Correlation does not equal causation…

Repeat it with me folks: “Correlation does NOT equal causation.”  I realized, whilst watching the news this morning, that this indeed a very important concept for people to really grasp.  I mean, I know it is important because I teach it in both my college Psych course and in Logic…but, I didn’t really realize how many people don’t really get the meaning of that phrase.

So, if two things are correlated, they do indeed have a relationship; for one example as one variable rises, so does the incident of another variable.  For a real world example, I’ll use this morning’s news blurb; A study has found a correlation between autism and rainy weather.

This means that kids raised in a place with more rainy weather had a greater incident of autism than those who were not.  Now…the common mistake is for someone to latch onto that and exclaim, “So, rainy weather causes autism!” No, no, no, no, no…no.

Just because something has a relationship does not speak to cause and effect.  You cannot make the claim that rainy weather causes autism, because that is not what the study found…all you can say is that rainy weather and autism rates are correlated.

This is something the news anchor didn’t seem to grasp and began to ridicule the study.  Now maybe the study was a poor one, I don’t know, but I do know that there was no claim made in the study that the rain causes autism.  Now, it is possible to study this further, which I’m sure will be done, but the point of this blog post is to hammer this fact into people’s head; just because there is a relationship between two things, it does not mean one caused the other.

The person being interviewed understood this, and gave an example of what could be happening (remember, this is just speculation); perhaps kids that live in rainy environs do not go outside nearly as much as those who live in sunny environs…that would have several possible impacts.  The rainy kids may watch much more TV, get less exercise, be vitamin D deficient, may not interact as much socially with others since they are stuck in the house, etc… OR perhaps there are chemicals in the rain, blah, blah, blah.

This mistake of thinking correlation equals causation is a very common one, especially in “scientific” research (“scientific” is in quotes because some supposedly scientific research is anything but).  Mainly it is misunderstood by people trying to interpret research, including reporting in the news media.

Here’s another example that every single one of my students knows, and loves: It is a fact that as ice cream sales rise, so does the number of deaths from drowning.  Now, the mistake would be to think that eating more ice cream causes more people to drown…what’s really going on?  Think about it.  That’s right; it’s summer time; ice cream sales and deaths from drowning are indeed correlated positively.  In summer, both variable are affected as more people buy ice cream in the summer, and more people also do water activities in the summer.

So, remember: Correlation does not equal causation.

Advertisements

3 Comments

Filed under Logic, Musings

3 responses to “Correlation does not equal causation…

  1. Carl Sachs

    True — but correlation and causation are highly correlated!

    (My apologies for a little Humean humor.)

  2. now, if they are indeed correlated, and rain does NOT cause autism, then there is only one possibility left: Autism causes rain 🙂

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s