The Curious Case of Correlation ≠ Causation

By marwanayache via Sensitive Context Blog

In my last post, I wrote about how not having enough contextual data can outright boggle the mind. Today, we’re going to read about something else that similarly boggles the mind, albeit not really related to any linguistic phenomena. It’s an interesting little logical fallacy in the field of statistics known as cum hoc ergo propter hoc, or more commonly, ‘correlation does not prove causation’.  Here, we define correlation as ‘when two things happen at the same time’, and causation as ‘when one thing causes the other’.

correlation causationThis logical fallacy is great at showing the glaring inaccuracies caused by lack of data on a specific subject, and how this lack can cause us to reach blindly for (often incorrect) conclusions in the proverbial fogginess of our mind. Additionally, the comedic factor here is amplified if you forego the law of parsimony (also known as Occam’s razor), which states that of all the possible solutions to a question or problem, the simplest one is most likely the truth.

Have you ever been in a recording booth or a really quiet place? If you’re in there for a long time your mind begins to create its own sounds. Essentially, you begin to hallucinate due to a lack of external stimuli. This is basically what goes on in the aforementioned logical fallacy: you end up compensating for a lack of data by drawing a perceived (and often inaccurate) connection between the sole items of data you have.

What does this have to do with a language blog? Essentially, it’s a great way of showing how a lack of the background information required for comprehension can yield wildly inaccurate knowledge. Dig this:

comparefeetDid you know that children with bigger feet are statistically better at spelling?  This is statistically true. Without additional contextual information, I could hypothesize that having larger foot-size means the children would perform better at sports and have better balance while carrying large and cumbersome schoolbags, making them less prone to falling over in bustling school hallways, making them less likely targets for bullies, leading to an inevitable increase in confidence, leading to better scholastic performance, and thereby, better spelling skills!

The truth is, it’s actually because children with larger feet are probably a lot older than children with smaller feet. Duh.

Did you know that you are more likely to get cancer if you always wear a seat-belt?

MORE . . .

4 responses

  1. […] The Curious Case of Correlation ≠ Causation […]

  2. Paper:
    “Cereal-induced gender selection? Most likely a multiple testing false positive”

  3. […] Cartoon via […]

Leave a Reply

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

You are commenting using your 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

%d bloggers like this: