The basis of this fallacy is in assuming that when two events occur together one of them must be causing the other. If you’re thinking that this sounds a bit like the post hoc fallacy then you’re right, but where as with that fallacy the link is merely a temporal one, here it is more statistical.
Examples of this particular fallacy can be quite fun, as we’ll come to see. As a starter for ten it is a fact that ice cream sales strongly correlate with drownings. Hopefully you haven’t immediately jumped to the conclusion that ice creams cause people to drown or vice versa. The reality is that ice cream sales mostly occur in the summer months, which is also when most people go swimming and partake of other water based activities. So the two phenomena are linked by a third factor which is causal to both: nice weather.
Those of you who belong to the Church of the Flying Spaghetti Monster may be familiar with their belief that global warming is caused by a lack of pirates in the world; whilst this is a deliberately tongue in cheek statement the two certainly do correlate.
It is important to remember that sometimes a correlation does turn out to be causal. The only way we can know for certain though is if we conduct a trial carefully designed to control all other possible variables. If the correlation is still present then you can begin to be confident that the relationship is, indeed, causal.
There is a wonderful website called Spurious Correlations that demonstrates the pitfalls of this fallacy perfectly. For example, it clearly shows that the number of people who drown by falling into swimming pools correlates with the number of films that Nicolas Cage appears in per year; and that per capita cheese consumption correlates with the number of people who die by becoming entangled in their bedsheets.
I hope by now you can see the potential perils of assuming that two phenomena are related just because they seem to happen together superficially. In healthcare and science in general a huge part of the job is making sure that any relationship is actually causal. Is a drug really improving a symptom or is the person getting better on their own or are they perhaps being healed by another drug they’re also taking. Picking apart these matters is a genuine skill. In every day life it should be a bit more simple than that but don’t let your guard down: it has been clearly shown that making false assumptions strongly correlates with making a fool of yourself.