1. Analysis of Variance(ANOVA) is a statistical method of comparing the ________ of several populations.
2. Standard deviations
3. Variances
4. Means
5. None of the above

2. If H0: b1= b2 = b3 =0 is rejected, then we can conclude that there is no linear relationship between the dependent (y) and independent (x) variables in the model.
3. True
4. False

3. In an ANOVA, when testing the significance at α= 0.05, the p-value is 0.003, it can be concluded that:
4. there is no statistical evidence that any population mean is different from any other
5. no two population means are equal
6. no two variances are equal
7. there is strong statistical evidence that not all the population means are equal

4. When independent variables are correlated with each other, multicollinearity is present.
5. True
6. False

5. The F-ratio used to test for the existence of a linear relationship between the dependent variable and any independent variable is:
6. MSR/MSE
7. MSR/MST
8. MSE/MSR
9. None of the above

Use the information below to answer Questions 6 to 9.

Students were randomly selected from TSU undergraduate ECON 3050-01 Class.  Their undergraduate performance was compared to their high school GPA and scores on standardized tests. The multiple regression analysis output was as follows:

Analysis of Variance (ANOVA)

6. Using degrees of freedom (df), the total number of students (n) in this sample was:
7. 7 students
8. 8 students
9. 5 students
10. Cannot be determined

7. Assuming that high school GPA is (x₁) and standard score (x₂), the regression equation to predict the independent variable (y) is:
8. y = 0.4775 x1+ 0.0013392×2
9. y = 0.2059 + 0.1630×1 + 0.0006693×2
10. y = 1.1066 + 0.4775×1+ 0.0013392×2
11. Not enough information given

8. At the 5% level of significance, are high school GPA scores and standard scores significant?
9. Both are significant
10. Neither are significant
11. Only high school GPA is significant
12. Only standard scores are significant

9. What is the value of R2using the information from the above table?
10. 4%
11. 6%
12. 8%
13. Insufficient information to determine

10. When all members of every block are randomly assigned to all factors/treatments, the design is called:
11. Repeated measures design
12. Tukey design
13. One-way ANOVA
14. Randomized complete block design

11. Dummy variables are used when:
12. Qualitative variables are involved in the model
13. Quantitative variables are involved in the model
14. Making transformations of quantitative variables
15. None of the above

12. Multicollinearity may cause the signs of some estimated regression parameters to be the opposite of what we would expect.
    1. True
    2. False

13. As more independent variables are added to a multiple regression model, ___________ will increase; this is not always so with ___________, which will only increase if the additional variables add substantial explanatory power to the model.
14. R2; adjusted R2
15. adjusted R2; R2
16. Both a) and b)
17. None of the above

14. TSU wanted to see if there is a relationship between installation of security cameras and crime rate in hostels. A randomly select 5 hostels had security cameras set up at TSU.  If we see that crime has decreased in all 5 hostels, we can conclude that the security cameras caused the decrease in crime rate.
15. True
16. False

15. When the null hypothesis, H₀: b1= b2 = b3 = 0, is not rejected, the interpretation should be:
16. There is no linear relationship between y and any of the three independent variables
17. There is a regression relationship between y and at least one of the three independent variables
18. All three independent variables have equal slopes
19. There is a regression relationship between y and all three independent variables

16. A variance inflation factor (VIF) of ______ means there is no correlation between independent variables while a VIF exceeds __________ shows that there is enough correlation
17. 5 and 0
18. 5 and 1
19. 1 and 5
20. None of the above

17. If the purpose of the regression model is to provide a prediction for the dependent variable (y), the presence of multicollinearity is not necessarily a problem in using the model.
    1. True
    2. False

18. When One-Way ANOVA F-test is found to be significant, which statistical method is used as a follow-up procedure to determine means that are statistically different?
    1. t-test for related mean
    2. Tukey-Kramer test
    3. t-test for differences between independent means
    4. None of the above

19. In a one-Way ANOVA, if the computed F-statistic exceeds the critical F value, we may reject the null hypothesis since there is evidence that at least one of the means differs.
    1. True
    2. False

Consider the one-way ANOVA table below and answer Questions 20 and 21

20. What is the mean square error (MSE)?
21. 56
22. 88
23. 70
24. None of the above

21. Assuming there are equal number of observations in each factor (treatment), then the facto consists _________ observations
22. 3
23. 4
24. 6
25. 23

22. Stepwise regression is one of the ways to prevent the problem of multicollinearity.
23. True
24. False

23. The range of feasible values for the multiple coefficient of determination is from
24. 0 to 1
25. – 1 to + 1
26. – 1 to 0
27. None of the above

24. In order to test the significance of a single independent variable, we use:
25. t-test
26. The overall F-test
27. Adjusted R²
28. All of the above

For Questions 25 and 26 use the table below.  In the table are absolute differences in pairs of population means for x₁, x₂ and x₃ and their associate critical ranges (CRs).

