money health weekly


Correlation Isn't Causation: Are People With Dogs Happier?

I read a book a while ago that said people with dogs were happier than people without dogs. It also said that people with cats were more educated than people without cats. If you are a dog person, your brain went straight to, "Yes! I am a happy person, but the heck with cat people; I'm educated." If you are a cat person you thought, "I knew I was smarter than those dumb dog people. Plus, I'm just as happy as anyone with a dog!" You can't help it; that's how we're wired!

Our brains perceive this as: dogs cause us to be happy and smart people get cats. That is no doubt what the person who wrote that wanted us to think. But is it true?


Correlation is basically a way to show how two different things move together. You hear this kind of language a lot in statistics. If something is positively correlated it means that when one things tends to happen, another thing tends to happen in the same direction. In economics, the demand for a product or service is positively correlated to the price. For example, the demand for college education is on the rise and so is the cost of college. If something is negatively correlated it means that when one thing tends to happen, another thing tends to happen in the opposite direction. Using economics as an example again, the supply of a product or service is negatively correlated with the price. For example, if a patent expires and more drug companies can supply a drug, the price of that drug tends to go down.


In both of the examples above, college education demand and college costs and patent expiration and drug prices, there is a strong case for a causation element as well. More people wanting college education means colleges can raise their prices. More companies selling a particular drug means that no one company can charge too much because the other companies will win more customers.

Causation means that something causes another thing to happen. Higher demand for college caused higher prices, more drug suppliers caused lower prices. A cue ball hitting an 8 ball causes the 8 ball to move.

Correlation ≠ Causation

Correlation and causation are not the same thing. If there is a cause and effect relationship, then those two things are correlated. But that doesn't always work in reverse. Just because there is a correlation does not mean that one caused the other. For example, if A and B are correlated, it doesn't mean that A caused B. It could be that B caused A, or there was a third variable, C, that caused both. Then again, it could be completely random with no connection.

For example, there is a positive correlation between the number of ice cream cones consumed (A) and swimming pool drowning deaths (B). I don't think anyone would think eating ice cream causes people to drown. It just so happens that both ice cream cone consumption and people going swimming tend to happen more in