Can Action-Oriented Research Help Social Science Be More Relevant?

In his posthumous address to his discipline, the late public administration scholar Christopher Pollitt lamented its diminishing relevance and standing, laying blame on a deep-seated institutional division. On one side he noted a tendency toward ‘detached scholasticism’ in which academics engaged in insular and increasingly abstracted conversations with little relevance, or even concern, for the practical utility of their research. On the other was a form of ‘high-brow management consultancy’, driven by salesmanship and superficiality and with little accompanying interest in ethical, informed and cumulative scholarship. Pollitt’s point was not that either tendency was wholly unhelpful, but that a relevant position in the applied social sciences lay somewhere in between, in what he earlier called the ‘poorly mapped shadowland’ bridging practice and theory.

What is action oriented research?

One approach following this middle-way is what we have taken to calling ‘action oriented research’ (AOR): research which explores socially relevant problems by bridging worlds of theory and practice, which involves collaboration between practitioners (and other research ‘consumers’) and academics and which shortens pathways to impact by combining research production with research use in situ. AOR is clearly far from a new approach and involves many methods which enable co-creation of knowledge of mutual benefit for both researchers and practitioners. Methodologies such as action research, action learning, co-production, and research roles like developmental evaluator and embedded researcher share a focus on relevance and praxis and conform to this conception of AOR.

Action oriented research: a route to relevance in the social sciences?

As a set of methodologies and research roles, AOR can make a strong claim as a route to relevance for the applied social sciences. This type of research requires a commitment and often investment to the research process by both researchers and the ‘researched’ and therefore, by design, tends to concern problems relevant to practice. AOR combines research production with research use, with findings informing practice changes on an immediate and ongoing basis within the research site, rather than relying solely on a lengthy and expensive publication model of research impact. COVID-19 has accelerated focus on this long standing issue; the need for collaborative, relevant and problem-focused research has never been greater.

Action oriented research: advancing academic theory

By focusing on significant and cutting edge problems but keeping theory development and testing as a key parameter of the research process, AOR is also well equipped to advance academic theory. Many large scale research projects underway across the UK have recognized this, adopting research methods with strong affinity to AOR, such as Smart Urban Intermediaries, Jam & Justice, and What Works Scotland. Many active and vibrant research networks exist for similarly inclined researchers, like the Collaborative Action Research Network, Action Learning Action Research Association, and the UK Participatory Research Network. AOR however occupies a very marginal position amongst leading journals and conferences. We found that just 2 papers have been published in the top 6 journals in public administration over the past 15 years which have utilized methods of Action Research, Action Learning, or Action Inquiry.

Roles like embedded researcher, researcher-in-residence or developmental evaluators are still more likely to be applied by consultancies and action oriented institutes, including for instance  Research for Real and the Institute for Voluntary Action Research.

How can action-oriented research achieve this middle way?

So can AOR provide a ‘middle way’ to Pollitt’s extremes of detached scientism and superficial consultancy? We think two areas remain to be addressed.

1 ) AOR needs clearer standards of quality and validity

While AOR can claim a strong connection with practice it remains a fragmented approach with practitioners spread across various research communities, each using different languages and different approaches (we note at least eight varieties of action research, for instance). A more significant problem is that these forms of research often use different languages and different standards of quality. In this way research therefore loses the opportunity to cross-pollinate between particular approaches and communities. As such research findings cannot easily be generalized, contextualized or adapted, and so currently AOR lacks the cumulative research impact to progress its disciplinary standing.

Validity is derived from direct involvement with research subjects, iterative testing and revision of the research process, and engaged rather than dispassionate observation.

This is also important because research approaches like co-production, Action Research and Action Learning adopt a different conception of quality and reliability than approaches deriving from the scientific method. Validity is derived from direct involvement with research subjects, iterative testing and revision of the research process, and engaged rather than dispassionate observation. No cross-cutting standards of evidence exist which can guide editorial boards or reviewers as they might for other approaches, leading to unclear standards for quality which evaluators might adopt, or which commissioning editors and reviewers might use to judge study quality. A significant question remains to be addressed – just how far different action-oriented research approaches can find enough common ground as a distinct research tradition.

2) AOR needs clearer pathways to impact

Relevance, along with ‘impact’, has become a watchword of research policy and higher education reform. AOR approaches are often undertaken with the express intent of improving research relevance, for example as informal models of knowledge exchange, or as ways to re-invigorate Universities’ civic role. While the Research Excellence Framework, and the Knowledge Exchange Framework can be criticized, they have at least created an environment for the comparative merits of AOR to be considered in that context. AOR shortens pathways to impact by combining research production with use, however pathways are less linear, emerging from the process of research rather than merely from particular research outputs. While discussions of research impact and usefulness often rely on linear accounts of improvement, there remains work to do on clarifying and increasing the visibility of AOR as a means of research impact, knowledge transfer, and civic engagement.