Correlation isn’t cause


  • English
  • Arabic

Hollywood star Nicolas Cage may be guilty of many things – including making some poor movie choices – but we can say with confidence that he is not responsible for the high number of drownings in US swimming pools. This is despite the fact that, as The National reported yesterday, there is a correlation between the annual number of deaths in pools and the number of movies Cage filmed in each year between 1999 and 2009.

This observation was made by Harvard criminology student Tyler ­Vigen, who has also noted a correlation between the amount of cheese consumed and the number of people who died after being tangled in their bed sheets. Mr Vigen’s point is that correlation is not causation.

There is a serious point to this. Measles outbreaks are increasing because many parents have stopped vaccinating their children in the erroneous belief that the measles, mumps and rubella (MMR) vaccine causes autism. The belief persists despite the retraction in 2010 of the widely ­debunked research paper that pointed out a correlation, and studies since that have found no evidence of a link between MMR and autism.

Confusion over correlation can be entertaining – there are websites dedicated to funny examples – but not when it is holding back good science.