Correlation does not imply causation
|Part of the series on|
|—This, expressed in classic formular logic.|
Correlation does not imply causation is a quip that events or statistics that happen to coincide with each other are not necessarily causally related. The reality is that cause and effect can be indirect and due to a third factor known as confounding variables, or entirely coincidental and random. The assumption of causation is false when the only evidence available is simple correlation. To prove causation, a controlled experiment must be performed.
 Explanations and examples
 Simple example
Two events can consistently correlate with each other but not have any causal relationship. An example is the relationship between reading ability and shoe size across the whole population of the United States. If someone performed such a survey, they would find that the larger shoe sizes correlate with better reading ability, but this does not mean large feet cause good reading skills. Instead it's caused by the fact that young children have small feet and have not yet (or only recently) been taught to read. In this case, the two variables are more accurately correlated with a third: age.
The part age plays in this example is known as a "confounding variable" or "confounding factor," and is something that is not being controlled for in the experiment. In this case, age influences both reading ability and shoe size quite directly. A confounding variable can be what the actual cause of a correlation is, hence any studies must take these into account and find ways of dealing with them, usually by searching them out and trying to alter this variable directly.
The most common method to control confounding variables is with controlled studies. In these studies, the differences between the observations and the control group are minimised as best as possible, so that one can be more confident that a correlation is a valid indicator of causation. This is extremely important in compensating for the placebo effect in medical trials, but it is also important in other branches of science. In the age/shoe-size/reading-ability example, a controlled experiment would look for a correlation between reading ability and shoe size given a sample of people all the same age - or alternatively the hypothesis could be further tested by correlating age and reading ability given a sample of similar shoe size.
 Somewhat murkier example
“”Correlation does not imply causation, but it does waggle its eyebrows suggestively and gesture furtively while mouthing "look over there."
It's generally accepted in medicine that a diet rich in carbohydrates and fats coupled with sedentary behaviour, over time, increases the likelihood of (note well: not "causes") both diabetes and obesity. Under these circumstances, a positive correlation between diabetes and obesity is to be expected (and its absence would be noteworthy, and possibly informative). In the popular mind and press, this correlation has been misinterpreted as meaning that obesity itself causes diabetes.
In actual fact, the two conditions have different sets of causes, but share a common feature; obesity and diabetes are more likely to affect individuals with a particular diet and activity-level profile. So, the way to treat and avoid diabetes on a more permanent level would be to target the cause of the illness — diet and activity level — which also happens to affect weight. As weight can be a clear proxy target for this that you can see declining as your health improves (again, another rough correlation) it makes "losing weight" a considerably easier to identify goal than "avoiding diabetes."
 Risk factor
The term "risk factor" is used in medicine to mean "something that is positively correlated." So, in the above example, obesity is a risk factor for diabetes. The term is often incorrectly understood to mean "cause," and this can lead to people objecting to something being listed as a risk factor, arguing that it's not a cause. The term "risk factor" is symmetric; since obesity is a risk factor for diabetes, diabetes is also a risk factor for obesity. It is not, however, transitive; if A is a risk factor for B, and B is a risk factor for C, A is not necessarily a risk factor for C.
 In science
In science, correlation studies are often used to test for the existence of interesting patterns, but they are never used exclusively to claim a cause. In order to make a causal claim you must run an experiment or series of experiments and further studies using the scientific method — i.e., test to see if it really is a cause by altering parameters and performing more experiments, making predictions and testing them. This is in order to validate that one event is indeed directly influencing the other and is the reason behind the detected correlation.
Many woo and pseudoscience pushers conflate correlation with causation in order to make a claim of validity but forget to attempt the later scientific steps of compensating for confounding variables and thoroughly testing the causal relationship. For example, if someone gets a cold, but takes vitamin C, their cold will go away in 5-7 days. The claim is then made that the vitamin C caused the cold to go away. However, the cold would have gone away anyway, whether or not the vitamin C was taken, and so the validity claim is false. The placebo effect is another correlation with "treatment" that quacks use to create false validity.
Correlations seem to tap into a deep part of human psychology. As pattern recognition machines, we are hyper-responsive to any potential signal in our environment. People will often take two completely unrelated events and decide that they must cause each other because they seem to correlate. Someone may decide that when they wear a given shirt they have good luck; this is often combined with a powerful confirmation bias to create magical thinking.
 In parody
In the "Church of the Flying Spaghetti Monster", a key "belief" is that global warming is caused by a lack of pirates sailing the oceans. This is shown by a graph correlating increasing surface temperatures of the earth with a decline in the number of pirates. While it is certainly true that piracy has decreased and temperatures have gone up, there is nothing directly connecting the two trends.
 Fallacy engineering
Care should be taken not to assume the opposite (that correlation never implies causation). Woo spinners and quacks may try to walk around sound science or statistical analysis by citing this fallacy. Anti vaccine crusaders often discount all evidence that shows the use of vaccine followed by a reduction in the infection rate by claiming "correlation does not imply causation" as though "correlation cannot imply causation". Ironically they may then cite their own statistics showing the increase of cases of autism when a new vaccine is introducted. In such cases their evidence is flawed and such correlations absolutely utterly does not imply causation. Be very wary of climate change deniers and fanatical free market economists when they throw this fallacy on the table.
 See also
- ↑ xkcd - Correlation
- ↑ Unfortunately, as it turns out, one particular kind of obesity -- central obesity, marked by the presence of visceral fat -- is more strongly correlated with adult-onset diabetes than other kinds of obesity are. And while a high caloric intake and a low activity level correlate with obesity in general, they don't correlate with which kind of obesity the individual has.
| Articles about logical fallacies|
|Affirming the consequent||Denying the antecedent|
|Fallacious argument styles|