---

What Are 'Correlation and Causation'?

Causation is a complete chain of cause and effect. Correlation means that the given measurements tend to be associated with each other.

Correlation is not Causation. Just because one measurement is associated with another, doesn't mean it was caused by it.

The more changes in a system, the harder it is to establish Causation.

The more you can isolate the change you make, the more you can tell if it really was the reason behind the results.

Josh Kaufman Explains 'Correlation and Causation'

Imagine a billiards table: if you know the exact position of every ball on the table and the details of the force applied to the cue ball (impact vector, impact force, location of impact, table friction, and air resistance), you can calculate exactly how the cue ball will travel and how it will affect other balls it hits along the way.

Professional billiards players are skilled at Mentally Simulating these relationships so well that they can rapidly clear entire tables.

That's Causality: a complete chain of cause and effect. Since it's possible to calculate a complete chain of causality, you can say that striking the cue ball caused a ball to fall into the corner pocket. If you hit the cue again in exactly the same way in exactly the same situation, you'll get the same result every single time.

Here's another thought experiment, using hypothetical data: people who suffer heart attacks eat, on average, 57 bacon double cheeseburgers every year.

Does eating bacon double cheeseburgers cause heart attacks? Not necessarily.

People who suffer heart attacks typically take 365 showers a year and blink their eyes 5.6 million times a year. Does taking showers and blinking your eyes cause heart attacks as well?

Correlation is not causation. Even if you notice that one measurement is highly associated with another, that does not prove that one thing caused the other.

Imagine you own a pizza parlor, and you create a 30-second television advertisement to air on local television. Shortly after the commercial goes live, you notice a 30% increase in sales. Did the advertisement cause the increase?

Not necessarily-the increase could be due to any number of factors. Maybe a convention was held in town that day, so there were more visitors in town than usual, and they needed a quick place to eat. Maybe school let out, and families went out to eat to celebrate. Maybe you offered a special 2-pizzas-for-the-price-of-1 promotion at the same time, and that's what actually brought more people in the door.

So many things happened at the same time, it's hard to be certain.

In fact, the commercial may have actually caused a decline in sales-maybe people found it unappealing or offensive, but another factor caused such a huge increase in sales that it overshadowed the decline.

Causation is always more difficult to prove than correlation.

When analyzing complex systems with many variables and Interdependencies, it's often extremely difficult to find true causality.

The more changes that happen in a system over a period of time, the higher the likelihood that more than one change had an impact on the result you're trying to analyze. Adjusting for known variables can help you isolate the potential causes of a change in your system.

For example, if you know that families go out to celebrate the end of school, or if an annual convention is coming up, you can adjust for that seasonality by using historical data.

The more you can isolate the change you made in the system from other factors, the more confidence you can have that the change you made intentionally actually caused the results you see.

Questions About 'Correlation and Causation'

• Are you finding causation, or correlation?
• How can you adjust for different variables to isolate the effects of the changes you make?

---

"Correlation isn't causation, but it sure is a hint."

Edward Tufte, statistician, information design expert , and professor at Yale University

---

---

Share this concept:

https://personalmba.com/correlation-causation/

---

---

---

---