Interpreting science 101

Alright, so I've talked a lot about increasing communication between the scientific community and the rest of the world, but how can we actually do that? As I pointed out in a previous post, increasing such communication is a two way street. Over the next few months, I'll be periodically introducing some basic tips on how to critically examine everything, from scientific papers to your local election advertisements! One of my favorite concepts in interpreting data was taught to me a very long time ago. In fact, this axiom is something I carry with me pretty much every day. ARE YOU READY?! Repeat after me...

Correlation does not imply causation.

Ah, isn't that beautiful? Believe it or not, applying this statement to anything you see on television or read in a newspaper will help you determine whether or not the claims they announce are true. To explain what this sentence means, I present an excellent example of this reasoning as described by the Simpsons:

Figure 1: Simpsons lessons are the best lessons!

Let's apply this concept to a different, more science oriented setting. Say your aunt gives you a new herbal tea, with the intention of curing your cold. If your cold goes away after drinking the tea for a week, did the tea cure your cold? Not necessarily! The cold could have resolved itself on its own, or you may have taken another medication that treated it! Correlation does not imply causation.

Alright, one more example. Let's say you're sitting in your car during traffic, and you're really hoping that the light will turn green. In desperation, you scream to yourself, "WILL YOU JUST TURN GREEN ALREADY?!" The light turns green. Did your screaming cause the light to change? Audience, now's your chance to write down your answer!

If you said no, you've clearly been paying attention! Correlation does not imply causation.

When you look at a news article, scientific or otherwise, you should always ask yourself two questions:

1. Do I believe that the claims made in this piece true? 2. If yes, how do I know they are true? If no, how do I know they are not true?

While correlation does not imply causation won't directly answer either of those questions, it's a great way of pumping the breaks and realizing that arguments or claims may not be as crystal clear as the author may want you to believe.

At the risk of having this post grow too long, the take away message is this: just because two events happen at the same time doesn't mean they influenced each other! Keep this point in mind and I will guarantee that your likelihood of getting bamboozled by some ridiculous claim will go down by at least 70%!