# Correlation Vs Causation

We know that correlation and causation talks about the relationship between 2 variables.

Correlation is a statistical measure to know the strength of relationship between 2 variables.

It will be measured using the below equation.

This formula states the measure of the strength of linear relationship between 2 variables.

We can think that, relationship means one variable makes some impact in another variable. This is called Causation.

But this does not mean that whenever the correlation coefficient is high, there is always a Causation.

Correlation does not mean Causation.

### Correlation With Causation:

The change in one variable does makes an impact in another variable.

For example height and weight.

### Correlation without Causation:

Sometimes the measure Correlation Coefficient is high but in real world scenario it does not mean anything. It means the data is purely a coincidence.

For example the correlation coefficient of a movie release of a famous actor and raining at the time of release is 0.9. But we know that this is purely a coincidence, and that actor or movie is nothing to do with the weather.

In this case Correlation does not mean any Causation. If we go with the correlation coefficients and make any model based on this statistics, that will fail for the new data set.

Hope you understand that Correlation need not to make a Causation

0

## 8 Replies to “Correlation Vs Causation”

Great example. Give an example for causation too. It will be nice

0
1. Thank you.
Below is the example for Causation:
For example the correlation coefficient of a movie release of a famous actor and raining at the time of release is 0.9. But we know that this is purely a coincidence, and that actor or movie is nothing to do with the weather.

0
2. Dinesh says:

Very well explained, thank you

0
3. Sri says:

Good Examples 🙂 Thanks for the explanation

0