A Comparison of Action Research and Scrum
A great benefit of studying at a British university in my opinion, is that they put emphasis on applying research methodologies in practice. This doesn’t end after the research is over. All these learnings about processes, ethics, etc. are potentially applicable in one’s profession. For example, applying Action Research in order to improve a company’s workflow.
Industries have established practices as well. The software industry has Agile methodologies and the software development framework Scrum. Having learned and applied both Action Research (for improving practice) and Scrum (for developing products), I have become aware of the many similarities between the two. My following research paper takes a methodical approach in analyzing both Action Research and Scrum, and highlights areas where one could benefit from transferring concepts from one to the other.
Here are four outcomes from my investigations of Action Research and Scrum:
| Releases: | In Scrum, there should be a fully functional version of the product at the end of a release cycle. Applying this to Action Research, the researcher could aim to have a near feature complete version of the research paper by the end of each Action Research cycle and reiterate as long as time permits. If needed, a stakeholder could receive immediate results based on the researcher’s current level of knowledge. For example, a paper with two completed Action Research cycles, rather than a draft version. |
| Agility: | A real-world problem situation may change during an Action Research cycle. For example, new team members or technologies may be introduced. The researcher may avoid redundant improvements to a practice by continuously updating the definition of the problem situation. Like a Scrum backlog, the definition of the problem situation could be decoupled from the other processes and continuously be updated by stakeholders whenever new data is received. |
| Knowledge distribution: | In Action Research, achieving both the practical and the academic outcomes are considered ideal. In Scrum, sharing the optimizations to the production process beyond the team could be an additional task for the Scrum Master. Like the backlog, documentation of the process could be treated as an artifact. |
| Avoiding self-delusion and groupthink: | Action researchers are expected to address the handling of personal bias. The “devil’s advocate procedure”, known from Action Research can be applied to Scrum events such as the sprint retrospective. One or more team members taking the role of a devil’s advocate can constantly try to show that a theory does not apply and possibly uncover any lack of evidence. |
It is was interesting to uncover these four extensions to Action Research and Scrum during my investigations. I believe there is potential for further improvements to both, and I will keep similarities in mind while applying Action Research and Scrum in the future.