Which statement best describes correlation coefficients?

• A. They can only be used for categorical data.
• B. They quantify the strength and direction of the linear relationship between two numeric attributes.
• C. They always imply causation.
• D. They are not affected by the scale of the attributes.

Answer: (B)

Correlation coefficients quantify how strongly two numeric attributes are related in a linear way, and in which direction that relationship goes. They give a value between -1 and 1: a positive value means both attributes tend to increase together, a negative value means one tends to rise while the other falls, and larger magnitudes indicate a tighter, more linear association. This is about linear relationship, not causation—having a high correlation doesn’t prove that one variable causes changes in the other. Since correlation relies on numeric data, it isn’t applicable to purely categorical data. Also, while correlation is typically robust to linear changes in scale (units) and location, extreme outliers or non-linear relationships can distort it.