Chapter 11

Mind on Statistics

Chapter 11

Sections 11.1 - 11.2

Questions 1 to 4: For each study, decide if the two samples are independent samples or paired samples.

1. A group of 50 students each measured the length of their right arm and the length of their left arm. The average
right arm lengths were compared to the average left arm lengths.
A. Independent samples
B. Paired samples
KEY: B

2. A study compared the average number of courses taken by a random sample of 100 freshmen at a university
with the average number of courses taken by a separate random sample of 50 freshmen at a community college.
A. Independent samples
B. Paired samples
KEY: A

3. A group of 100 students were randomly assigned to receive vitamin C (50 students) or a placebo (50 students).
The groups were followed for 2 weeks and the proportions of students with colds were compared.
A. Independent samples
B. Paired samples
KEY: A

4. A group of 50 students had their blood pressures measured before and after watching a movie containing
violence. The mean blood pressure before the movie was compared with the mean pressure after the movie.
A. Independent samples
B. Paired samples
KEY: B

5. Which of the following helps determine the standard error for a confidence interval for a mean?
A. The sample size(s).
B. The sample estimate.
C. The true value of the population parameter.
D. The confidence level.
KEY: A

6. The distinction between a sampling distribution and a confidence interval is:

A. A confidence interval gives possible values for a sample statistic when the population parameter is assumed
known, while a sampling distribution gives possible values for a population parameter when only a single
value of a sample statistic is known.
B. A sampling distribution gives possible values for a sample statistic when the population parameter is
assumed known, while a confidence interval gives possible values for a population parameter when only a
single value of a sample statistic is known.
C. Sampling distributions exist only for situations involving means, while confidence intervals can be
computed for situations involving means and proportions.
D. Confidence intervals exist only for situations involving means, while sampling distributions can be
computed for situations involving means and proportions.
KEY: B

219
Chapter 11

7. A randomly selected sample of 100 students had an average grade point average (GPA) of 3.2 with a standard
deviation of 0.2. The standard error of the sample mean is
A. 0.020
B. 0.200
C. 1.600
D. 2.000
KEY: B

8. A randomly selected sample of 30 students spent an average amount of $40.00 on a date, with a standard
deviation of $5.00. The standard error of the sample mean is
A. 0.063
B. 0.167
C. 0.913
D. 5.000
KEY: C

9. A random sample of 250 third graders scored an average of 3.2 on a standardized reading test. The standard
deviation was 0.95. What is the standard error of the sample mean?
A. 0.95
B. 0.202
C. 0.0038
D. 0.060
KEY: D

10. A randomly selected sample of 60 mathematics majors spent an average of $200.00 for textbooks one term,
while during the same term, a randomly selected sample of 40 literature majors spent an average of $180.00 for
textbooks. The standard deviation for each sample was $20.00. The standard error for the difference between
the two sample means is
A. 0.057
B. 4.082
C. 5.744
D. 16.663
KEY: B

11. A random sample of 40 men drank an average of 20 cups of coffee per week during finals, while a sample of 30
women drank an average of 15 cups of coffee per week. The sample standard deviations were 6 cups for the
men and 3 cups for the women. The standard error for the difference between the two sample means is
A. 1.095
B. 1.200
C. 1.549
D. 2.400
KEY: A

12. Conscientiousness is a tendency to show self-discipline, act dutifully, and aim for achievement. The trait shows
a preference for planned rather than spontaneous behavior. A random sample of 650 students is asked to fill out
the Hogan Personality Inventory (HPI) to measure their level of conscientiousness. The 300 undergraduate
students scored an average of 145 with a standard deviation of 16. The 350 graduate students had a mean score
of 153 with a standard deviation of 21. What is the standard error for the difference between the two sample
means?
A. 18.5
B. 2.11
C. 1.45
D. 2.05
KEY: C

220
 Chapter 11

13. Which of the following is not true about the standard error of a statistic?
A. The standard error measures, roughly, the average difference between the statistic and the population
parameter.
B. The standard error is the estimated standard deviation of the sampling distribution for the statistic.
C. The standard error can never be a negative number.
D. The standard error increases as the sample size(s) increases.
KEY: D

14. For a randomly selected sample of n = 36 men’s heights, it is reported that the standard error of the mean is 0.5
inches. Three of the following statements are true, while one is false. Which statement is false?
A. The standard error (0.5 inches) is an estimated value of the standard deviation of the sample mean.
B. If a new sample of n = 36 men’s heights is collected, the standard error of the mean might not equal 0.5
inches.
C. Over many different samples of n = 36 men’s heights, the average difference between the sample mean and
population mean will be roughly 0.5 inches.
D. In about 95% of all samples of n = 36 men’s heights, the sample mean will be within 0.5 inches of the
population mean.
KEY: D

