1. A normal distribution is a basis for classical statistical inference that is bell-shaped and symmetrical in appearance.

True (moderate, page 355)

2. The measures of central tendency for a normal distribution are all identical.

True (moderate, page 355)

3. A summary description of a fixed characteristic or measure of the target population is called a standard deviation.

False (moderate, page 355)

4. According to the text, a parameter denotes the true value that would be obtained if a census, rather than a sample, were undertaken.

True (moderate, page 355)

5. A statistic is a summary description of a characteristic or measure of the sample.

True (moderate, page 355)

6. The sample statistic is used as an estimate of the population parameter.

True (moderate, page 355)

7. When estimating a population parameter by using a sample statistic, the precision level is the desired size of the estimating interval.

True (moderate, page 355)

8. Standard deviation is the maximum permissible difference between the sample statistic and the population parameter.

False (moderate, page 355)

9. The range into which the true population parameter will fall, assuming a given level of confidence, is called the sampling distribution.

False (moderate, page 356)

10. According to the text, the confidence level is the probability that a confidence interval will include the population parameter.

True (moderate, page 356)

11. The error that results when the particular sample selected is an imperfect representation of the population of interest is called nonrandom sampling error.

False (moderate, page 356)

12. The distribution of the values of a sample statistic computed for each possible sample that could be drawn from the target population under a specified sampling plan is called the sampling distribution.

True (moderate, page 356)

13. Statistical inference is the process of generalizing the sample results to the population results.

True (moderate, page 356)

14. The standard deviation of the sampling distribution of the mean or proportion is called random error.

False (moderate, page 357)

15. According to the text, the sampling distribution of the means is a normal distribution.

True (moderate, page 357)

16. The number of standard errors that a point is away from the mean is called the z value.

True (moderate, page 357)

17. The area under the sampling distribution curve between any two points can be calculated in terms of p values.

False (moderate, page 356)

18. Confidence intervals around sample means or proportions are estimated using the variance formula.

False (moderate, page 358)

19. The American Customer Satisfaction Index is a national measure of satisfaction with quality that provides a resource for examining the differences in satisfaction among customers of different industries by their demographic and socioeconomic characteristics.

True (moderate, page 360)

20. Sample size can be computed by using the formula for the standard error.

True (moderate, page 360)

21. Sample size is directly affected by the variability of the characteristic in the population.

True (easy, page 362)

22. Proportions are estimated when the choice is dichotomous.

True (moderate, page 362)

23. When computing confidence intervals, maximum variation occurs when a proportion of interest is 0.5.

True (difficult, page 360)

24. The two factors that affect the task of interviewing a much larger number of respondents than originally anticipated to achieve the desired net sample are random rate and standard rate.

False (difficult, page 365)

25. When studying a characteristic with a high incident rate the researcher will have to initially contact many more people than if the characteristic has a low incident rate.

False (difficult, page 365)

26. The rate of occurrence of persons eligible to participate in the study expressed as a percentage is called the completion rate.

False (moderate, page 365)

27. The percentage of qualified respondents who complete the interview is called the incidence rate.

False (moderate, page 365)

28. The completion rate enables researchers to take into account anticipated refusals by people who qualify for a research study.

True (moderate, page 366)

29. Response rate = number of completed interviews / number in the population.

False (moderate, page 366)

30. As the number of qualifying respondents for a research study increases, the incidence rate will fall.

False (moderate, page 365)

31. According to the text, the statistical approaches to determining sample size based on confidence intervals may involve the estimation of the mean or the proportion.

True (moderate, page 358)

32. Statistical calculation of sample size requires determining the z value.

True (moderate, page 362)

33. When conducting marketing research in foreign countries, statistical estimation of sample size may be difficult since estimates of the population variance may be unavailable.

True (difficult, page 368)

34. According to the text, one of the biggest challenges in online research is the ability to generate a truly random sample that is representative of the target population.

