Fishbone diagrams – methodological issues

Abstract

This article discusses methodological issues concerning the use of cause-effect analysis, in particular the fishbone diagram, also known as root cause analysis (RCA) in the areas of quality management and accident prevention. The method was tested in six organisations in a research project about prevention of occupational stress in knowledge work. The fishbone diagrams were used to tracking down causes of strain and enthusiasm. A total of 22 workshops were conducted with employees, middle managers and representatives from different staff functions. The qualitative analysis of selected bones from the fishbone diagrams shows that the diagrams only partially reflect the discussions at the workshops. This is partially caused by lack of appropriate stringency in the use of the method, but it is also caused by the participants’ lack of cognitive ability to model the complexity of the organizational and social domain. Consequently, it is difficult for the method to live up to the objective aspirations found in the method description. Despite the problems with precision and inner logic, the participants found the produced diagrams plausible and sensible. The article argues that is essential to decide on an ontological perspective before initiating fishbone workshops. If an objectivist perspective is chosen and the diagrams are seen as representations, the workshop moderator should stringently control the workshop and the participants should be carefully chosen to represent expertise within the problem domain – not just experts in their own working environment. If a generative, constructivist perspective is chosen and the diagrams are seen as reflections, the role of the moderator should be more reflexive and leave more work to the participants. Independent of the perspective, the moderator should carefully make sure that the diagrams are sufficiently detailed to retain the essence of the discussions after the workshop. The article concludes that it seems impossible to attain the objectives ideal of the method, but that by using the method it is possible to produce program theories that the participants find sensible and plausible. The critique of the method has wider implication. When fishbone diagrams are used in quality management and accident prevention (root cause analysis), the same type of methodological limitations apply. Therefore, the same type of caution should be taken when root cause analysis leaves the technical domain and attempts to model the organizational and social domain.