Correlation is a necessary, but not sufficient condition for causation. If a is correlated with b, it means a might causes b, but you don't know for sure. Some

Correlation is a necessary, but not sufficient condition for causation. If a is correlated with b, it means a might causes b, but you don't know for sure. Some dum-dums think that if a is correlated with b, then it cannot cause b. in order to show causation, you need to go back in time and change things. This is called a counter-factual. Obviously we cannot do this. the best we can do is to control everything as closely as possible, changing only one factor at a time, or at least thinking we do. This is called the scientific method. Basically, you cannot really ever prove causation, but you can provide overwhelming evidence, along with isolated control over changing factors. Related http://stats.stackexchange.com/questions/36/correlation-does-not-mean-causation/77#77 http://stats.stackexchange.com/questions/36/correlation-does...