Multiple cause diagrams

Multiple cause diagrams are used to explore why changes or events happen in systems. They do not predict behaviour, but may give insights into the multiple causes of system behaviour and how to make undesirable behaviour less likely.

The elements of multiple cause diagrams are phrases, and arrows between them.

The rules for drawing multiple cause diagrams are:

• phrases may be things but as the diagram develops it is preferable to use variables associated with them, e.g. ‘poor teaching material’ might become ‘35% of teaching material is substandard’;
• arrows do not necessarily mean causes, but can be read as ‘contributes to’, ‘leads to’, ‘enables’, or similar terms;
• the diagram may be entirely sequential, or it may contain loops.

The guidelines for using multiple cause diagrams include:

• begin at the factor or event to be explained and work backwards;
• the arrows should be labelled;
• it is not necessary to put blobs around phrases;
• ensure that each causal link is clear, inserting any necessary intermediate variables or factors as necessary;
• these diagrams do not distinguish necessary and sufficient causes – if this is required the diagram will need annotating to show this;
• it is not necessary to draw a system boundary, but drawing the diagram may guide ideas about where the boundary lies;
• although these diagrams are similar to influence diagrams, they are different because they can be read sequentially rather than being a snapshot representation and they do not begin with the structure of the system.