25. From the table above and, using the decision rule for comparing pairs of population means, which of the following statement is correct?
26. Population means for are different
27. Population means for
28. Population means for
29. Neither of the above statements are correct

26. From the above table, we can conclude that population means for pairs
27. True
28. False

27. No matter how many groups are being compared, the F test from the one-way ANOVA uses only one significance test.
    1. True
    2. False

28. Qualitative data cannot be incorporated into linear regression models.
29. True
30. False

29. Which of the following iterative search procedures for model-building in a multiple regression analysis adds variables to model as it proceeds?
30. a) Backward elimination
31. b) Stepwise regression
32. c) Forward selection
33. d) All possible regressions

30. A one-way ANOVA uses 5 factors/treatments and a total of 40 observations. This means there are 35 degrees of freedom (df) within group variation
31. True
32. False

31. The ______ sum of squares measures the variability of the observed values around their respective treatment means.
32. Factor/treatment
33. Residual/error
34. Interaction
35. Total

32. One-way ANOVA partitions the total variation into “between” and “within” groups.
33. True
34. False
35. It is possible to test the effect of each factor in a two-way ANOVA
36. True
37. False

34. The table below shows correlation coefficients for variables in a multiple regression analysis. If the correlation coefficient was the chosen criterion to build regression model using forward selection procedure. The fist variable to be selected is

1. x1
2. x2
3. x3
4. x4

35. The table below shows correlation coefficients for variables in a multiple regression. The analysis reveals potential multicollinearity between which variables?

1. x4and x5
2. x1and x2
3. x1 and x4
4. x4and x3

36. The null hypothesis for conducting a one-way ANOVA is that “not all the means are equal.”
37. True
38. False
39. A two-way ANOVA examines the simultaneous effect that two main factors have on the observed data
40. True
41. False

For Questions 35 to 39, use the ANOVA summary table below

38. The value of sum squares between (a) is:
39. 76
40. 16
41. 60
42. 24

39. What is the total number of observations in this ANOVA analysis?
40. 24
41. 25
42. 20
43. None of the above

40. The mean sum of squares between(MSB) as represented by (b) is:
41. 15
42. 19
43. 4
44. 6

41. The mean sum squares within (MSW) shown as (d) is:
42. 20
43. 3
44. 8
45. Cannot be determined

42. What is the value of F-calculated?
43. 33
44. 1
45. 3
46. None of the above

43. The one-way ANOVA partitions the total variance into four components. These include variance attributable to (i) Factor A, (ii) Factor B, (iii) interaction (Factor A & B), and (iv) that which is unaccounted for.
44. True
45. False

44. How is the degree of association between a set of independent variables and a dependent variable measured?
45. Confidence intervals.
46. Autocorrelation
47. Coefficient of multiple determination
48. Standard error of estimate
49. The best example of a null hypothesis for testing an individual regression coefficient is:
53. None of the above

46. The Mean Square Error (MSE) is a biased estimator for the variance of the population, denoted by s2.
47. True
48. False

47. In testing for interaction between two factors (A and B) under a two-way ANOVA, the null hypothesis statement reads — H₀: Factors A and B do not interact. If we fail to reject the null hypothesis we can proceed testing Factors A and B.
48. True
49. False

48. Suppose that in a multiple regression the F is significant, but none of the t-ratios for independent coefficients are significant. This means that:
49. Multicollinearity may be present
50. The regression is good
51. Either a) or b)
52. None of the above

49. A factor in ANOVA describes the cause of variation in the data
50. True
51. False

50. The process of deciding which independent variable should be part of the final regression is known as
51. Model building
52. Residual analysis
53. Multicollinearity
54. None of the above