Okay. Today’s “Things I’ve Learned @ Brescia” is just a short one, but it’s a VERY important one, especially for those of us in psychology.

#4 Correlation does not necessarily equal causation.

I feel like this sums up SO much of every class I have every taken in psychology, and is one of the main concepts that students (even in the upper years) still struggle with. What does it mean? Basically:

Just because two things happen together does NOT mean that one causes the other.

Consider the following (made up) statement:

• People who are nice tend to have more friends.

When we read a statement like this, we tend to read it causally. In other words, we tend to read it as the first thing causing the second thing, like:

• If you are nice, then you will have more friends.

However, just because  niceness and number of friends are related does NOT mean that one causes the other. For example, what if:

• If you have lots of friends, THEN you are more likely to be nice? (This is reverse causation)
• If you have positive attitude, THEN you are more likely to both have friends and be nice. (This is a “third variable”- something that causes both the number of friends and the niceness.)

I think we can get especially tricked because often causal statements seem “common sense”. When a statement about a correlation seems to “flow”, then we just assume it must be causal. However, when we pick a correlation that seems a little more bizarre, this starts to become clearer. For example:

• On the days that people eat more ice cream, more people drown.

Wait a minute. Is this saying that if we eat ice cream, we are more likely to drown? And that if we stopped eating ice cream, maybe we would reduce our risk of drowning?

You probably agree that it doesn’t make sense that eating more ice cream CAUSES drownings. It also doesn’t seem likely that the reverse is true: that people eat more ice cream because of drownings. However, as the weather gets hotter, people buy more ice cream. As well, more people go swimming. And it is the increase in SWIMMING that causes more people to drown.

As another example, consider this (made up) statement:

• People who don’t exercise are more likely to be sick.

Again, we would often read this as “not exercising makes you sick”. However, consider how the reverse may be true. If you are ill (especially if you have a severe illness), how likely are YOU to exercise? (I don’t know about you, but when I have the stomach flu I am not exactly in the mood to hit the treadmill). Maybe it’s that, if you are sick, then that CAUSES you to exercise less. Or maybe it’s a third variable, like “tiredness”. Maybe people who don’t get enough sleep tend not to exercise (because they are tired) and also tend to get sick (because the exhaustion makes them ill).

See if you can pick out correlational statements that might be  interpreted as causal (Hint: MUCH of the media does this, especially when they are reporting on “scientific” findings). When you see it, ask: is there REALLY evidence that one causes the other? Or are these two things just related? You’ll be surprised at how much we mistake correlation for causation!