- Coefficient of determination - Glossary Term
In a linear model between two variables, the coefficient of determination (R2) is the proportion of the total variation that can be explained by the linear relationship existing between the two variables, usually expressed as a percentage. For two variables only, the coefficient of determination is numerically equal to the square of the correlation coefficient (r2).
A study finds that the correlation between the heart weight and body weight of a sample of mice is r = 0.765. The coefficient of determination = r2 = 0.7652 = 0.5852 … or approximately 59%
From this information, it can be concluded that approximately 59% of the variation in heart weights of these mice can be explained by the variation in their body weights.
Note: The coefficient of determination has a more general and more important meaning in considering relationships between more than two variables, but this is not a school level topic.