Statistical tool used to measure the strength of a relationship between two (or more) variables.
A correlation can be :
– a coincidence.
Example : antibiotic consumption rate in United States correlated with violence in Bulgaria.
– caused by an other variable :
– A and B are both affected by C.
– excellent grades in High school (A) and excellent grades in University (B) are both affected by continuously hard working (C).
– a negative correlation :
– when A increases then B decreases.
– a positive correlation :
– when A increases then B increases.
Relation between a cause and it’s effect.
– A causes B.
– On Earth, dropping an object (A) causes it to fall (B).
We often say “correlation doesn’t imply causality“, which means that most of correlations won’t be a cause-effect matter, and so, won’t imply a causality.
Cum hoc ergo propter hoc is very common sophism in which a correlation is mistook as a causality, this leads to huge mistakes in thinking and establishing conclusion.
It can be also used as a manipulation, “correlation” or “causality” have to be used very carefully.
A correlation is only a statistical tool and is not useful as an argument, but, it’s always very interesting to study its possible causal link as it could lead to an interesting finding, and then, to an interesting argument.
Note : A causality WILL imply a correlation, in any case.
More : Find on this website funny correlations that will show you how a correlation differs from any cause-effect relation, such as : “US spending on science, space and technology” correlates with “suicide by hanging, strangulation and suffocation”
It also shows how easily graphics can be manipulated to make you believe that two things that have NOTHING in common can be related.