Chapter - 08 Con Interval

CHAPTER 8: Confidence Interval Estimation
MULTIPLE CHOICE
1. The chi-square and F-distributions are used primarily to make inferences about population
___________.
a. means
b. variances
c. medians
d. modes
e. proportions
ANS: B PTS: 1 MSC: AACSB: Analytic | AACSB: Interval Estimation
2. If you increase the confidence level, the confidence interval _____.

a. decreases
b. increases
c. stays the same
d. may increase or decrease, depending on the sample data
3. If you are constructing a confidence interval for a single mean, the confidence interval will _____ with
an increase in the sample size.
a. decrease
b. increase
c. stay the same
d. increase or decrease, depending on the sample data
ANS: A PTS: 1 MSC: AACSB: Analytic | AACSB: Interval Estimation
4. When you calculate the sample size for a proportion, you use an estimate for the population
proportion; namely . A conservative value for n can be obtained by using = _____.
a. 0.01
b. 0.05
c. 0.10
d. 0.50
e. 1.00
ANS: D PTS: 1 MSC: AACSB: Analytic | AACSB: Interval Estimation
5. Confidence intervals are a function of which of the following three things?

a. The population, the sample, and the standard deviation
b. The sample, the variable of interest, and the degrees of freedom
c. The data in the sample, the confidence level, and the sample size
d. The sampling distribution, the confidence level, and the degrees of freedom
e. The mean, median, and mode
ANS: C PTS: 1 MSC: AACSB: Analytic | AACSB: Interval Estimation
6. Suppose there are 500 accounts in a population. You sample 50 of them and find a sample mean of
$500. What would be your estimate for the population total?
a. $5,000
b. $50,000
c. $250,000
d. $500,000
e. None of the above
7. When we replace with the sample standard deviation (s), we introduce a new source of variability
and the sampling distribution becomes the _____.
a. t -distribution c. chi-square distribution
b. F- distribution d. normal distribution
8. There are, generally speaking, two types of statistical inference. They are:
a. sample estimation and population estimation
b. confidence interval estimation and hypothesis testing
c. interval estimation for a mean and point estimation for a proportion
d. independent sample estimation and dependent sample estimation
9. If the odds of a horse winning a race are 2 to 1, then the probability of this horse winning the race
is_____.
a. 1/4
b. 1/3
c. 1/2
d. 2/3
10. The t-distribution for developing a confidence interval for a mean has _____ degrees of freedom.
a. n + 2
b. n +1
c. n
d. n – 1
e. n - 2
11. As the sample size increases, the t-distribution becomes more similar to the ________ distribution.
a. normal
b. exponential
c. multinominal
d. chi-square
e. binomial
12. When the samples we want to compare are paired in some natural way, such as pretest/posttest for each
person or husband/wife pairs, a more appropriate form of analysis is to not compare two separate
variables, but their _____.
a. difference
b. sum
c. ratio
d. total
e. product
13. A parameter such as is sometimes referred to as a ________ parameter, because many times we
need its value even though it is not the parameter of primary interest.
a. special
b. random
c. nuisance
d. independent
e. dependent
14. Two independent samples of sizes 50 and 50 are randomly selected from two populations to test the
difference between the population means, . The sampling distribution of the sample mean
difference is:
a. normally distributed
b. approximately normal
c. t - distributed with 98 degrees of freedom
d. chi-squared distributed with 99 degrees of freedom
15. Two independent samples of sizes 20 and 25 are randomly selected from two normal populations with
equal variances. In order to test the difference between the population means, the test statistic is:
a. a standard normal random variable
b. approximately standard normal random variable
c. t-distributed with 45 degrees of freedom
d. t-distributed with 43 degrees of freedom
16. After calculating the sample size needed to estimate a population proportion to within 0.05, you have
been told that the maximum allowable error (B) must be reduced to just 0.025. If the original
calculation led to a sample size of 1000, the sample size will now have to be:
a. 2000 c. 1000
b. 4000 d. 8000
17. From a sample of 500 items, 30 were found to be defective. The point estimate of the population
proportion defective will be:
a. 0.06 c. 16.667
b. 30.0 d. None of the above
18. The number of degrees of freedom needed to construct 90% confidence interval for the difference
between means when the data are gathered from paired samples, with 15 observations in each sample,
is:
a. 30 c. 28
b. 15 d. 14
19. The general form of a confidence interval is:

