Affinity diagram

The affinity diagram is a business tool used to organize ideas and data. It is one of the Seven Management and Planning Tools. People have been grouping data into groups based on natural relationships for thousands of years; however, the term affinity diagram was devised by Jiro Kawakita in the 1960s and is sometimes referred to as the KJ Method.

The tool is commonly used within project management and allows large numbers of ideas stemming from brainstorming to be sorted into groups, based on their natural relationships, for review and analysis. It is also frequently used in contextual inquiry as a way to organize notes and insights from field interviews. It can also be used for organizing other freeform comments, such as open-ended survey responses, support call logs, or other qualitative data.

This method, identified by the initials of its author, a Japanese anthropologist , Pr. Jirō Kawakita , is very frequently used in the animation of the working groups after the phase of ideation or generation of ideas, to organize the ideas and structure them by themes.

The KJ makes it possible to identify the origin of a problem based on the facts, to identify priorities.

The KJ is a powerful tool for structuring language data, a simple method for clearly modeling and grouping complex problems, ideas, issues or opinions related to various issues and creating links between them.

The KJ diagram is part of the 7 quality or management tools (called the “7M”) released in 1976 by the Japanese as well as the Pareto diagram, the cause-effect diagram, stratification, checklist, histogram, scatter plot, graph / control chart.

It is a method used by managers, which has greatly contributed to the wealth and organization of collective thinking and the establishment of a true team spirit, and especially a field method.

In practice, the exercise consists in answering a question on cards or Post-it notes, taking care to have only one information per support. The latter will then be grouped by theme into clusters and renamed by all the actors in order to quickly organize ideas and structure them.

In 1990, Prof. Shoji Shiba from Tsukuba University and MIT developed a more detailed version, which he called the Language Processing LP method.

The affinity diagram organizes ideas with following steps:

Record each idea on cards or notes.
Look for ideas that seem to be related.
Sort cards into groups until all cards have been used.
Once the cards have been sorted into groups the team may sort large clusters into subgroups for easier management and analysis. Once completed, the affinity diagram may be used to create a cause and effect diagram.

In many cases, the best results tend to be achieved when the activity is completed by a cross-functional team, including key stakeholders. The process requires becoming deeply immersed in the data, which has benefits beyond the tangible deliverables.

Phase diagram building steps
For the affinity diagrams, the following main steps are taken: ,

description and formalization of the problem, in order to establish the topic;
the idea or generation of ideas, which are recorded on cards (cards);
aggregating ideas (from charts) across families and naming titles for each group of charts;
relationships between all records are determined;
mapping (diagramming) for a global vision;
evaluation and conclusion.
The affinity chart is useful when addressing a topic that is not well understood or on which no conclusion is reached. It can also be used to determine the probable results of a problem being solved.

Stage description
1. Identify the topic that is proposed to be addressed, the objectives or the problem involved.
2. The verbal data relating to the proposed theme or case is collected. The verbal data consists of facts, reasoning, ideas or opinions resulting from a brainstorming process or session organized by the project manager.
3. Create data cards (cards). Each verbal date (information item) is written on a separate piece of paper.
4. Organize data cards. On a larger sheet of paper, place the data cards so they do not overlap with each other.
5. Affinity cards are created . For this purpose, a label (one header) is given for each group or cluster in the form of a short but complete title referring to the group’s characteristics. This new card labeled for a group is called an affinity card .
6. Order in groups of affinity cards. Identifying the affinity relationships between the constituted groups is made. By aggregating the groups created by several participants, you can see how close the relationship between the concepts is.
7. Prepare the affinity chart that provides a global view of the problem so detailed.
8. Finally, a decision / conclusion process takes place, assuming a weighting of the elements of the problem in the relationship.

The 5 key steps of the KJ method
The method has 3 aspects: a model for problem solving, a model for qualitative data formulation and analysis, and a new proactive concept for field research.

The basis of realization of a KJ is a cycle, of 5 or 6 stages, which can be renewed until satisfaction. Two cycles will be necessary in problem solving: the first to define the problem, the second to find the solution (s).

All facts and information are recorded on cards (or possibly post-its): – information by medium, – factual data, which can be verified, important.

Group and name
– The cards are mixed or battered, then read carefully. – Then the cards are sorted and grouped when they seem to belong to the same group of information. The cluster thus formed receives a name that is written on a different color map that is placed on the cluster. – This group formation process will have to be repeated up to 9 clusters.

– At this point (in a team work) the cards are retrieved by the facilitator and redistributed so that they do not return to the hands of the person who wrote them. – Each card will be read aloud. Each contributor will then check if in his “hand” is not a card that forms a “group” with the card that has just been read. – A name will then be chosen carefully by large clusters, neither too vague nor aggregated cards constituting it, the most precise possible.

Mapping an overview
Now that the information is clustered in up to 9 groups, each consisting of subgroups, sub-subgroups … move to the defining step of presenting this dataset in a visual form – on a large sheet of paper attached to a wall in a dedicated place for the time that will solve the problem, – in a harmonious way. As in mind mapping, this mapping will allow everyone to have a global vision of the problem as well as detailed. The reason why the number of final clusters should not exceed 9 is that our brains can not process on average 7 ± 2 pieces of information.

Everyone will express what they feel by reading this card, keeping it in writing, while paying attention to distinguishing the facts from the interpretation he / she makes. This step may take several days.

Ideas for solutions are often found and developed explaining the structure of the problem.

Source from Wikipedia