Question

Which of the following points are not true when conducting a multiple regression?

Select one:
a. Multiple regression can be used to assess quadratic relationships
b. Multiple regression can be used to assess linear relationships
c. Data must be homogeneous for a multiple regression
d. Data must be normally distributed for multiple regression
e. Data must be free from outliers for a multiple regression

Note: Answer C is NOT the correct answer. Please find the correct answer. Any answer without justification will be rejected automatically

Answer 1

The correct answer is d. Data must be normally distributed for multiple regression.

Justification: In multiple regression, the assumption of normality applies to the residuals (the differences between observed and predicted values). The residuals should be approximately normally distributed for the regression analysis to be valid. However, it is not necessary for the independent variables (predictors) or the dependent variable to be normally distributed. The other points listed are true for conducting multiple regression: a. Multiple regression can be used to assess quadratic relationships, b. Multiple regression can be used to assess linear relationships, and e. Data should be free from outliers for a multiple regression to be reliable.