What is the statistical measure of how strong the relationships are between attributes in a data set?

• A. Correlation
• B. Covariance
• C. Regression
• D. Distribution

Answer: (A)

The main idea here is how strongly two numeric attributes are related. Correlation provides a single, standardized measure of that linear relationship between two attributes. The correlation coefficient, r, ranges from -1 to 1: values near 1 mean a strong positive linear association (as one attribute increases, the other tends to increase), values near -1 mean a strong negative linear association (as one increases, the other tends to decrease), and values near 0 suggest little or no linear relationship. Because it standardizes by the variables’ variability, correlation is unitless and comparable across different scales, unlike covariance which depends on the units of measurement. Regression is about predicting one attribute from another and gives a slope and intercept rather than a pure strength measure, while distribution describes how a single attribute’s values are spread, not how two attributes relate to each other. Remember that correlation captures linear relationships well but can miss non-linear associations and can be affected by outliers.