Ch13 Test Bank + Answers Controlling For A Third Variable - Answer Key + Test Bank | Statistics for Criminology and Criminal Justice 5e by Bachman by Ronet D. Bachman. DOCX document preview.
Chapter 13: Controlling for a Third Variable: Multiple OLS Regression
Test Bank
Multiple Choice
1. Which of the following is NOT one of the things that need to exist before we conclude that a causal connection exists between two variables?
a. nonspuriousness
b. correlation
c. temporal order
d. multicollinearity
Learning Objective: 13.1. Explain the importance of controlling for multiple independent variables when predicting a dependent variable.
Cognitive Domain: Knowledge
Answer Location: What do we mean by controlling for other important variables?
Difficulty Level: Easy
2. Spuriousness occurs when ______.
a. a third variable causes both the independent and dependent variable
b. a third variable causes the dependent variable only
c. a third variable causes the independent variable only
d. a third is caused by both the independent and dependent variables
Learning Objective: 13.1. Explain the importance of controlling for multiple independent variables when predicting a dependent variable.
Cognitive Domain: Knowledge
Answer Location: What do we mean by controlling for other important variables?
Difficulty Level: Easy
3. In a study, researchers find that ice cream consumption predicts increased crime. However, once they control for summertime (which also predicts crime), the relationship goes away. This is an example of ______.
a. nonspuriousness
b. temporal order
c. correlation
d. spuriousness
Learning Objective: 13.1. Explain the importance of controlling for multiple independent variables when predicting a dependent variable.
Cognitive Domain: Application
Answer Location: What do we mean by controlling for other important variables?
Difficulty Level: Medium
4. Which of the following is NOT an assumption of the multivariate regression model?
a. The dependent variable is measured at the interval/ratio level.
b. The relationship between the independent and dependent variables is linear.
c. The independent variables are not highly correlated among themselves.
d. The error term is correlated with each of the independent variables.
Learning Objective: 13.1. Explain the importance of controlling for multiple independent variables when predicting a dependent variable.
Cognitive Domain: Knowledge
Answer Location: The multiple regression equation
Difficulty Level: Medium
5. A partial slope coefficient is ______.
a. the effect of the independent variable on the dependent variable
b. the effect of the independent variable on the dependent variable after controlling for other independent variables
c. the partial effect of the independent variable on the dependent variable
d. the effect of the dependent variable on the independent variable
Learning Objective: 13.2. Describe the difference between a bivariate correlation and a partial correlation coefficient.
Cognitive Domain: Knowledge
Answer Location: The multiple regression equation
Difficulty Level: Easy
6. When the independent variables are too highly correlated in a regression equation, they are considered to be?
a. partial coefficients
b. spurious
c. inter-leveraging
d. multicollinear
Learning Objective: 13.2. Describe the difference between a bivariate correlation and a partial correlation coefficient.
Cognitive Domain: Knowledge
Answer Location: The multiple regression equation
Difficulty Level: Easy
7. Why is multicollinearity a problem?
a. It induces a spurious relationship between the independent variables and the dependent variables.
b. It is difficult to identify which of the collinear variables are influencing the dependent variable.
c. It increases type I error.
d. It reduces the standard errors.
Learning Objective: 13.4. Interpret the hypothesis test for a multiple regression model, and understand how it is different from the hypotheses tests for partial slope coefficients.
Cognitive Domain: Knowledge
Answer Location: The multiple regression equation
Difficulty Level: Easy
8. One remedy for multicollinearity is to ______.
a. remove all of the troublesome variables
b. increase the sample size
c. remove one of the troublesome variables and re-estimate the equation
d. obtain the partial slope coefficient
Learning Objective: 13.4. Interpret the hypothesis test for a multiple regression model, and understand how it is different from the hypotheses tests for partial slope coefficients.
Cognitive Domain: Comprehension
Answer Location: The multiple regression equation
Difficulty Level: Medium
9. Given a multiple regression model with four independent variables, which of the following standardized partial slopes would have the strongest relationship with the dependent variable.
a. 3.25
b. 1.26
c. -2.36
d. -4.65.
Learning Objective: 13.7. Describe how the beta weights are different from unstandardized slope coefficients.
Cognitive Domain: Application
Answer Location: Comparing the strength of a relationship using beta weights
Difficulty Level: Easy
10. One can convert all variables in a regression equation for comparative purposes by calculating the ______.
a. beta weight
b. partial slope coefficient
c. partial regression coefficient
d. R2
Learning Objective: 13.7. Describe how the beta weights are different from unstandardized slope coefficients.