True (moderate, page 368)

35. The statistical approaches to determining sample size are based on confidence intervals.

True (moderate, page 368)

Multiple Choice

36. A ______________ is a basis for classical statistical inference that is bell-shaped and symmetrical in appearance.

a. normal distribution (moderate, page 355)

b. standard deviation

c. variance

d. confidence interval

e. classic symmetry

37. The measures of central tendency for a normal distribution are all ______________.

a. inverted

b. random

c. identical (moderate, page 355)

d. uncertain

e. binary

38. Which of the following measures of central tendency for a normal distribution is not identical?

a. mean

b. median

c. mode

d. All are identical for a normal distribution. (difficult, page 355)

e. standard deviation

39. A summary description of a fixed characteristic or measure of the target population is called a ______________.

a. parameter (moderate, page 355)

b. standard deviation

c. statistical inference

d. normal distribution

e. variance

40. According to the text, a ______________ denotes the true value that would be obtained if a census, rather than a sample, were undertaken.

a. standard deviation

b. statistical inference

c. normal distribution

d. parameter (moderate, page 355)

e. standard normal

41. A ______________ is a summary description of a characteristic or measure of the sample.

a. statistic (moderate, page 355)

b. confidence interval

c. sampling distribution

d. confidence level

e. summary judgment

42. The ______________ is used as an estimate of the population parameter.

a. random sampling error

b. sample statistic (moderate, page 355)

c. confidence level

d. precision level

e. standard deviate

43. When estimating a population parameter by using a sample statistic, the ______________ is the desired size of the estimating interval.

a. confidence interval

b. precision level (moderate, page 355)

c. random sampling error

d. standard deviation

e. mode

44. Which of the following is the maximum permissible difference between the sample statistic and the population parameter?

a. confidence interval

b. random sampling error

c. precision level (moderate, page 355)

d. standard deviation

e. range

45. The range into which the true population parameter will fall, assuming a given level of confidence, is called the _____________.

a. confidence interval (moderate, page 356)

b. precision level

c. sampling distribution

d. standard deviation

e. standard range

46. According to the text, the ______________ is the probability that a confidence interval will include the population parameter.

a. precision level

b. standard deviation

c. variance

d. confidence level (moderate, page 356)

e. under/over

47. The error that results when the particular sample selected is an imperfect representation of the population of interest is called _______________.

a. random sampling error (moderate, page 356)

b. symmetrical error

c. standard error

d. nonrandom sampling error

e. under/over error

48. The distribution of the values of a sample statistic computed for each possible sample that could be drawn from the target population under a specified sampling plan is called the _______________.

a. confidence level

b. sampling distribution (moderate, page 356)

c. normal distribution

d. standard deviation

e. standard values distribution

49. ______________ is the process of generalizing the sample results to the population results.

a. Randomizing

b. Hypothesizing

c. Statistical inference (moderate, page 356)

d. Standardizing

e. parsing

50. The standard deviation of the sampling distribution of the mean or proportion is called ______________.

a. unsymmetrical error

b. standard error (moderate, page 357)

c. variance

d. random error

e. median

51. In cases where the population standard deviation is unknown, it is estimated by the sample standard deviation. In this case, the standard error of the mean becomes:

a. mean / proportion

b. standard variate / standard deviate

c. the z-value

d. p (1 – p)

e. sample standard deviation / square root of sample size (difficult, page 357)

52. According to the text, as the sample size increases the sampling distribution of the mean becomes a(n) _______________.

a. unsymmetrical distribution

b. normal distribution (moderate, page 356)

c. inverted distribution

d. random distribution

e. chi-square distribution

53. The symbol for the sample standard deviation is _______________.

a. p

b. n^{2}

c. n

d. s^{2}

e. s (difficult, page 356)

54. The number of standard errors that a point is away from the mean is called the ______________.

a. F value

b. P value

c. z value (moderate, page 357)

d. t value

e. chi-square value

55. The area under the sampling distribution curve between any two points can be calculated in terms of ______________.

a. z values (moderate, page 357)

b. F values

c. p values

d. t values

e. chi-square values

56. Which of the following is NOT a qualitative factor that should be considered when determining the sample size?

a. the importance of the decision

b. the nature of the research

c. incidence rates

d. All of the selections are qualitative factors that should be considered when determining the sample size. (moderate, page 358)

e. the number of variables

57. Confidence intervals around sample means or proportions are estimated using the ______________.

a. standard error formula (moderate, page 358)

b. variance formula

c. standard deviation formula

d. confidence interval formula

e. reflexive value equation

58. The _______________ is a national measure of satisfaction with quality that provides a resource for examining the differences in satisfaction among customers of different industries by their demographic and socioeconomic characteristics.

