Deck 7: Multiple Regression

Question 1

Regression analysis procedures have as their primary purpose the development of an equation that can be used for predicting values on some DV for all members of a population.

Answer 1

True

Question 2

A secondary purpose is to use regression analysis as a means of explaining causal relationships among variables.

Answer 2

True

Question 3

In order to make predictions, three important facts about the regression line must be known. One of them is: The point at which the line crosses the X-axis.

Answer 3

False

Question 4

The regression line is essentially an equation that expresses X as a function of Y.

Question 5

Residuals (errors of prediction) are essentially calculated as the difference between the actual value and the predicted value for the IV.

Question 6

The reason that we obtain the best-fitting line as our regression equation is that we mathematically calculate the line with the smallest amount of total squared error.

Question 7

Multiple regression is used to predict the value of a single DV from a weighted, linear combination of IVs.

Question 8

The coefficient of determination in multiple regression is the proportion of DV variance that can be explained by at least one IV.

Question 9

Multicollinearity is desirable in multiple regression.

Question 10

Multicollinearity tends to increase the variances in regression coefficients, which ultimately results in a more stable prediction equation.

Question 11

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

Question 12

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).

Question 13

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

Question 14

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

Question 15

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

Question 16

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

Question 17

Multiple regression can be very sensitive to extreme outliers.

Question 18

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

Question 19

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

Question 20

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

Question 21

Partial correlation is a measure of the relationship between an IV and a DV, holding all other IVs constant.

Question 22

The other main calculation in multiple regression is the determination of the value for R² and its associated significance test.

Question 23

Interpretation of multiple regression focuses on determining the inadequacy of the regression model that has been developed.

Question 24

The multiple correlation (R) is a Pearson correlation coefficient between the predicted and actual scores of the IVs.

Question 25

The F test in multiple regression examines the degree to which the relationship between the IVs is linear.

Question 26

The final part of the multiple regression output is the coefficients table that represents the following:

A) The unstandardized regression coefficient (B).
B) The standardized regression coefficient (beta or β).
C) t and p values.
D) All the above.

Question 27

Three correlation coefficients are displayed in the coefficients table. They include the following:

A) The zero order correlation coefficient.
B) The partial correlation coefficient.
C) The part correlation coefficient.
D) All of the above.

Question 28

If the value for tolerance is acceptable, one should proceed with interpreting the:

A) Model summary.
B) ANOVA table.
C) Table of coefficients.
D) All of the above.

Question 29

The summary of multiple regression results should always include a description of how:

A) Variables have been transformed.
B) Cases have been deleted.
C) Both of the above.
D) None of the above.

Question 30

Typically, unless only a few variables are analyzed, which of the following descriptive statistics are presented in tables?

A) Correlation matrix.
B) Means.
C) Standard deviations for each variable.
D) All of the above.