15. What is the primary purpose of a 95% confidence interval for a mean?
A. to estimate a sample mean
B. to test a hypothesis about a sample mean
C. to estimate a population mean
D. to provide an interval that covers 95% of the individual values in the population
KEY: C

16. The confidence level for a confidence interval for a mean is

A. the probability the procedure provides an interval that covers the sample mean.
B. the probability of making a Type 1 error if the interval is used to test a null hypothesis about the population
mean.
C. the probability that individuals in the population have values that fall into the interval.
D. the probability the procedure provides an interval that covers the population mean.
KEY: D

17. Which of the following will not result in paired data?

A. The same measurement is taken twice on each person, under different conditions or at different times.
B. Similar individuals are paired before giving the treatments in an experiment. Each member of a pair then
receives a different treatment. The same response variable is measured for all individuals.
C. Two different variables are measured for each individual. There is interest in the amount of difference
between the two variables.
D. One random sample is taken, and a variable is recorded for each individual, but then units are categorized
as belonging to one population or another.
KEY: D

18. Which of the following statements is most correct about a confidence interval for a mean?
A. It provides a range of values, any of which is a good guess at the possible value of the sample mean.
B. It provides a range of values, any of which is a good guess at the possible value of the population mean.
C. It provides a good guess for the range of values the sample mean is likely to have in repeated samples.
D. It provides a good guess for the range of values the population mean is likely to have in repeated samples.
KEY: B

221
Chapter 11

19. The weights of a sample of n = 8 college men will be used to create a 95% confidence interval for the mean
weight of all college men. What is the correct t* multiplier involved in calculating the interval?
A. 1.89
B. 2.00
C. 2.31
D. 2.36
KEY: D

20. The heights of a sample of n = 18 female college students will be used to create a 98% confidence interval for
the mean height of all female college students. What is the correct t* multiplier for this interval?
A. 2.11
B. 2.55
C. 2.57
D. 2.90
KEY: C

21. The WISC scores (similar to IQ test scores) of a sample of n = 20 5th graders will be used to create a 99%
confidence interval for the mean WISC score of all 5th graders. What is the correct t* multiplier for this
interval?
A. 2.86
B. 3.55
C. 2.54
D. 2.85
KEY: A

22. The heights of a random sample of 100 women are recorded. The sample mean is 65.3 inches and the sample
standard deviation is 3 inches. Which of the following is an approximate 95% confidence interval for the
population mean?
A. 65.3 (2)(0.03)
B. 65.3  (2)(0.3)
C. 65.3  (2)(3)
D. 65.3  (2)(30)
KEY: B

23. A random sample of 100 students had a mean grade point average (GPA) of 3.2 with a standard deviation of
0.2. The standard error of the sample mean in this case is 0.02. Calculate an approximate 95% confidence
interval for the mean GPA for all students.
A. (2.8, 3.6)
B. (3.16, 3.24)
C. (3.18, 3.22)
D. None of the above
KEY: B

24. A random sample of 30 students spent an average amount of $40.00 on a date, with a standard deviation of
$5.00. The standard error of the sample mean is 0.913. Calculate an approximate 95% confidence interval for
the average amount spent by all students on a date.
A. (40.00, 90.00)
B. (79.10, 80.90)
C. (78.20, 81.80)
D. None of the above
KEY: C

222
 Chapter 11

25. The cholesterol levels of a random sample of 100 men are measured. The sample mean is 188 and the sample
standard deviation is 40. Which of the following provides a 95% confidence interval for the population mean?
A. 188  (1.98)(0.4)
B. 188  (1.98)(4)
C. 188  (1.98)(40)
D. 188  (1.98)(4000)
KEY: B

26. A random sample of 30 airline flights during a storm had a mean delay of 40 minutes. The standard deviation
was 5 minutes, and the standard error of the mean is 0.9129. Calculate a 90% confidence interval for the
average delay for all flights during a storm.
A. (38.2, 41.8)
B. (38.4, 41.6)
C. (31.5, 48.5)
D. None of the above
KEY: B

27. A random sample of 30 airline flights during a storm had an average delay of 40 minutes. The standard
deviation was 5 minutes. Calculate a 98% confidence interval for the average delay for all flights during a
storm.
A. (37.8, 42.2)
B. (38.2, 41.8)
C. (27.7, 52.3)
D. None of the above
KEY: A