a. Arbitron Rating Index

b. American Customer Satisfaction Index (moderate, page 360)

c. A.C. Nielsen Diary Panel Index

d. American Regulated Index

e. J. D. Power Loyalty Index

59. Sample size is influenced by which of the following?

a. confidence level changes

b. population variability

c. precision level changes

d. a and b only

e. a, b, and c (difficult, page 361)

60. The two factors that affect the task of interviewing a much larger number of respondents than originally anticipated to achieve the desired net sample are ______________ and ______________.

a. standard rate; incidence rate

b. incidence rate; completion rate (difficult, page 365)

c. completion rate; random rate

d. random rate; standard rate

e. tolerance rate; error rate

61. When studying a characteristic with a ______________, the researcher will have to initially contact many more people than if the characteristic has a ______________.

a. low incidence rate; high incidence rate (difficult, page 365)

b. low random rate; high random rate

c. high incidence rate; low incidence rate

d. high random rate; low random rate

e. high diversity; low diversity

62. The rate of occurrence of persons eligible to participate in the study expressed as a percentage is called the ______________.

a. standard rate

b. random rate

c. completion rate

d. incidence rate (moderate, page 365)

e. diversity rate

63. If the incidence rate is 0.25, then on average how many potential respondents must be approached in order to obtain one qualified respondent?

a) .25

b) .75

c) 1.25

d) 3

e) 4 (moderate, page 365)

64. The percentage of qualified respondents who complete the interview is called the ______________.

a. standard rate

b. random rate

c. completion rate (moderate, page 366)

d. incidence rate

e. torsion rate

65. Which of the following enables researchers to take into account anticipated refusals by people who qualify for a research study?

a. completion rate (moderate, page 366)

b. standard rate

c. random rate

d. incidence rate

e. alarm rate

66. The reason sample size for online surveys generally exceeds that determined statistically is because _______________.

a. the population variance is larger online

b. the population variance is smaller online

c. federal regulations require larger samples online

d. statistics become less relevant online because statistical theory is weaker online

e. the marginal cost of collecting and analyzing larger sample is low (moderate, page 367)

67. As the number of qualifying respondents for a research study increases, the ______________ will fall.

a. completion rate

b. standard rate

c. random rate

d. incidence rate (moderate, page 365)

e. activation rate

68. According to the text, the statistical approaches to determining sample size based on confidence intervals that may involve the estimation of the _______________ or _______________.

a. standard deviation ; variance

b. variance;mean

c. mean;proportion (difficult, page 368)

d. proportion;standard deviation

e. mode;median

69. Statistical calculation of sample size requires specification of all of the following EXCEPT:

a. precision level.

b. confidence level.

c. population standard deviation.

d. z value (moderate, page 368)

e. c and d

70. When conducting marketing research in foreign countries, statistical estimation of sample size may be difficult since estimates of the ______________ may be unavailable.

a. population variance (difficult, page 368)

b. population demographics

c. population density

d. population error

e. census boundaries

71. All of the following are true EXCEPT:

a. Sample size is one of the major costs in a research project.

b. Sample size is directly proportional to the variance of the variable.

c. Estimates of the population variance are based on small pilot studies, related research and judgment of the researcher.

d. The researcher should disclose to the client unexpected differences between expected and actual sample standard deviations.

e. Inflating population variance estimates does not lead to inflating the sample size. (difficult, page 369)

72. For simple applications, appropriate sample size formulas can be programmed using ______________ programs.

a. word processing

b. spreadsheet (moderate, page 369)

c. database

d. publishing

e. GIS

73. Which of the following factors DOES NOT need to be specified when using a statistical package to determine the appropriate sample size for a research study?

