In a regression with several predictors, which statistic is commonly used to quantify the proportion of variance explained by the model?

• A. Eta-squared
• B. F-statistic
• C. p-value
• D. R-squared

Answer: (D)

R-squared tells you the proportion of variance in the outcome that the regression model accounts for when multiple predictors are included. It’s calculated as 1 minus the residual sum of squares divided by the total sum of squares, so it directly measures how much the model reduces error compared with simply predicting the mean. That makes it the standard way to express how much of the outcome’s variability is explained by the set of predictors as a whole. The F-statistic, while important for testing whether the model provides any improvement over a baseline, isn’t a direct measure of variance explained. The p-value indicates the likelihood of seeing the observed data if there were no real effect, not how much variance is explained. Eta-squared is a related concept used more commonly in ANOVA contexts and isn’t the usual measure for regression with several predictors.