28. A randomly selected sample of n = 51 men in Brazil had an average lifespan of 59 years. The standard deviation
was 10 years, and the standard error of the mean is 1.400. Calculate a 90% confidence interval for the average
lifespan for all men in Brazil.
A. (42.2, 75.8)
B. (56.6, 61.4)
C. (56.2, 61.8)
D. None of the above
KEY: B

29. A randomly selected sample of n =51 men in Brazil had an average lifespan of 59 years. The standard deviation
was 10 years. Calculate a 98% confidence interval for the average lifespan for all men in Brazil.
A. (55.6, 62.4)
B. (56.2, 61.8)
C. (35.0, 83.0)
D. None of the above
KEY: A

30. A random sample of 25 third graders scored an average of 3.2 on a standardized reading test. The standard
deviation was 0.95. Calculate a 99% confidence interval for the mean score on this test.
A. (3.01, 3.39)
B. (2.49, 3.91)
C. (2.81, 3.59)
D. None of the above
KEY: D

223
Chapter 11

Questions 31 to 33: For a random sample of n = 203 adults in a survey, here is a summary of responses to “How
long did you sleep last night?”

Variable N Mean StDev SE Mean

sleep 203 6.42 1.56 0.11

31. Based on the output, what value estimates, roughly, the average difference that would be found between the
sample mean x and the population mean μ in repeated samples of this size?
A. 1
203
B. 6.42
C. 1.56
D. 0.11
KEY: D

32. Which of the following provides an approximate 95% confidence interval for the mean hours of sleep the
previous night in the population represented by the sample?
A. 6.42  20.11
B. 6.42  2 0.11
203
C. 6.42  0.11
D. 6.42  21.56
KEY: A

33. What confidence level would be associated with the interval 6.42  0.11 as a confidence interval for the
population mean ?
A. 50%
B. 68%
C. 90%
D. 95%
KEY: B

224
 Chapter 11

Questions 34 to 36: Spatial perception is measured on a scale from 0 to 10. Two groups of 8th grade children are
tested for spatial perception. The students in group 1 were first given a short course on spatial concepts. The students
in group 2 received no such instruction. SPSS was used to calculate descriptive statistics for the two samples.

34. What is the value of the standard deviation for measurements made on the group 1 students, which is missing in
the output?
A. 1.732
B. 3.807
C. 1.269
D. 0.141
KEY: C

35. Calculate a 95% confidence interval for the mean spatial perception score for students with instruction.
A. (6.47, 7.31)
B. (5.91, 7.87)
C. (6.04, 7.74)
D. None of the above
KEY: B

36. Calculate a 90% confidence interval for the mean spatial perception score for students without instruction.
A. (4.75, 5.91)
B. (4.00, 6.66)
C. (4.26, 6.40)
D. (4.18, 6.48)
KEY: C

225
Chapter 11

37. Explain what the difference between a parameter and a statistic is. Give an example of each.
KEY: A parameter is a characteristic of a population; a statistic is a characteristic of a sample. An example of a
parameter would be the proportion of all students who will vote for a particular candidate for student body
president. An example of a statistic is the proportion of students in a sample who will vote for the candidate.

38. Explain what is wrong with the following statement: A t-interval is a confidence interval estimate of a
sample mean.
KEY: A confidence interval is computed based on a sample statistic and the result is used to estimate a population
parameter. The statement would be correct if the phrase “sample mean” were replaced by the phrase
“population mean”.

39. In a random sample of 100 students (60 women and 40 men), the average hours of sleep during finals week
were compared. Are the samples of men and women independent samples or paired samples?
KEY: They are independent samples.

40. In a random sample of 100 students, the change in hours of sleep for each student during and after finals week
were compared. Are the samples of students’ sleep during finals and after finals independent samples or
paired samples?
KEY: They are paired samples.

Questions 41 to 43: Based on a random sample of 41 flights, the average minutes of airborne time for an airline was
106 minutes, with a standard deviation of 5 minutes.

41. What is the standard error for the average airborne time?
KEY: 0.7809 minutes

42. Give a 90% confidence interval for the average airborne time for this airline.
KEY: (104.7, 107.3)

43. Would a 95% confidence interval for the average airborne time for this airline be wider or narrower than the
90% confidence interval?
KEY: It would be wider.

Questions 44 to 46: Refer to the following statistical summary of the left foot lengths (in centimeters) for a random
sample of 76 college men.

