Deck 18: Regression Analysis

Question 1

The partial correlation coefficient shows the portion of correlation between two variables that is not shared with other variables.

Answer 1

True

Question 2

Suppose that independent variable A explains 24 percent of the variation in a dependent variable.Suppose that independent variable B has a bivariate correlation of .40 with that same dependent variable, but that after controlling for variable A, it explains only 1 percent of the variation in the dependent variable.Then:

A)The partial r for variable B is .10.
B)The multiple correlation between both variables and the dependent variable is .50.
C)Multicollinearity exists between the two independent variables.
D)All of these.

Answer 2

All of these.

Question 3

We can find the multiple correlation between a set of independent variables and a dependent variable by adding up the bivariate correlations that each has with the dependent variable.

Answer 3

False

Question 4

Multivariate analysis looks simultaneously at the interrelationships of three or more variables.

Question 5

The standardized betas in a multiple regression equation:

A)Represent the slope of each predictor variable.
B)Represent the relative influence each predictor variable has in explaining the variation in the dependent variable when other predictor variables are controlled.
C)Represent the values on variable Y predicted from known particular X values.
D)Show the amount of dependent variable variation accounted for by each predictor variable.

Question 6

We can square the partial correlation coefficient to find the proportion of dependent variable variation uniquely attributable to the independent variable, when the effects of other variables are partialed out.

Question 7

Suppose we have hypothesized that even after controlling for variables A and B, variable C will have a significant and strong relationship with a dependent variable.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.

Question 8

Which of the following statements is/are true about the regression equation?

A)We use the symbol Y' instead of just Y to signify that we are dealing with a predicted value, not an actual value.
B)The a in the equation shows where the regression line intersects the vertical y-axis line.
C)The b in the equation shows us the slope we would use as we draw the line moving from our starting point along the y-axis to the right as the line moves higher or lower above the various x-axis values.
D)All of these.

Question 9

Which of the following statements is most correct about the multiple regression equation?

A)The dependent variable must be at the nominal level of measurement.
B)The dependent variable must be at the interval or ratio level of measurement.
C)All independent variables must be at the interval or ratio level of measurement.
D)At least one independent variable must be at the nominal level of measurement..

Question 10

The regression line is drawn where the sum of squared distances (or deviations) of the actual data points from the line will be minimized.

Question 11

Regression analysis is most useful when the correlation between X and Y is very weak.

Question 12

If three interrelated independent variables each has a .20 bivariate correlation with a dependent variable, then the multiple correlation is:

A).60
B).63
C).12
D)None of these.

Question 13

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.

Question 14

The more overlap among the independent variables in a multiple correlation analysis, the greater the sum of each variable's unique amount of shared variance with the dependent variable.

Question 15

Multiple correlation refers to the degree of correlation between a group of independent variables and a dependent variable.

Question 16

Which of the following statements is most correct about the multiple regression equation?

A)It predicts a variable Y value in light of a known value on variable X, the point where the regression line intersects the y-axis, and the slope of the regression line.
B)It predicts a variable Y value in light of a known value on variable X and the point where the regression line intersects the y-axis.
C)It predicts a variable Y value in light of a known value on variable X and the slope of the regression line
D)It predicts a variable Y value in light of the point where the regression line intersects the y-axis and the slope of the regression line.

Question 17

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.

Question 18

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.

Question 19

The regression equation predicts a variable Y value in light of: a known value on variable X, the point where the regression line intersects the y-axis, and the slope of the regression line.

Question 20

The regression equation:

A)Is used to predict one variable's values based on another variable's values.
B)Provides a precise prediction in every case.
C)Requires that all variables be at the nominal level of measurement.
D)All of these.

Question 21

The multiple regression equation extends the bivariate regression formula by adding more independent variables after bX, and each additional variable gets multiplied by its own slope.

Question 22

When 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, we should use the hierarchical multiple regression method.

Question 23

The greater the standardized beta weight, the greater influence a variable has in explaining the variation in the dependent variable when other variables are controlled.

Question 24

When we have hypothesized in advance which variables, or which sets of variables, are more influential than others in predicting the dependent variable, we should use the stepwise multiple regression method.

Question 25

15, In multiple regression analysis, the dependent variable must be at the interval or ratio level of measurement, but not all independent variables need be at that level.

Question 26

The greater the multicollinearity among the independent variables, the less distortion in the multiple regression findings.