When I gave the talks on “What is quality” I found it was not possible to provide a definition of quality on which everyone would agree. I recently read a book by Kaoru Ishikawa which included him describing quality using a cause-and-effect diagram[1]. It occurred to me that using a cause-and-effect diagram to describe something on which people have different opinions could be a way of building consensus. The effect the diagram was describing was quality and the causes the diagram showed were the causes of quality. This type of diagram is known as an Ishikawa or fishbone diagram. An illustrative example cause-and-effect diagram is at the top of this post.
It can be difficult to achieve a consensus can when defining quality. It can be that some team members have definitions of quality that they believe are true. Differences of opinion over definitions of quality can make discussion difficult and this may make it hard to gain agreement on a form of words to define quality.
If a cause and effect diagram is used to define quality then you can use the diagram to create consensus on what quality is. It could be that even if people disagree on the words that they use to describe quality when they add items to a diagram there is agreement on the items they add. A facilitator can then use this as a way to create a consensus.
The first step in creating a cause-and-effect diagram is to define the effect for which causes are to be identified. In this case, this is “Quality”. The name “Quality” should be placed in the box on the diagram. Each of the causes of quality can then be placed in a box off the spine of the diagram. Subsidiary causes can be added as branches of the causes that have already been defined.
If people break down the ways to describe quality to specific items such as the use of continuous integration, they can add the items to the diagram as causes of quality. The creation of the diagram could start by adding those causes of quality on which everyone agrees. It could be that the group want to add code reviews as a cause of quality and this can be added to the diagram. Once items that everyone agrees on have been added the group can discuss adding other causes on which they do not agree. This process narrows down the areas of disagreement.
Defining quality is important because if you can not define quality then how can you achieve quality? Using a cause-and-effect diagram to define quality could be a way to create a consensus on an important subject.
References
[1] What is Total Quality Control The Japanese Way by Kaoru Ishikawa (1985, p47)
TestAndAnalysis 