Sunday, September 12, 2010

Reasoning: Association vs. Causation

I came across this concept in my Macro-Economics course about errors in reasoning in confusing association (or correlation) and causation.  The book concluded the following:
The fact that one event follows another does not necessarily mean that the first event caused the second event.
Association is where these two events occur, like voodoo dancing and stock market increases, and concluding that they are related.  Many people do this in their reasoning, especially in religious context, and it is flawed.  I know some people that do this with computers, and they struggle in building technical knowledge, because they build a lot of assumptions based on these associative reasoning.

Causation is when the cause-and-effect relationship is stable over time.  From a technical perspective, it would mean one can empirically prove the relationship, such as through a controlled environment.  It that is difficult to set up, at least document the consistent behavior.

So which side are you: Association or Causation?  Or are some things association and causation?