Chapter 16 – Quantitative Data Analysis: Hypothesis Testing

1. To determine the F-ratio, the variance between groups (‘between-groups mean square’) is divided by the variance within groups (‘within-groups mean square’).

a. T

b. F

2. ANOVA is used to determine if a non-metric variable influences a metric variable.

a. T

b. F

3. Which test could best be used to test the hypothesis that the average grade of students is higher than a 7 on a 10-point scale?

a. Independent samples t-test.

b. One sample t-test.

c. Paired samples t-test.

d. ANOVA.

4. Which test could best be used to test the hypothesis that the quality of tangible elements (measured on ratio-level) influences satisfaction on the University restaurant (measured on ration-level), but that the relationship between these variables differs between students and faculty?

a. Multiple regression analysis with an interaction-effect.

b. ANOVA.

c. Three multiple regression models so that mediation can be tested.

d. MANOVA.

5. A researcher is interested in whether customer loyalty (measured on a 5-point Likert scale) of people with a bonus card differs from loyalty of people without a bonus card. To test this he can make use of:

a. An independent samples t-test.

b. ANOVA.

c. A paired-samples t-test.

d. A Chi-square test.

6. Multiple regression analysis measures the effect of an independent variable on the dependent variable, while the other independent variables are held constant.

a. T

b. F

7. A researcher wants to know if the distribution of income in his sample corresponds to the real distribution in his population (60% < the standard, 40% >= above the standard). Which of the following tests could he use for this?

a. An independent samples t-test.

b. ANOVA

c. A paired-samples t-test.

d. A Chi-square test.

8. Which test could best be used to test the hypothesis that students that for a certain course attended tutorials from lecturer X are happier (measured on interval-level) than students who attended tutorials from lecturer Y and Z?

a. Regression analysis.

b. Independent samples t-test.

c. Chi-square test.

d. ANOVA.

9. Multicollinearity can be verified by a correlation matrix.

a. T

b. F

10. Multicollinearity is a problem when the VIF values are higher than 10.

a. T

b. F

11. VIF = 1 divided by the tolerance.

a. T

b. F

12. An ANOVA conducted on 2 groups essentially is the same as conducting a t-test.

a. T

b. F

13. When you want to test if the average score of an exam significantly differs from 7.3 (the average score of last year), an independent samples t-test is most appropriate.

a. T

b. F

14. A researcher want to test if a cause-effect relationships exists between someone’s age and the number of hours of sleep someone needs. She measures age as follows:

0 < 20 years.

0 20 - 50 years.

0 > 50 years.

Furthermore, participant of the research can fill in the number of hours they sleep on average in a day on a dotted line.

Which test could best be used to test the hypothesis that age influences the number of hours someone sleeps?

a. A independent samples t-test.

b. A paired samples t-test.

c. ANOVA.

d. Discriminant analysis.

15. The statistical null hypothesis is rejected when the p-value is small enough.

a. T

b. F

16. To be able to conduct a t-test it is important:

a. That the independent- as well as the dependent variable is measured on a nominal or ordinal scale.

b. That the independent variable is measured on an interval- or ratio scale and the dependent variable is measured on a nominal- or ordinal scale.

c. That the independent- as well as the dependent variable is measured on an interval- or ratio scale.

d. That the dependent variable is measured on an interval- or ratio scale and the independent variable is measured on a nominal- or ordinal scale.

17. A possible solution to deal with multicollinearity is to collect more data.

a. T

b. F

18. Regression analysis is used when the independent variable is nominal or ordinal in nature.

a. T

b. F

19. Regression analysis is used when the dependent variable is nominal or ordinal in nature.

a. T

b. F

20. The objective of regression analysis is to explain a dependent variable based on 1 or more independent variables.

a. T

b. F

21. A Type I error, also referred to as alpha (α) is the probability of failing to reject the null hypothesis given that the alternative hypothesis is actually true.

a. T

b. F

22. A Type II error, also referred to as beta (β), is the probability of rejecting the null hypothesis when it is actually true.

a. T

b. F

23. Statistical power depends on alpha (α), beta (β), and the size of the sample.

a. T

b. F

24. The most appropriate way to test if social class influences the political party for which people vote is regression-analysis.

a. T

b. F

25. ANOVA is used to determine if the mean between three or more groups significantly differ from each other.

a. T

b. F

26. The most appropriate technique to establish whether the average age of first-year students is higher than 19 is an independent samples t-test.

a. T

b. F

27. The most appropriate technique to test the hypothesis that the average price that students are willing to pay for a meal at the University restaurant differs between American and Chinese students is a Chi-square test.

a. T

b. F

28. A researcher tests if the loyalty (measured on a 5-point Likert scale) of people with a loyalty card significantly differs from the loyalty of people without a loyalty card. The most appropriate technique to test this is an independent samples t-test.

a. T

b. F

29. A researcher estimates his model by means of regression analysis. The significance level of his model is .000 and the R-squared is .115. Based on these data, the researcher can conclude that the model is not significant.

a. T

b. F

30. A researcher estimates his model by means of regression analysis. The significance level of his model is .000 and the R-squared is .115. Based on these data, the researcher can conclude that at least one of the independent variables or the constant of the model is significantly related to Y.

a. T

b. F

31. Multiple regression analysis measures the effect of an independent variable on the dependent variable, while the other independent variables are held constant.

a. T

b. F

32. A researcher wants to know if the distribution of gender in his sample corresponds to the real distribution in his population (60% is female, 40% is male). The most appropriate technique to test this is:

a. A Chi-square test.

b. An independent samples t-test.

c. ANOVA

d. A paired-samples t-test.

33. Partial mediation exists when an independent variable influences the dependent variable both directly and indirectly.

a. T

b. F

34. Multicollinearity occurs when any single independent variable is highly correlated with the dependent variable.

a. T

b. F

35. The statistical null hypothesis is rejected when the p-value is big enough.

a. T

b. F

36. To be able to conduct a t-test it is important that the dependent variable is measured on an interval- or ratio scale and the independent variable is measured on a nominal- or ordinal scale.

a. T

b. F

37. An independent samples t-test is used to examine the differences in the same group before and after a treatment.

a. T

b. F

38. The Wilcoxon signed-rank test is a non-parametric test for examining significant differences between two related samples on a single sample.

a. T

b. F

39. McNemar's test is a parametric method used on nominal data.

a. T

b. F

40. MANOVA is similar to ANOVA, with the difference that MANOVA is able to handle more than one independent variable.

a. T

b. F