Which of the following is an assumption of bivariate regression analysis, but NOT an assumption of multiple regression analysis?
a. The dependent variable must be at the interval or ratio level of measurement.
b. The data must be reasonably linear.
c. Each independent variable must be at the interval or ratio level of measurement.
d. The degree of multicollinearity must not be excessive.
Suppose we have a large number of possible independent variables that we want to explore, and we want to find a smaller set of these variables to use in predicting the dependent variable, eliminating other variables that add only an insignificant or trivial amount of explained variation in the dependent variable beyond the smaller set. Which method should we use regarding the order in which predictor variables are entered into our multiple regression analysis?
a. The all-possible-subsets method.
b. The hierarchical method.
c. The stepwise method.
d. The partial method.
Suppose two interrelated independent variables each has a .30 bivariate correlation with a dependent variable, and their combined multiple correlation with the dependent variable is .40 . Then:
a. The two variables combined account for 60 percent of the variation in the dependent variable.
b. The two variables combined account for 18 percent of the variation in the dependent variable.
c. The two variables combined account for 16 percent of the variation in the dependent variable.
d. The two variables combined account for 70 of the variation in the dependent variable.
Quick Reply
