Researchers found that, when ice cream sales go up, murder rates in cities go up, and when ice cream sales go down, murder rates in cities go down. But there's no connection between those two things. They might be indirectly connected because they both increase with outdoor temperature, so that more ice cream is sold and more murders occur on hot days than on cold ones.
That's just one of the weird correlations that statisticians come across, which are part of the old truism that "correlation does not mean causation." There's even a website listing some of them, with graphs: http://tylervigen.com/. My favorite is the one about consumption of mozzarella and issuance of doctorates in civil engineering.
That's just one of the weird correlations that statisticians come across, which are part of the old truism that "correlation does not mean causation." There's even a website listing some of them, with graphs: http://tylervigen.com/. My favorite is the one about consumption of mozzarella and issuance of doctorates in civil engineering.