One of the first things statisticians are taught is that "Correlation is not causation." Correlation is often based on a range of 1 to minus 1. A strong negative correlation is minus 1. A perfect correlation would be 1. 0 represents no relationship between two numbers whatsoever.
However, people in general are not statisticians and hang on to the habit of confusing evemts that correlate with an event that is caused by the other event. A simple alternative is that a third factor causes both of the observed factors; this third wheel is hard to find and often called a "confounder."
The UK Guardian blog notes, "A famous example of correlation that was hard to prove existed in the 1950s in comparing smoking with lung cancer. There was a correlation, but there might be a confounder lurking to create the apparent causation. The Guardian blog notes, "As a seasonal example, just because people in the
UK tend to spend more in the shops when it's cold and less when it's hot doesn't mean cold weather causes frenzied high-street spending. A more plausible explanation would be that cold weather tends to coincide with Christmas and the new year sales."
Arguably the most well known and important example of a correlation being clear but caustion being in doubt concerned smoking and lung cancer in the 1950s.
There might be a confounder that was responsible for the correlation between smoking and lung cancer. The Guardian blog noted, "The increased rate could have been the result of better diagnosis, more industrial pollution or more cars on the roads belching noxious fumes. Perhaps people who were more genetically predisposed to want to smoke were also more susceptible to getting cancer?"
Finally, a study involving over 40,000 UK doctors demonstrated conclusively that smoking really does cause cancer.
Summarized from:
http://www.guardian.co.uk/science/blog/2012/jan/06/correlation-causation
However, people in general are not statisticians and hang on to the habit of confusing evemts that correlate with an event that is caused by the other event. A simple alternative is that a third factor causes both of the observed factors; this third wheel is hard to find and often called a "confounder."
The UK Guardian blog notes, "A famous example of correlation that was hard to prove existed in the 1950s in comparing smoking with lung cancer. There was a correlation, but there might be a confounder lurking to create the apparent causation. The Guardian blog notes, "As a seasonal example, just because people in the
UK tend to spend more in the shops when it's cold and less when it's hot doesn't mean cold weather causes frenzied high-street spending. A more plausible explanation would be that cold weather tends to coincide with Christmas and the new year sales."
Arguably the most well known and important example of a correlation being clear but caustion being in doubt concerned smoking and lung cancer in the 1950s.
There might be a confounder that was responsible for the correlation between smoking and lung cancer. The Guardian blog noted, "The increased rate could have been the result of better diagnosis, more industrial pollution or more cars on the roads belching noxious fumes. Perhaps people who were more genetically predisposed to want to smoke were also more susceptible to getting cancer?"
Finally, a study involving over 40,000 UK doctors demonstrated conclusively that smoking really does cause cancer.
Summarized from:
http://www.guardian.co.uk/science/blog/2012/jan/06/correlation-causation
No comments:
Post a Comment