Section 14 Rules for joining maps

Very often we will have different fragments of information which share common variables, which we can join up to create more comprehensive maps. In this section we will look at four different ways to snap together causal maps. These four are enough to build up arbitrarily complicated causal maps.

This section introduces the four cases from a purely structural perspective. The rules in this section will probably seem trivial. They basically all say “yes, of course you can join together maps on their common variables”.

In the subsequent sections we will look at the more tricky question of how the functions which contain the causal content of the mini-maps can be combined.

14.1 The chaining rule

Given two causal maps with one common variable which is an influence variable in one map and a consequence variable in the other, you can derive the single map which chains them together, providing this does not create a loop.

So from this:

you can deduce this (and vice-versa):

This illustration uses a mini-map with two influence variables combining with another mini-map with three influence variables, but the number of influence variables is not important. The same applies to the other rules in this section.

We will deal with loops or cycles later.

14.2 The shared consequence rule

Given two causal maps with one common variable which is a consequence variable in both maps, you can derive the single map which combines them.

From this …

… we can derive this:

14.3 The shared influence rule

Given two causal maps with one common variable which is an influence variable in both maps, you can derive the single map which combines them.

From this:

derive this:

14.4 The shared arrow rule

Finally:

Given two maps which contain a co-terminal arrow, i.e. with a common influence variable and a common consequence variable, we can combine the maps on these two variables.

So if we hear:

heart disease has a causal effect on alcohol consumption

and again

heart disease has a causal effect on alcohol consumption

like this:

..… we can combine the mini-maps, like this:

Technical note

We could present these rules more formally, to provide a recursive definition of what we mean by “causal map”, like this:

  • If M is a mini-map (created by the mini-map coding rule), it is a causal map.
  • If C is a causal map and M is a mini-map, the map constructed by joining C to M by any of the rules in this section is a causal map.
  • Nothing else is a causal map.