Question

1. Tolerance is a measure of collinearity among IVs, where possible values range from 0–1. T F

2. The variance inflation factor (VIF) for a given predictor “indicates whether there exists a strong linear association between it and all remaining predictors” (Stevens, 2001). T F

3. In standard multiple regression, the IV that has the highest correlation with the DV is entered into the analysis first. T F

4. Sequential multiple regression is also sometimes referred to as statistical multiple regression. T F

5. Stepwise multiple regression is often used in studies that are explanatory in nature. T F

6. Model validation, sometimes called model cross-validation, is an important issue in multiple regression. T F

7. Multiple regression can be very sensitive to extreme outliers. T F

8. One of the assumptions in multiple regression with regard to the raw scale variables is that the IVs are normally distributed. T F

9. Another assumption in multiple regression with regard to the residuals is that the errors are correlated with the IVs. T   F

10. In cases that involve moderate violations of linearity and homoscedasticity, one should be aware that these violations weaken and invalidate the regression analysis. T F

Answer 1

1.Tolerance is a measure of collinearity among IVs, where possible values range from 0–1.

TRUE

2.The variance inflation factor (VIF) for a given predictor “indicates whether there exists a strong linear association between it and all remaining predictors” (Stevens, 2001).

TRUE

Note: VIF is a used to measure the multicollinearity in the model. Hence in turn it calculates the collinearity between the predictor and all remaining predictors

3. In standard multiple regression, the IV that has the highest correlation with the DV is entered into the analysis first.

FALSE. (We remove the variables that has no correlation or has a less significant contribution to the DV but the highest correlated IV doesnt go into analysis first).

4. Sequential multiple regression is also sometimes referred to as statistical multiple regression.

FALSE. (Statistical multiple regression is also known as simultaneous multiple regression)

5. Stepwise multiple regression is often used in studies that are explanatory in nature.

FALSE. (sequential multiple regression is used in studies that are explanatory in nature)

6. Model validation, sometimes called model cross-validation, is an important issue in multiple regression.

TRUE. (yes it is important since we check for all the assumptions to be satisfied in model validation).

7. Multiple regression can be very sensitive to extreme outliers.

TRUE. (YES, it can be very sensitive since the outlier may pull the least squares line towards it which might affect the model.)

8. One of the assumptions in multiple regression with regard to the raw scale variables is that the IVs are normally distributed.

FALSE. (We make no assumption about the distribution of IVs but we assume that the errors are normally distributed.)

9. Another assumption in multiple regression with regard to the residuals is that the errors are correlated with the IVs.

FALSE. (Errors need not be correlated with the IVs)

10. In cases that involve moderate violations of linearity and homoscedasticity, one should be aware that these violations weaken and invalidate the regression analysis.

TRUE. ( One of our important assumptions is the homoscedastic nature of the regression analysis which in turn if not satisfy invalidates the regression analysis.)