Variable N Mean Median TrMean StDev SE Mean

LeftFoot 76 27.4 27.5 27.300 2.64 0.3

44. In terms of the notation of this course, what is the correct way to express the “Mean” shown in the output?
KEY: x = 27.4

45. Complete the following sentence: In about 95% of all samples of this size from this population, the
difference between the sample mean and the population mean would be ____.
KEY: no more than 0.6

46. Give an approximate 95% confidence interval for the average left foot length for college men.
KEY: (26.8, 28)

226
 Chapter 11

Questions 47 to 49: Spatial perception is measured on a scale from 0 to 10. A group of 9 th grade students are tested
for spatial perception. SPSS was used to calculate descriptive statistics of the spatial perception scores in the sample.

47. In terms of the notation of this course, what is the correct way to express the value 0.428 in the output?
KEY: s.e.( x )

48. Explain in your own words what the value 0.428 in the output measures.
KEY: The average distance between the possible values of the sample mean (obtained by taking all samples of this
size from this population) and the population mean is roughly 0.428.

49. Give a 90% confidence interval for the average spatial perception score for all 9th graders.
KEY: (5.26, 6.74)

227
Chapter 11

Section 11.3

50. Which of the following studies describes a paired data design?

A. The testosterone levels of male doctors and male college professors are compared
B. 40 students measure their blood pressure twice--first while resting and then again after running in place for
10 minutes.
C. The mean blood pressure of men is compared to the mean blood pressure of women
D. All of choices A, B, and C involve paired data.
KEY: B

51. Which of the following examples involves paired data?

A. A psychologist compares two methods of memorizing information. Twenty people use one method and
twenty other people use the second method.
B. A researcher estimates the difference between the mean forearm lengths of men and women based on a
random sample of each.
C. A medical researcher measures the cholesterol levels of each of 30 heart attack patients 2 days after the attack
and again 4 days after the attack.
D. A marketing research expert estimates the proportion of the 22 to 29 year-old age group that says they might
buy a new car in the next two years
KEY: C

52. Which one of the following ways of collecting data would not result in paired data?
A. Each person is measured twice.
B. Similar individuals are paired prior to an experiment. Each individual in a pair receives a different
treatment.
C. Two different variables are measured for each person.
D. Two independent samples are selected and the same response variable is compared between samples.
KEY: D

53. The multiplier for a confidence interval for one mean or the mean of paired differences depends on
A. The confidence level only.
B. The degrees of freedom only.
C. The confidence level and the degrees of freedom.
D. The confidence level and the sample standard deviation.
KEY: C

54. The amount of time the husband and the wife spend on house work is measured for 15 women and their 15
husbands. What are the appropriate degrees of freedom for the multiplier t* for a confidence interval for the
difference in mean time spent on housework between husbands and wives?
A. 14
B. 15
C. 29
D. 30
KEY: A

55. The head circumference is measured for 25 girls and their twin sisters. What is the value of the multiplier t* for
a 90% confidence interval for the difference in mean head circumference between girl twins?
A. 1.68
B. 2.02
C. 1.71
D. 2.06
KEY: C

228
 Chapter 11

Questions 56 to 58: An experiment is conducted with 15 seniors who are taking Spanish at Oak View High School.
A randomly selected group of eight students is first tested with a written test and a day later with an oral exam. To
avoid order effects, the other seven students are tested in reverse order. The instructor is interested in the difference
in grades between the two testing methods. SPSS is used to obtain descriptive statistics for the grades of the two
tests.

N Mean Std. Deviation

Talk 15 1.523 1.530
Write 15 4.166 2.047
Talk - Write 15 -2.643 2.182

56. What are the degrees of freedom for this design?

A. 7
B. 8
C. 14
D. 29
KEY: C

57. What is the value of the standard error for the difference in mean grades between the two testing methods?
A. 0.395
B. 0.529
C. 0.563
D. 0.660
KEY: C

58. Calculate a 98% confidence interval for the difference in mean grades between the two testing methods.
A. (−3.85, −1.44)
B. (−4.03, −1.26)
C. (−4.12, −1.17)
D. (−4.33, −0.95)
KEY: C

Questions 59 and 60: “Sunlight makes everyone feel better.” To test this claim, 30 college students receive special
lights for their bedrooms that slowly turn on in the morning with imitation sunlight. The experiment takes place in
January, when it is still dark out in the morning. Prior to participation, the students fill out a questionnaire to
measure their general mood. The average score was 2.73. After three weeks with the special light they fill in the
questionnaire again. The average score this time was 3.13. The standard deviation of the differences was 0.822.

