Maybe this doesn’t go in GQ, and I’ll seek it out elsewhere if the mods move it - but, I’m looking for a couple of simple and dramatic examples to show that correlation does not necessarily mean causation. Dopers - any help? xo, C.
The trees moving does not make the wind blow.
the famous example is that July has the most ice cream sales and also the most rapes. The sale of ice cream does not cause people to be raped.
Correlation is not causation when there is no demonstrable causative link between two phenomenon other than a co-occurance. IOW, you have to see the transmission turn in order to say unequivocally that the engine drives the wheels of a car. Another example I heard several years ago was that there was an increase in DUIs and in the number of Baptist ministers. Corr=Caus would imply that Baptist ministers were responsible for the DUIs, when in fact both were a reflection of an increasing population in the US.
Vlad/Igor
40% of automotive collisions are caused by drunk drivers. Therefore we can conclude that sobriety causes 60% of collisions.
Outside of a properly designed and controlled experiment, correlation never implies causation.
Wiki has some examples: http://en.wikipedia.org/wiki/Correlation_causation
On Monday, I drank gin and tonic, and got drunk.
On Tuesday, I drank vodka and tonic, and got drunk.
On Wednesday, I drank scotch and tonic, and got drunk.
On Thusday, I drank rum and tonic, and got drunk.
On Friday, I drank rye and tonic and got drunk.
Clearly, tonic water makes me drunk.
The more ice cream one eats, the lower the chance one will commit a crime. Thefore, ice cream stops crime.
8th graders are taller than 3rd graders. Therefore, one must grow tall in order to graduate.
Just think of any scenario where Z causes both X and Y. Then make the ridiculous statement that X causes Y.
At the risk of turning this into a great debate, another classic case might be that where the number of pirates in the world is inversely correlated with the rise in average global temperature.
Every day the sun comes up, a man is murdered.
That’s been true 365 days a year, probably for the last 10,000 years.
A more pernicious example is one in which the two elements being compared are correlated because they are not causal but are in fact two wordings (and/or two experimental operationalizations) of the exact same thing.
Being a freshman neither causes nor is caused by your being a 1st year student in college.
While this is obvious, there are plenty of less obvious situations. Language is messy and the experimental mechanism by which Variable 1 is implemented (“in our experiment, this is what it means to ‘be a freshman’…”) can look enough different from that used for Variable 2 that unless you are sufficiently familiar with the topic area of study you may not notice that the study has essentially proven that there’s a very high “correlation” between being a 1st year student and being a freshman among college students studied.
There are also those cases where the occurrence of thing x doesn’t prevent thing y – the fact that everybody that eats bread will die eventually doesn’t mean that bread is lethal.
TheDow Jones Average tends to rise in periods when womens’ hemlines also rise.
Blimey, that’s a bit grim. When I was a schoolkid they used a graph showing number of stork sightings and number of births in sweden over a many-year period, which had a close correlation even though the two don’t have anything to do with each other. A quick google suggests it’s still very widely used in teaching, although I’m not 100% sure the example is factual.
Thank you, all. Could you also come up with some good, dramatic examples of the *post hoc *falacy?
Post Hoc:
Often, when a lot of people bring umbrellas, it will start to rain.
The Super Bowl winner predicts if the stock market goes up or down.
Post hoc: The rooster crowing causes the sun to rise.
In children, there is a strong correlation between shoe size and reading ability.
Johnny Damon playing for the Red Sox correlated well with the Yankees losing games against them. Since baseball is not basketball, soccer, hockey, or any other symmetrical team sport, there doesn’t seem to have been a causal link between his presence and the Yankees inferior performance as a team-- unless you believe in juju, karma, voodoo, or effecting metaphysical change by having really long hair.