Cognitive Domain: Knowledge
Answer Location: Comparing the strength of a relationship using beta weights
Difficulty Level: Easy
11. Calculation of the standardized partial slopes is obtained by ______.
a. dividing the standard deviation of the dependent variable by the standard deviation of the independent variable
b. dividing the standard deviation of the independent variable by the standard deviation of the dependent variable
c. dividing the variance of the dependent variable by the variance of the independent variable
d. dividing the variance of the independent variable by the variance of the dependent variable
Learning Objective: 13.7. Describe how the beta weights are different from unstandardized slope coefficients.
Cognitive Domain: Knowledge
Answer Location: Comparing the strength of a relationship using beta weights
Difficulty Level: Medium
12. With a regression equation with two independent variables, an R2 of .42 indicates that ______.
a. both of the independent variables explain 18% of the variation in the dependent variable
b. both of the independent variables cause 42% of the dependent variable
c. both of the independent variables explain 42% of the variation in the dependent variable
d. both of the independent variables cause 18% of the dependent variable
Learning Objective: 13.3. Interpret a multiple coefficient of determination and be able to calculate the change in the multiple R2.
Cognitive Domain: Comprehension
Answer Location: Multiple coefficient of determination, R2
Difficulty Level: Easy
13. For Model 1, the R2 equals .23 and for Model 2 the R2 = .35. This means that ______.
a. Model 2 explains an extra 12% of the variation in the dependent variable
b. Model 1 explains an extra 12% of the variation in the dependent variable
c. Model 1 was spurious and thus not explaining enough of the variation in the dependent variable
d. Model 1 was multicollinear and thus not explaining enough of the variation in the dependent variable
Learning Objective: 13.3. Interpret a multiple coefficient of determination and be able to calculate the change in the multiple R2.
Cognitive Domain: Application
Answer Location: Multiple coefficient of determination, R2
Difficulty Level: Medium
14. The null hypothesis of a regression equation with three independent variables can be stated as ______.
a. 
b. 
c. 
Learning Objective: 13.4. Interpret the hypothesis test for a multiple regression model, and understand how it is different from the hypotheses tests for partial slope coefficients.
Cognitive Domain: Knowledge
Answer Location: Hypothesis testing in multiple regression
Difficulty Level: Easy
15. To obtain the t-statistic of a regression equation, one must ______.
a. divide the standard error of the slope by the partial slope coefficient
b. divide the partial slope coefficient by the standard error of the slope
c. divide the mean square for the regression model by the mean square for the residuals
d. divide the mean square for the residual by the mean square for the regression model
Learning Objective: 13.4. Interpret the hypothesis test for a multiple regression model, and understand how it is different from the hypotheses tests for partial slope coefficients.
Cognitive Domain: Comprehension
Answer Location: Hypothesis testing in multiple regression
Difficulty Level: Medium
16. To obtain the F-statistic of a regression equation, one must ______.
a. divide the standard error of the slope by the partial slope coefficient
b. divide the partial slope coefficient by the standard error of the slope
c. divide the mean square for the regression model by the mean square for the residuals
d. divide the mean square for the residual by the mean square for the regression model
Learning Objective: 13.4. Interpret the hypothesis test for a multiple regression model, and understand how it is different from the hypotheses tests for partial slope coefficients.
Cognitive Domain: Comprehension
Answer Location: Hypothesis testing in multiple regression
Difficulty Level: Medium
17. As compared to the R2, the adjusted R2 ______.
a. takes into account the number of observations
b. takes into account multicollinearity
c. takes into account spuriousness
d. takes into account the number of independent variables in the regression equation
Learning Objective: 13.3. Interpret a multiple coefficient of determination and be able to calculate the change in the multiple R2.
Cognitive Domain: Knowledge
Answer Location: Hypothesis testing in multiple regression
Difficulty Level: Medium
True/False
18. Since researchers cannot always conduct true experiments, one must control for spurious relationships.
Learning Objective: 13.1. Explain the importance of controlling for multiple independent variables when predicting a dependent variable.
Cognitive Domain: Knowledge
Answer Location: What do we mean by controlling for other important variables?
Difficulty Level: Easy
19. To use the multiple regression equation the dependent variables and all of the independent variables have to be measured at the interval or ratio level.
Learning Objective: 13.1. Explain the importance of controlling for multiple independent variables when predicting a dependent variable.
Cognitive Domain: Knowledge
Answer Location: The multiple regression equation
Difficulty Level: Easy
20. In order to compare the effects of two or more independent variables on a dependent variable the beta weight would be the proper statistic to use.
Learning Objective: 13.7. Describe how the beta weights are different from unstandardized slope coefficients.
Cognitive Domain: Knowledge
Answer Location: Comparing the strength of a relationship using beta weights
Difficulty Level: Easy
21. The magnitude of R2 does not change depending on which independent variable the researcher enters into the equation first.
Learning Objective: 13.3. Interpret a multiple coefficient of determination and be able to calculate the change in the multiple R2.
Cognitive Domain: Comprehension
Answer Location: Multiple coefficient of determination, R2
Difficulty Level: Medium
22. You can determine which independent variable has the strongest effect on the dependent variable by comparing unstandardized partial slope coefficients.