59. What is the value of the standard error for the difference in average mood score?
A. 0.106
B. 0.150
C. 0.153
D. 0.822
KEY: B

60. Calculate a 95% confidence interval for the difference in average mood scores (after − before).
A. (0.14, 0.66)
B. (0.09, 0.71)
C. (0.19, 0.61)
D. None of the above
KEY: B

229
Chapter 11

Questions 61 to 63: A manufacturer of cell phones is interested in evaluating two alternative existing layout and
flow path designs. Design 1 is the current one in use and Design 2 is a new proposed plan. Both designs were
implemented for eight consecutive working days and the output of each design was recorded.

61. What are the degrees of freedom for this paired design analysis?
A. 8 (since there are eight days).
B. 16 (since there are 16 observations in all).
C. 15
D. 7
KEY: D

62. Suppose the sample mean difference in output (Design 2 – Design 1) was –174 phones and the standard error of
the mean difference was 28. What is the 95% confidence interval for the population mean difference in output?
A. –174  2.3628
B. –174  2.36 28
8
C. –174  28
D. –174  0.9528
KEY: A

63. How would a 90% confidence interval for the mean difference in output compare to the 95% confidence
interval?
A. It would be wider.
B. It would be narrower.
C. It would stay the same.
D. You cannot tell until you look up the multiplier value.
KEY: B

Questions 64 to 67: Managerial training programs often teach supervisory skills. Are these training programs
effective? A company psychologist administers a test to each of 10 supervisors before and after attending such a
training program. Higher scores indicate increased supervisory skills. The results are used to estimate the mean
improvement in supervisory skills.

64. Explain why this is a paired data design.

KEY: Two measurements are made on each of the 10 supervisors.

65. Suppose the mean improvement in supervisory skills was 3 point and the standard deviation of the sampled
differences was 2 points. Give the standard error of the sample mean difference.
KEY: 0.632

66. Give a 95% confidence interval for the population mean improvement in supervisory skills for all supervisors
who take the training program.
KEY: (1.57, 4.43)

67. Does it appear that the training program is effective? Explain how you came to your decision.
KEY: Yes, the training program does appear to be effective, since the 95% confidence interval for the mean
improvement did not include the value of zero and in fact contains only positive values.

230
 Chapter 11

Section 11.4

68. When constructing a confidence interval for the difference in two population means, it is appropriate to use the
pooled standard error only when
A. the population standard deviations can be assumed to be equal.
B. the sample standard deviations are exactly equal.
C. the population means can be assumed to be equal.
D. the sample means are exactly equal.
KEY: A

69. Which one of the following ways of collecting data would result in two independent samples?
A. A sample of college students is measured for stress both before and after taking an important exam.
B. A group of college students are paired based on gender, age, major, and grades. One student in a pair
receives stress management training and the other one does not. After a few weeks all students are
measured for stress right before an important exam.
C. A sample of English majors is measured for stress before taking an important exam. A sample of
engineering majors is also measured for stress before taking their important exam.
D. A sample of college students is measured for stress before taking an important exam. Their exam grade is
also recorded.
KEY: C

70. The amount of time single men and women spend on house work is measured for 15 single women and 20
single men. What are the appropriate degrees of freedom for the multiplier t* for a pooled confidence interval
for the difference in mean time spent on housework between single men and women?
A. 14
B. 19
C. 33
D. 35
KEY: C

71. The head circumference is measured for a sample of 15 girls and a separate sample of 15 boys. What is the
correct combination of degrees of freedom and the value of the multiplier t* for a 90% confidence interval for
the difference in mean head circumference between girls and boys?
A. df = 13, t* = 1.77
B. df = 28, t* = 1.70
C. df = 28, t* = 2.05
D. df = 30, t* = 1.70
KEY: B

72. The amount of time single men and women spend on house work is measured for 15 single women and 25
single men. For the women the mean was 7 hours/week with a standard deviation of 1.5. For the men the mean
was 4.5 hours/week with a standard deviation of 1.1. What is the value of the pooled standard deviation for the
difference in mean time spent on housework between single men and women?
A. 1.30
B. 0.45
C. 1.59
D. 1.26
KEY: D

231
Chapter 11

73. The head circumference is measured for a sample of 17 girls and a separate sample of 12 boys. The mean for
the girls was 49 cm with a standard deviation of 1.25 cm. The mean for the boys was 50 cm with a standard
deviation of 0.95 cm. What is the value of the pooled standard error for the difference in mean head
circumference between girls and boys?
A. 0.41
B. 1.29
C. 0.43
D. 1.14
KEY: C

