We're taught from a young age that correlation doesn't mean causation. But it's easy to get sucked in by studies claiming that, say, a more active sex life means a fatter wallet: Although it's often much more complicated than A-equals-B, correlational studies are much easier to process and share. But two brave researchers recently published a paper aiming to show how ridiculous we all look bragging about our chocolate consumption to all our closest Facebook friends.
The paper, which bears the catchy title "Linguistic Diversity and Traffic Accidents: Lessons from Statistical Studies of Cultural Traits," aims to link a country's linguistic diversity to the number of traffic accidents in the country, going through a whole host of irrelevant (but totally correlated) associations to get there. You see, linguistic diversity is associated with climate, which is influences siestas, which are linked to a language's morphological complexity, which is impacted by group size, which predicts levels of extramarital sex in a community...and so on and so forth, until we get to the obvious and indisputable fact that the presence of acacia trees is correlated with a greater number of traffic fatalities.
Part of the reason this study is so fun is that it points out how easy correlations are to find in an era when all kinds of databases are readily available to researchers looking to prove just about anything. After all, the paper's authors did control for "linguistic diversity, length of road network, GDP, distance from the equator, population size and population density." If you can't trust them, who can you trust?
But more importantly, it's a fantastic way to troll media outlets—including, ahem, us—that are often all too ready to take questionable studies at face value and need some linkbait. The authors warn that "the media typically exaggerate the implications of this type of finding and try to link it to current events rather than emphasize the long-term change implied in most studies." So remember: linguistic diversity definitely causes traffic accidents, and you should never trust a study you read about in a non-peer-reviewed outlet. Except for us. You can totally trust us.
[image via AP]