a. precision level

b. confidence level

c. population standard deviation

d. population demographics (moderate, page 369)

e. c and d

74. According to the text, one of the biggest challenges in online research is the ability to _______________.

a. generate quality responses

b. generate a truly random sample (moderate, page 368)

c. generate sufficient responses

d. generate a convenience sample

e. generate analysis of variance

75. The statistical approaches to determining sample size are based on ______________.

a. confidence intervals (difficult, page 368)

b. random sample error

c. regression analysis

d. product characteristics

e. geometric principles

Essay Questions

76. In a short essay, list and discuss at least four statistical concepts used in sample size determination.

Answer

a. Parameter – a summary description of a fixed characteristic or measure of the target population. A parameter denotes the true value that would be obtained if a census, rather than a sample, were undertaken.

b. Statistic – a summary description of a characteristic or measure of the sample. The sample statistic is used as an estimate of the population parameter.

c. Precision level – when estimating a population parameter by using a sample statistic, the precision level is the desired size of the estimating interval. This is the maximum permissible difference between the sample statistic and the population parameter.

d. Confidence interval – the range into which the true population parameter will fall, assuming a given level of confidence.

e. Confidence level – the probability that a confidence interval will include the population parameter.

f. Random sampling error – the error that results when the particular sample selected is an imperfect representation of the population of interest.

(moderate, pages 355-356)

77. In a short essay, discuss the concept of the sampling distribution.

Answer

The sampling distribution is the distribution of the values of a sample statistic computed for each possible sample that could be drawn from the target population under a specified sampling plan. Under a specified sampling plan, it is possible to draw several different samples of a given size from the target population. The sampling distribution should not be confused with the distribution of the values of the elements in a sample.

(easy, page 356)

78. In a short essay, discuss the concept of statistical inference.

Answer

An important task in marketing research is to calculate statistics, such as the sample mean and sample proportion, and to use them to estimate the corresponding true population values. This process of generalizing the sample results to the population results is referred to as statistical inference. In practice, a single sample of predetermined size is selected, and the sample statistics are computed. The sample statistics become the basis for making inferences about the population values. This inferential process is possible because sampling distribution enables a researcher to use probability theory to make inferences about the population values.

(moderate, page 356)

79. In a short essay, define incidence rate and completion rate. Next, discuss the use of incidence and completion rates in determining sample size.

Answer

Incidence rate refers to the rate of occurrence. It is influenced by the proportion of people in the population. When studying a characteristic with a low incident rate, the researcher will have to initially contact many more people than if the characteristic has a high incidence rate, screening out those who do not meet the study requirements. The number of initial contacts must also be increased in anticipation of refusals coming from people qualified to participate. The completion rate denotes the percentage of qualified respondents who complete the interview. It enables researchers to take into account anticipated refusals by people who qualify.

(moderate, page 365)

80. In a short essay, list and discuss at least two specific sampling issues that are unethical in marketing research.

Answer

The sample size is one of the major determinants of cost in a marketing research project. It may seem that if the sample size is statistically determined, this procedure is free from ethical conflicts. However, this may not be true. The sample size is directly proportional to the variance of the variable and estimates of the population variance are based on small pilot studies, related research, and the judgment of the researcher. Because judgment is involved, the researcher has the ethical responsibility to not use large estimates of the population variance simply to increase the cost of the project by inflating the sample size. Furthermore, the researcher may be faced with ethical dilemmas when the sample standard deviation varies widely from that assumed. In this case, the confidence interval will be larger than required if the actual sample standard deviation turns out to be much larger than that used to estimate the sample size. If this happens, the researcher should disclose the larger confidence interval to the client and jointly arrive at a corrective action.

(moderate, pages 369-370)

81. In a short essay, discuss sampling issues facing researchers involved in international marketing research.

Answer

When conducting marketing research in foreign countries, statistical estimation of sample size may be difficult since estimates of the population variance may be unavailable. Hence, the sample size is often determined by qualitative considerations. When statistical estimation of sample size is attempted, the differences in estimates of population variance should be recognized and factored in if possible. For example, consumer preferences for certain products may be relatively heterogeneous in markets where those products have been newly introduced. Thus, it may be a mistake to assume that the population variance is the same, or to use the same sample size across countries.