74. A random sample of 60 mathematics majors spent an average of $200.00 for textbooks for a term, with a
standard deviation of $22.50. A random sample of 40 English majors spent an average of $180.00 for textbooks
that term, with a standard deviation of $18.30. What is the value of the pooled standard error for the difference
in mean amount spent?
A. 20.93
B. 4.27
C. 16.81
D. 4.10
KEY: B

75. Random samples of 200 men and 200 women were collected and their resting pulse rates were measured, to
estimate how much mean resting pulse rates differ for men and women in the population. An analyst mistakenly
paired the observations and constructed an approximate 95% confidence interval for the mean difference to be
(5  20.2) beats per minute. If the data had been analyzed correctly, finding an approximate 95% confidence
interval for the difference in population means, which of the following parts of the interval would be different?
A. The sample statistic of 5 beats per minute.
B. The multiplier of 2.
C. The standard error of 0.2.
D. None of the parts would be different; it is an equivalent analysis.
KEY: C

76. A researcher asked random samples of 50 kindergarten teachers and 50 12 th grade teachers how much money
they spent out-of-pocket on school supplies in the previous school year, to see if teachers at one grade level
spend more than the other. A 95% confidence interval for K  12 is $30 to $50. Based on this result, it is
reasonable to conclude that
A. 95% of all kindergarten and 12th grade teachers spend between $30 and $50 on average.
B. 95% of all kindergarten teachers spend between $30 and $50 more then 95% of all 12th grade teachers.
C. kindergarten teachers spend more on average than do 12 th grade teachers.
D. 12th grade teachers spend more on average than do kindergarten teachers.
KEY: C

77. A 95% confidence interval for the difference between the mean handspans of men and the mean handspans of
women is determined to be 2.7 centimeters to 3.3 centimeters. Which of the following statements is the best
interpretation of this interval?
A. It is likely that the difference in the population mean handspans of men and women is covered by the
interval 2.7 centimeters to 3.3 centimeters.
B. It is likely that the difference in the sample mean handspans of men and women is covered by the interval
2.7 centimeters to 3.3 centimeters.
C. It is likely that if new samples of the same size were to be taken, the difference in sample means would be
contained in the interval 2.7 centimeters to 3.3 centimeters.
D. It is likely that for 95% of married couples, the husband's handspan is between 2.7 and 3.3 centimeters
longer than the wife's hand span.
KEY: A

232
 Chapter 11

78. A random sample of 60 mathematics majors spent an average of $200.00 for textbooks for a term, while a
random sample of 40 literature majors spent an average of $180.00 for textbooks that term. The standard
deviation for each sample was $20.00. The standard error for the difference between the two sample means is
4.082. Calculate an approximate 95% confidence interval for the difference in average amounts spent on
textbooks (math majors – literature majors).
A. (7.8, 32.2)
B. (11.8, 28.2)
C. (15.9, 24.1)
D. None of the above
KEY: B

79. A random sample of 40 men drank an average of 20 cups of coffee per week during finals, while a sample of 30
women drank an average of 15 cups of coffee per week. The sample standard deviations were 6 cups for the
men and 3 cups for the women. The standard error for the difference between the two sample means is 1.095.
Calculate an approximate 95% confidence interval for the difference in average cups of coffee drunk
(men –women).
A. (2.81, 7.19)
B. (3.02, 6.98)
C. (3.91, 6.10)
D. None of the above
KEY: A

80. A random sample of 60 mathematics majors spent an average of $200.00 for textbooks for a term, with a
standard deviation of $22.50. A random sample of 40 English majors spent an average of $180.00 for textbooks
that term, with a standard deviation of $18.30. Calculate a 90% pooled confidence interval for the difference in
average amounts spent on textbooks (math majors – English majors).
A. (12.91, 27.09)
B. (11.50, 28.50)
C. (13.19, 26.81)
D. None of the above
KEY: A

81. An experiment is conducted with two groups of 25 college students. The treatment group scored an average of
3.67 with a standard deviation of 0.86. The placebo group scored an average of 3.29 with a standard deviation
of 0.97. Calculate a 95% pooled confidence interval for the difference in means between the treatment and the
placebo.
A. (−0.44, 0.82)
B. (−0.14, 0.90)
C. (−0.25, 1.01)
D. None of the above
KEY: B

82. Reaction time is measured in a driving simulator for a random sample of 16 year-old boys (n = 12, x = 4.5,
s = 1.1) and a random sample of 24 year-old young men (n = 9, x = 2.9, s = 1.3). Calculate a 99% pooled
confidence interval for the difference in mean reaction time (16 year-olds – 24 year-olds).
A. (0.48, 2.72)
B. (0.51, 2.70)
C. (0.07, 3.13)
D. None of the above
KEY: D

233
Chapter 11

Questions 83 and 84: For a randomly selected sample of 20 new mothers in the year 2000, the mean age was 24.6
years. For a randomly selected sample of 10 new mothers in 1970, the mean age was 21.4 years. The difference
between the mean ages is 3.2 years, and the standard error of the difference is 1.366. Assume that the ages of new
mothers are normally distributed, do not assume the population variances are equal, and use the conservative “by
hand” estimate for the degrees of freedom.

