We’ve explored the relationship of correlation and causation before on the blog.
In particular, this commentary on DeBoer’s blog notes that — while correlation doesn’t prove causation — it might be a useful first step in discovering causation.
DeBoer argues for a difficult middle ground. He wants us to know (say it with me) that “correlation doesn’t prove causation.” AND he wants us to be reasonably skeptical, not thoughtlessly reactive.
On some occasions, we really ought to pay attention to correlation.
More Fun
I recently stumbled across a livelier way to explore this debate: a website called Spurious Correlations.
If you’d like to explore the correlation between — say — the number of letters in the winning word of the Scripps National Spelling Bee and — hmmm — the number of people killed by venomous spiders: this is definitely website for you.
Just so you know, the correlation of the divorce rate in Maine with per-capita consumption of margarine is higher than 99%.
