What does 'Multiple R' quantify in regression analysis?

What does 'Multiple R' quantify in regression analysis?

• A. The correlation between predictors and the outcome
• B. The correlation between observed and predicted outcomes
• C. The correlation among predictors
• D. The overall multiple correlation coefficient

Answer: (B)

Multiple R is about how well the model’s predictions line up with what was actually observed. After fitting a regression model, you get predicted values ŷ for each case. Multiple R is the Pearson correlation between those observed values Y and the predicted values ŷ. It captures the strength of the linear relationship between what happened and what the model predicts. A higher Multiple R means the predictions track the actual data closely; a lower one means the predictions are more off. This value is the square root of R², which tells you the proportion of variance in Y explained by the model. It’s not about how the predictors relate to each other or about the raw correlation between predictors and the outcome.

Multiple R is about how well the model’s predictions line up with what was actually observed. After fitting a regression model, you get predicted values ŷ for each case. Multiple R is the Pearson correlation between those observed values Y and the predicted values ŷ. It captures the strength of the linear relationship between what happened and what the model predicts. A higher Multiple R means the predictions track the actual data closely; a lower one means the predictions are more off. This value is the square root of R², which tells you the proportion of variance in Y explained by the model. It’s not about how the predictors relate to each other or about the raw correlation between predictors and the outcome.