Correlation is not causation.

Correlation is not causation.

Perhaps the four most important words to say during any discussion of statistics, data science, analytics, or any other time one attempts to find relationships between things.

Correlation is not causation. (Emphasis added)

What does this mean? What is correlation?

Correlation is simply a mathematical way to describe the relationship between two things or two objects or two ideas.

Suppose we discuss my tax rate and my happiness. Taxes are very easy to measure, simply check my financial records! Happiness is more difficult, though nevertheless it may be measured. Think of the one-to-five Likert (pronounced ‘lick- ert’) scale from strongly disagree to strongly agree for the “Are you happy now?” question. We would say (unless you’re rather odd) that taxes and happiness are negatively correlated. In other words, as taxes go up, happiness goes down. Alternatively, when taxes decrease, happiness increases!

Ice cream and happiness, on the other hand, are positively correlated. As quantity of ice cream consumed increases, happiness does too. And, as all denizens of Texas know, when a certain brand of ice cream goes away, happiness decreases fairly drastically.

Correlation in statistics is calculated as a number between [-1 , 1] which is of course inclusive of the endpoints. A perfect positive correlation is 1, while a perfect negative correlation is -1. Anything that is 0 has no correlation (I’ll save for a far distant post that I’m discussing linear correlation here).

There are rather clever statistical ways to properly measure correlation of various data. That being said, perhaps higher than 0.45 to 1 might be considered trending towards significant and we might label correlations between -1 and -0.45 to the left of 0 as trending towards significant as well.

Correlation is not causation.

Just because two things have a mathematically linear relationship does not mean there is cause and effect.

This is important!

Correlation is not causation!

The next time I or someone close to you tells you two things are related, ask them what the lurking variables might be. If we’re honest, they won’t mind having a discussion with you about their thoughts or listen to yours on the matter.

If they’re not, well, always remember this: 100% of people who get divorced got married first. Also, cancer might cause cell phones:

Never be shy about asking why someone thinks one part of a puzzle causes or influences another part. If I can only tell you it’s because they’re correlated, feel free to ignore me.

Correlation is not causation.



Matt Wiley is a tenured, associate professor of mathematics with awards in both mathematics education and honour student engagement. He earned an assortment of degrees in computer science, business, and pure mathematics from the University of California and Texas A&M systems. He is the director of quality enhancement at Victoria College, assisting in the development and implementation of a comprehensive assessment program to enhance institutional performance outcomes. A programmer, a published author, a mathematician, and a transformational leader, Matt has always melded his passion for writing with his joy of logical problem solving and data science. From the boardroom to the classroom, he enjoys finding dynamic ways to partner with interdisciplinary and diverse teams to make complex ideas and projects understandable and solvable.

One comment

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 )

Google photo

You are commenting using your Google 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 )

Connecting to %s