83. Calculate a 90% confidence interval for the difference in population mean ages of new mothers in the two years
(year 2000 – year 1970).
A. (0.84, 5.56)
B. (0.78, 5.67)
C. (0.70, 5.70)
D. None of the above
KEY: C

84. Calculate a 99% confidence interval for the difference in population mean ages of new mothers in the two years
(year 2000 – year 1970).
A. (−1.24, 7.64)
B. (−1.13, 7.53)
C. (0.47, 5.93)
D. None of the above
KEY: A

Questions 85 to 87: A Statistics instructor asked a random sample of female and male students how many hours
they exercise each week. Minitab output for a comparison of females and males is shown below.

Gender N Mean StDev SE Mean

female 129 6.16 4.18 0.37
male 73 6.42 4.39 0.51

Difference = mu (female) - mu (male )

Estimate for difference: -0.262
95% CI for difference: (-1.511, 0.988)

85. This is an example of data collected using

A. paired data.
B. independent samples.
C. a binomial experiment.
D. two binomial experiments.
KEY: B

86. What is the correct notation for the parameter that is estimated by the confidence interval (95% CI) given in the
output?
A. p
B. p1  p2
C. 
D. 1  2
KEY: D

87. Based on the output given, what can be concluded about the difference between the mean amounts that females
and males exercise per week (female  male)?
A. It is not unreasonable to conclude that the difference in sample means is 0.
B. It is not unreasonable to conclude that the difference in population means is 0.
C. It is not unreasonable to conclude that the difference in sample means is positive.
D. All of A, B and C can be concluded.
KEY: B

234
 Chapter 11

88. Random samples of 5 Japanese women and 10 Japanese men showed an average life span of 83 years for the
women and 77 years for the men. The standard deviation was 2 years for the women and 1 year for the men.
Calculate a 95% confidence interval for the difference in average life spans (women  men). Assume that the
life spans are normally distributed, but do not assume the population variances are equal, and use the
conservative “by-hand” estimate for the degrees of freedom.
KEY: (3.36, 8.64)

89. Reaction time is measured in a driving simulator for a random sample of 16 year-old boys and a random
sample of 24 year-old young men. SPSS was used to obtain descriptive statistics.

N Mean Std. Deviation Variance

16 year 12 4.52 1.090 1.1886
24 year 9 2.86 1.347 1.8157

Calculate a 95% confidence interval for the difference in mean reaction time (16 year-olds – 24 year-olds).
Assume that t reaction time is normally distributed, but do not assume the population variances are equal, and
use the conservative “by-hand” estimate for the degrees of freedom.
KEY: (0.39, 2.93)

90. An experiment is conducted with two groups of college students. The 25 students in the treatment group
scored an average of 3.7 with a standard deviation of 0.8. The 18 students in the placebo group scored an
average of 3.3 with a standard deviation of 1.0. Compare the degrees of freedom from Welch’s approximation
with the conservative “by-hand” estimate for the degrees of freedom.
KEY: Conservative estimate: df = 17; Welch’s approximation: df = 31.5

Questions 91 to 94: A union representative has collected some data on the hourly pay of workers in the area. He is
curious to find out if hourly pay actually increases with age. He does not have the actual ages of the workers in his
sample, but he does have age group information. For now, he wishes to compare the young workers (18-30 years old)
with the middle aged workers (31-45 years old) by calculating a confidence interval for the difference in population
mean salaries.

91. The two sample standard deviations are very similar so the pooled procedure will be used. If the researcher
has collected data on 32 young workers and 40 middle aged workers, what would be the degrees of freedom
for the t* multiplier?
KEY: 70

92. What is the appropriate t* multiplier for a 90% confidence interval estimate?
KEY: 1.67

93. The 90% confidence interval for the difference in average hourly pay (young – middle aged) was found to be
(2.55, 1.43). Based on this interval, what conclusion can we draw about the average hourly pay for these
two age groups?
KEY: Since zero is included in the confidence interval, there does not appear to be a significant difference in the
average hourly pay for the two age groups.

