Although it is one of the most repeated warnings, especially in the field of science, it also constitutes one of the most frequent cognitive errors or illusions. We refer to ” correlation does not imply causation ” (CINAC, Correlation is not a cause; acronyms that we should wear stamped on the shirt).
In a nutshell, what this warning describes is that if two facts occur at the same time or appear to be related to each other, it does not necessarily mean that one of the facts is the cause of the other. The classic example, which we talked about a few days ago, is homeopathy and “it works for me”: the patient takes homeopathy, his pathology improves and the patient infers that the cause of his improvement has been homeopathy (when there could be been anything else, such as a simple spontaneous remission). From Spurious Correlations, they show us how far we can reach absurd conclusions if we establish this kind of relationship with statistical data … like Nicolas Cage’s films avoid traffic accidents.
US spending on science and suicides
Cheese consumption and death from getting tangled in the sheets
The divorce rate in Maine and consumption of margarine
Consumption of mozzarella and doctorates in civil engineering
Apart from the cases already presented, other graphics can be created on the site by clicking on the names of the data sets.
The CINAC is one of the great assets of the scientific spirit, which together with the following, are the requirements that every scientist should also wear, printed on the shirt, as Alan Sokal points out in Beyond intellectual impostures :
To maintain a scientific perspective, a constant intellectual and emotional struggle against illusions is required; teleological and anthropomorphic thinking; erroneous assessments of probability, correlation and causation; the conception of non-existent models, and the tendency to seek confirmation rather than the refutation of our favorite theories.