Correlation vs. Causation Lecture Notes

Vocabulary flashcards covering fundamental concepts in quantitative reasoning regarding statistical correlation, causation, and causal fallacies.

Correlation

A relation existing between phenomena, things, or mathematical/statistical variables that tend to vary, be associated, or occur together in a way not expected on the basis of chance alone, meaning when one changes, the other tends to change in a consistent manner.

Positive Correlation

A statistical relationship between two variables where they change in the same direction, such that when one variable increases, the other variable also tends to increase (e.g., height and weight).

Negative Correlation

A statistical relationship between two variables where an increase in one variable corresponds to a decrease in the other variable (e.g., height above sea level and temperature).

Causation

A cause-and-effect relationship between variables where changes in one variable bring about changes in the other, meaning the two variables are correlated and share a causal link.

Reverse Causation

A causal fallacy that occurs when the cause and effect of a situation are confused or inverted, characterized by the belief that $X \rightarrow Y$ ($X$ causes $Y$) when the reality is that $Y \rightarrow X$ ($Y$ causes $X$).

Common Causal Variable

A situation where two correlated events or measurements are falsely assumed to cause one another ($X \rightarrow Y$), but a third variable ($Z$) actually causes both ($Z \rightarrow X$ and $Z \rightarrow Y$).

Lurking Variable

A hidden or unobserved variable (represented as $Z$) that is not the presumed cause between two correlated events, but is actually responsible for causing both events to occur.

Oversimplification (Multiple Causes)

A fallacy that occurs when an effect is assumed to have a single cause ($A \rightarrow Z$), whereas in reality multiple contributing factors lead to the outcome ($A \rightarrow Z$, $B \rightarrow Z$, $C \rightarrow Z$, etc.).

Bidirectional Cause

A relationship where two events simultaneously cause one another, resulting in the reality $X \rightarrow Y$ and $Y \rightarrow X$ (exemplified by the predator/prey model of lions and gazelles).

Coincidence

A circumstance in which two events or measurements are correlated purely by chance, without any underlying causal relationship existing between them (belief: $X \rightarrow Z$, reality: $Y \rightarrow Z$).