94. How would the 99% confidence interval for the difference in average hourly pay compare to the 90%
confidence interval?
KEY: The 99% confidence interval would be wider, and thus still contain the value of zero.

235
Chapter 11

Section 11.5

95. A cola-tasting experiment is conducted in which 100 people are each given a half-cup of Cola A and a half-cup
of Cola B. All 100 participants taste both colas and then rate the taste of each cola on a 0 (horrible) to 50 (tastes
great) scale. For each person, a coin is tossed to determine the order of tasting the two colas. To estimate the
mean difference between ratings of the two products, what procedure should be used?
A. A confidence interval for one proportion.
B. A confidence interval for a difference between two proportions.
C. A paired t (or 1-sample t) confidence interval for a mean difference.
D. A 2-sample t confidence interval for a difference in two means.
KEY: C

96. For which one of the following situations was a procedure for finding a confidence interval not covered in the
textbook?
A. A mean difference for paired data.
B. A difference in means for two independent samples.
C. A difference in proportions for paired data.
D. A difference in proportions for two independent samples.
KEY: C

97. A research study compares the risk of lung disease for men who smoke to the risk of lung disease for women
who smoke. The sample relative risk is 1.1 and a 95% confidence interval for the relative risk is 0.5 to 1.9.
Which of the following statements is most correct about the comparison of the two groups?
A. The population risks may be the same because the interval includes the value 1.
B. The population risks may be the same because the interval doesn't include the value 0.
C. The population risks might differ because the interval includes the value 1.
D. The population risks might differ because the interval doesn't include the value 0.
KEY: A

98. A study on the relationship between vertex baldness and heart attacks in men reported that “For any vertex
baldness [compared with no baldness], the age-adjusted relative risk [of heart attack] was 1.4
(95% CI, 1.2 to 1.9).” Based on this, it can reasonably (with 95% confidence) be concluded that
A. men with any baldness are between 1.2 and 1.9 times more likely to have a heart attack than men of the
same age with no baldness.
B. men with no baldness are between 1.2 and 1.9 times more likely to have a heart attack than men of the
same age with any baldness.
C. the percent of men with any baldness likely to have a heart attack is between 1.2% and 1.9%.
D. the risk of having a heart attack for men with no baldness is not significantly different from the risk of
having a heart attack for men with any baldness.
KEY: A

99. A 95% confidence interval for the difference in the population proportions of men (group 1) and women
(group 2) who exercise regularly, is [0.06 to +0.10]. Which of the following is not a reasonable conclusion
from this confidence interval?
A. It is plausible that men spend 10% more time exercising, on average, than do women.
B. It is plausible that equal proportions of men and women exercise regularly.
C. It is plausible that the difference in proportions of men and women who exercise regularly is at least 0.05.
D. It is not plausible that the difference in proportions of men and women who exercise regularly is at least
0.15.
KEY: A

236
 Chapter 11

100. A prospective study on the relationship between sleep deprivation and heart disease was done by Ayas, et. al.
(Arch Intern Med 2003). Women who slept 6 hours a night were compared to women who slept for 8 hours a
night (reference group). After adjusting for potential confounding variables like smoking, a 95% confidence
interval for the relative risk of heart disease was (0.98, 1.42). Based on this confidence interval, state what
the association between sleep deprivation and heart disease is.
KEY: There is no conclusive evidence, based on this study, that the association (as measured by relative risk) is
different from 1 (equal risk).

Questions 101 to 104: As the number of men with lung cancer declines, the American Cancer Society estimates that
73,020 women will die in the United States of lung cancer this year, more than those who will die from breast,
ovarian and uterine cancers combined. While no national studies have yet been done, many lung cancer specialists
say they are seeing a disturbing trend of more and more non-smoking women with the disease. Based on one study,
it is estimated that 10 to 15 percent of lung cancer victims are non-smokers. Suppose the range from 10 to 15
percent actually represents a 95% confidence interval.

101. What was the proportion of lung cancer victims in the sample that were non-smokers?
KEY: The midpoint of the confidence interval provides the sample statistic, here it is 12.5%.

102. What was the margin of error associated with this estimate of the proportion of lung cancer victims that are
non-smokers?
KEY: 0.025 or 2.5%

103. Is the following statement true or false? There is a 95% probability that the population proportion of lung
cancer victims that are non-smokers is between 0.10 and 0.15.
KEY: False.

104. Is the following statement about the 95% confidence level true or false? If repeated samples of the same sizes
were obtained, we would expect 95% of the resulting intervals to contain the population proportion of lung
cancer victims that are non-smokers.
KEY: True.

237
Chapter 11

238