Structured abstract PurposeThe main ambition with action research is being able to solve organisational problems through intervention and to contribute to scientific knowledge. The main emphasis has so far been on the "research part". Here we focus on the "action part" of action research in order to generate rigorous research, to solve local problems and to deal with evident dilemmas in action research. The purpose of the article is to elaborate on the concept of action by addressing actions and roles in the practice of action research, illustrated by dilemmas in an action research project on information systems development in public sector.
Design/methodology/approachThis is a qualitative case study. The empirical illustrations of this article originate from an action research project that focused the two e-service development initiatives analysed below. The analysis is structured using key aspects and phases proposed by Avison et al. (2001). As a result of the analysis the concept of action is elaborated. The action elements action, actor, motive, space, and time are analysed together with different roles. This goes beyond the existing action research literature.
FindingsThe conclusions show that there is a need to understand actions and roles within action research projects -not separating action from research. Research is also seen as action. The practice of action research is also discussed as context bounded interactive social action: action research as a recurrent, interactive, and dynamic activity. It is also identified that the understanding of roles, actions and interaction can help us to handle dilemmas in action research.
Research limitations/implicationsWe contribute to the body of knowledge concerning action research in the information systems research field and in general by exploring the need to study the concept of action (e.g. situations and elements), to be explicit concerning the different phases, roles and responsibilities and management of different dilemmas in action research. The inter-organizational development character in this study adds an extra dimension into the practice of actions research only partially highlighted. This one limitation. Another limitation is focus on public agencies. However, this is not critical for the results on action elements and the AR dilemmas that are studied.
Practical implicationsThe understanding of roles, actions and interaction can solve the dilemmas and challenges linked to the practice of action research in the information systems field, but such understanding can help us to discover and handle dilemmas in action research.1 This article is a rewritten and significantly expanded version of a former conference paper: Melin, U., Axelsson, K. (2007)
Originality/valueThe originality in this research is an illustration of and a perspective on action research as a context bounded interactive social action: action research as a recurrent, interactive, and dynamic activity. The value is that this knowledge can help us to handle dilemmas in action research.