Learning Objective: 13.7. Describe how the beta weights are different from unstandardized slope coefficients.
Cognitive Domain: Comprehension
Answer Location: Comparing the strength of a relationship using beta weights
Difficulty Level: Easy
23. The following diagram suggests that the relationship between X and Y is spurious due to Z
Learning Objective: 13.1. Explain the importance of controlling for multiple independent variables when predicting a dependent variable.
Cognitive Domain: Application
Answer Location: What do we mean by controlling for other important variables?
Difficulty Level: Medium
24. The following diagram suggests that the relationship between X and Y is properly temporally ordered where the independent variables comes before the dependent variable.
Learning Objective: 13.1. Explain the importance of controlling for multiple independent variables when predicting a dependent variable.
Cognitive Domain: Application
Answer Location: What do we mean by controlling for other important variables?
Difficulty Level: Medium
25. The following diagram suggests that the relationship between X1 and Y would be spurious due to X2. Thus, to get accurate estimates of the effect of X1 on Y one must control for X2.
Learning Objective: 13.1. Explain the importance of controlling for multiple independent variables when predicting a dependent variable.
Cognitive Domain: Application
Answer Location: What do we mean by controlling for other important variables?
Difficulty Level: Medium
Essay
26. Explain the issues that arise with multicollinearity.
Learning Objective: 13.4. Interpret the hypothesis test for a multiple regression model, and understand how it is different from the hypotheses tests for partial slope coefficients.
Cognitive Domain: Comprehension
Answer Location: The multiple regression equation
Difficulty Level: Hard
27. A researcher is analyzing whether the rate of unlawful breaking is effected by the percent of poverty, percent male, percent nonwhite, and the percent of individuals between the ages of 15 and 24 residing in the census block. Given the following SPSS output, what would the researcher conclude? State the null and alternative hypotheses, discuss the F-statistic, the R2, R2 change, the beta coefficients, and which variable significantly affect the dependent variable.
Model Summary | |||||||||
Model | R | R Square | Adjusted R Square | Std. Error of the Estimate | Change Statistics | ||||
R Square Change | F Change | df1 | df2 | Sig. F Change | |||||
1 | .195a | .038 | .031 | 1843.30459 | .038 | 5.294 | 2 | 267 | .006 |
2 | .407b | .166 | .153 | 1723.12310 | .128 | 20.272 | 2 | 265 | .000 |
a. Predictors: (Constant), percentnonwhite, percentpoverty | |||||||||
b. Predictors: (Constant), percentnonwhite, percentpoverty, percent15to24, percentmale | |||||||||
ANOVAa | ||||||
Model | Sum of Squares | df | Mean Square | F | Sig. | |
1 | Regression | 35977621.414 | 2 | 17988810.707 | 5.294 | .006b |
Residual | 907205072.766 | 267 | 3397771.808 | |||
Total | 943182694.179 | 269 | ||||
2 | Regression | 156357090.146 | 4 | 39089272.537 | 13.165 | .000c |
Residual | 786825604.033 | 265 | 2969153.223 | |||
Total | 943182694.179 | 269 | ||||
a. Dependent Variable: rateofubev | ||||||
b. Predictors: (Constant), percentnonwhite, percentpoverty | ||||||
c. Predictors: (Constant), percentnonwhite, percentpoverty, percent15to24, percentmale | ||||||
Coefficientsa | ||||||
Model | Unstandardized Coefficients | Standardized Coefficients | t | Sig. | ||
B | Std. Error | Beta | ||||
1 | (Constant) | 76.230 | 258.075 | .295 | .768 | |
percentpoverty | 27.276 | 8.606 | .239 | 3.169 | .002 | |
percentnonwhite | -5.544 | 4.157 | -.101 | -1.334 | .183 | |
2 | (Constant) | -5080.529 | 1006.335 | -5.049 | .000 | |
percentpoverty | 21.254 | 8.173 | .186 | 2.601 | .010 | |
percentnonwhite | .246 | 3.991 | .004 | .062 | .951 | |
percent15to24 | -39.467 | 17.899 | -.126 | -2.205 | .028 | |
percentmale | 116.906 | 19.763 | .340 | 5.916 | .000 | |
a. Dependent Variable: rateofubev | ||||||
Learning Objective: 13.4. Interpret the hypothesis test for a multiple regression model, and understand how it is different from the hypotheses tests for partial slope coefficients.
Cognitive Domain: Comprehension; 13. 5. Calculate the multiple regression equation and be able to identify it from SPSS output.
Answer Location: Hypothesis testing in multiple regression
Difficulty Level: Hard
Document Information
Connected Book
Answer Key + Test Bank | Statistics for Criminology and Criminal Justice 5e by Bachman
By Ronet D. Bachman