Friday, June 8, 2012

Correlation vs Causation


Earlier we posted about a study that showed a correlation between dark chocolate and several health benefits.  Despite what the headlines read, I want to point out that the study showed correlation, not proof of causation.  It does not necessarily mean that eating more dark chocolate will make you healthier any more than sunspots affect skirt lengths or ice cream causes murder (above).  That is preposterous!  Clearly, ice cream does not cause murder.  Murder causes ice cream.  No wait.  That doesn't sound right either.  I guess it's possible that they both correlate to something else, like warm weather.

I'd just like to remind everyone that correlation is not the same as, nor does it imply causation.  Although it is true that causation correlates strongly to correlation, but that really isn't important.

Thank you for your attention.  Now back to your regularly scheduled program.

from G+

No comments:

Post a Comment