes that recognize participants as “co-researchers”.

The origins of action research are wide, but there is agreement that its genesis goes back to the work of Kurt Lewin and other social science research at the end of World War II. Lewin, who coined the expression “if you really want to understand something, tries to change it”, is also considered the father of social psychology and the concept of field theory, which explains human behaviour based on patterns of interaction between the individual and his environment. What is relevant for the purpose of this post is the connection of action research with what has been the discipline of organizational learning and, as I explain below, basic principles of the development of Agile.

Because Agile can be understood as a set of practices, values ​​and principles for the development of projects (not only software, as I described in a previous post), and in synthesis a form of organizational learning based on work as an iterative team, with requirements and solutions that evolve through the collaborative work of “interfunctional” teams that self-organize. As Tobias Mayer explains and explores about Scrum, one of the most popular Agile frameworks, we can understand Agile beyond a set of “rules” or “precepts” (which often makes it frustrating or heavily criticized), and see it as a process of empowerment of teams that want to work better and more effectively. This is, improving their productive relations and therefore the relations with the rest of their organization and the products, processes or services to which they are dedicated.

In a context of similar reflections (like this post by Christiaan Verwijs, or this article explaining Agile from the perspective of theory of change), there seems therefore to be a more than clear connection between Agile and previous work dynamics in small-scale autonomous teams in the ’60, especially from the perspective of sociotechnical systems, characterized by positive interdependence and cooperative developments. I refer to a whole corpus of theories about self-management and organizational change that started from the systemic study of the workplace by Kurt Lewin.

More than a coincidence (in terms of individual values, objectives and capabilities as key aspects of the continuous improvement of the conditions of people in social contexts), Lewin’s theories and practices in the fundamentals of action research coincide fully with those of Agile in the principle of the collaborative participation of workers in the choice of options and the evaluation of the results. Quoting Lewin, when he defines the systems model in an action research as an empirical and logical process of problem solving, involving cycles of action and reflection: “a spiral of steps, each of which is composed of a circle of planning, action and fact-finding about the result of the action” (Lewin, 1946), we can find there the basic concepts of feedback loops and the Scrum shared workflow. More resonances of Agile, action research and systems thinking can be found in the work of one of Lewin’s disciples, Eric Trist, from his approach to organizational change in his work with self-managed miners’ teams. In this context, he described how workers informally created systems that allowed them to be multidisciplinary and self-directed, adjusting to the circumstances as they evolved to improve their daily work.

Considering the development of organizations as a flexible spiral process that can implement changes through organizational learning, and Agile teams to some extent as the basic units of “socio-technical teams” usually defined in  literature about action research, the other clear connection in this regard is the influential work of Donald Schön. With established basic principles of reflection in action to emerge complexity and unlock the potential of people, the need to design and manage the environment as a bilateral task, based on direct and observable categories and how “doing and thinking are complementary. Doing extends thinking in the tests, moves, and probes of experimental action, and reflection feeds on doing and its results. Each feeds the other, and each sets boundaries for the other” (Schön, 1983). Some observations that I think completely connect with the fundamentals and rationale of Agile, and that I consider equally important as the already widely recognized and documented relationship between “agilism” and the precursor work by the same time of Takeuchi and Nonaka on flexible and autonomous teamwork.

Other references on how Agile can also be understood as organizational learning based on the “double loop learning” process, as defined by Schön and Argyris, would be the articles by McAvoy & Butler (2009) and McAvoy (2015). In this case, pointing to other considerations that also interest me and I want to discuss in parallel to what I observe articulated by research and evaluation organizations such as CECAN. A distributed academic center based on Agile principles for the evaluation of public policies around the Nexus, which regularly experiences the adoption of methodologies derived from action research, from the theory of change to the systemic mapping of complex environments.