Back to Causal Thinking
Types of junctions
A three-node network with two links is called a junction. These are the building blocks of all Bayesian- and causal networks. There are three basic types of junctions, with the help of which we can characterise any pattern of arrows in a three node network.
- A → B → C (mediator or chain)
- A ← B → C (see confounder or fork)
- A → B ← C (see collider)
Mediator or chain
A → B → C
This is a 'chain' (or mediation). We think of B as the mechanism, or "mediator," that transmits the effect from A to C.
A well-known example is the smoke detector. Although we sometimes call them 'fire detectors', they are actually smoke detectors. The fire itself does not trigger an alarm, so there is no direct arrow from 'fire' to 'alarm'. The fire also does not cause a warning due to heat. It only works by releasing smoke molecules into the air. If we switch off that link in the chain, for example, by sucking all the smoke molecules away with a suction system, no alarm will go off.
Let us consider A = Innovation, B = Entrepreneurship, C = success. A → B → C is a business chain that can't succeed without entrepreneurship.
Another observation leads to an important conceptual point about chains: the mediator B “screens off” information about A from C, and vice versa. The core of an innovation (A) get lost when it is bought (B) and brought to succes (C). (1)
|(1)||The Book of Why - J. Pearl & D. Mackenzie - Basic Books - 2018|
Examples from the field
- Long term thinking → Ethical decision making → Operational Strategy
- Long term thinking will not transform into a sound strategy without decision making.