causal loop diagram

You are here


Causal loop diagrams show the cause and effect relationships between the variables of a system. There are two basic feedback loops at the root of all systems behavior -- balancing and reinforcing loops. Balancing loops tend to keep the system in its current stage and reinforcing loops tend to compound change in one direction. These two loops are the building blocks for describing all complex social and economic systems.

A graphic language --
The power of causal loop diagrams is in their ability to capture the reasons systems behave the way they do and portray this understanding in a power graphic manner. Causal diagrams can be though of as a language. This language's syntax is built up from causal loops, which are like sentences constructed by linking together variables of importance and showing the causal relationships between them. Multiple loops can then form paragraphs that tell a story with the graphic representation.

Causal loop diagrams focus --
Causal loop diagrams concisely capture and communicate cause and effect relationships that can explain dynamic issues in a concise manner. What they do not do is provide a detailed representation of the structure producing the dynamics. That purpose is served by stocks and flows diagrams.

Why diagram a system --

  • To understand the actual workings of a system, how its outcomes are produced by the circular cause and effect relationships
  • To reveal the interrelationships between the parts of the system
  • To use this knowledge to make better decisions about how to achieve the desired results from changing a system
  • People can better manage relationships and systems that are visible and explicit rather than invisible and assumed
  • Produced by a group, a representation of a system can produce a revealing and full picture of reality and produce significant learning
  • Understanding the system enables the system to be worked on vs. working within the system - enabling the system to be improved in the results it produces

See: Kim, 1999. See feedback loop.