Correlation does not imply causation

A Fundamental Logical Error

"Correlation does not imply causation" is one of the most important principles of critical thinking and statistics.

A correlation simply means that two events or variables tend to occur together.

The fallacy is committed when we assume that because two things are correlated, one must have caused the other.

Clear Examples: The world is full of spurious correlations that can lead us to erroneous conclusions.

For example, there is a positive correlation between the number of ice creams sold and the number of shark attacks.

Does this mean that eating ice cream causes shark attacks? Obviously not.

There is a third variable, called a confounding variable, that causes both: summer heat.

In summer, people eat more ice cream and also swim more in the ocean.

The Direction of Causality: Even when there is a causal relationship, correlation alone does not tell us the direction of that causality.

For example, it has been observed that people with high self-esteem tend to be more successful.

Is it high self-esteem that causes success, or is it success that causes high self-esteem? The relationship is likely bidirectional, but correlation alone cannot determine this.

The Importance of Experimental Design

To establish causation, it is not enough to observe a correlation.

More rigoro