a. Point Estimate = Multiple × Standard Error
b. Point Estimate = Multiple +Standard Error
c. Point Estimate ± Multiple × Standard Error
d. Point Estimate = Multiple ± Standard Error
20. The approximate standard error of the point estimate of the population total is:
a. c.
b. d.
21. The chi-square distribution for developing a confidence interval for a standard deviation has_____
degrees of freedom.
a. n + 2
b. n +1
c. n
d. n – 1
e. n - 2
22. An example of a problem where the sample data would be paired is:
a. Difference between the means of appraised and sales house prices
b. Difference between the proportion of defective items from two suppliers
c. Difference in the mean life of two major brands of batteries
d. Difference in the mean salaries for graduates in two different academic fields at a
university
e. None of these options
23. The shape of a chi-square distribution

a. is symmetric c. is skewed to the right
b. is skewed to the left d. depends on the sample data
24. For a given confidence level, the procedure for controlling interval length usually begins with the
specification of
a. the point estimate c. the sample standard deviation, s
b. the population standard deviation, s d. the interval half-length, B
TRUE/FALSE
25. The standard error of the sampling distribution of the sample proportion , when the sample size n =
50 and the population proportion p = 0.25, is 0.00375.
ANS: F PTS: 1 MSC: AACSB: Analytic | AACSB: Interval Estimation
26. In developing a confidence interval for the population standard deviation , we make use of the fact
that the sampling distribution of the sample standard deviation s is not the normal distribution or the t-
distribution, but rather a right-skewed distribution called the chi-square distribution, which (for this
procedure) has n – 1 degrees of freedom.
ANS: T PTS: 1 MSC: AACSB: Analytic | AACSB: Interval Estimation
27. The interval estimate 18.5 2.5 was developed for a population mean when the sample standard
deviation s was 7.5. Had s equaled 15, the interval estimate would be 37 5.0.
28. As a general rule, the normal distribution is used to approximate the sampling distribution of the
sample proportion only if the sample size n is greater than 30.
29. In general, the paired-sample procedure is appropriate when the samples are naturally paired in some
way and there is a reasonably large positive correlation between the pairs. In this case, the paired-
sample procedure makes more efficient use of the data and generally results in narrower confidence
intervals.
30. If the standard error of the sampling distribution of the sample proportion is 0.0324 for samples of
size 200, then the population proportion must be 0.30.
31. The lower limit of the 95% confidence interval for the population proportion p, given that n = 300; and
= 0.10 is 0.1339.
32. If a random sample of size 250 is taken from a population, where it is known that the population
proportion p = 0.4, then the mean of the sampling distribution of the sample proportion is 0.60.
33. If two random samples of size 40 each are selected independently from two populations whose
variances are 35 and 45, then the standard error of the sampling distribution of the sample mean
difference, , equals 1.4142.
34. The degrees of freedom for the t and chi-square distributions is a numerical parameter of the
distribution that defines the precise shape of the distribution.
35. If two random samples of sizes 30 and 35 are selected independently from two populations whose
means are 85 and 90, then the mean of the sampling distribution of the sample mean difference,
, equals 5.
36. A confidence interval is an interval estimate for which there is a specified degree of certainty that the
actual true value of the population parameter will fall within the interval.
37. The 95% confidence interval for the population mean , given that the sample size n = 49 and the
population standard deviation = 7, is .
38. In order to construct a confidence interval estimate of the population mean , the value of must be
given.
39. We can form a confidence interval for the population total T by finding a confidence interval for the
population mean in the usual way, and then multiplying the lower and upper limits the confidence
interval by the population size N.
40. A 90% confidence interval estimate for a population mean is determined to be 72.8 to 79.6. If the
confidence level is reduced to 80%, the confidence interval for becomes narrower.
41. When samples of size n are drawn from a population, then the sampling distribution of the sample
mean is approximately normal, provided that n is reasonably large.
42. In developing confidence interval for the difference between two population means using two
independent samples, we use the pooled estimate in estimating the standard error of the sampling
distribution of the sample mean difference if the populations are normal with equal variances.
43. In general, increasing the confidence level will narrow the confidence interval, and decreasing the
confidence level widens the interval.

44. The upper limit of the 90% confidence interval for the population proportion p, given that n = 100; and
= 0.20 is 0.2658.
45. The t-distribution and the standard normal distribution are practically indistinguishable as the degrees
of freedom increase.
46. In determining the sample size n for estimating the population proportion p, a conservative value of n
can be obtained by using 0.50 as an estimate of p.
47. The mean of the sampling distribution of the sample proportion , when the sample size n = 100 and
the population proportion p = 0.15, is 15.0.
48. If a sample has 20 observations and a 95% confidence estimate for is needed, the appropriate value
of t-multiple is 2.093
49. The confidence interval for the population standard deviation s is centered at the point estimate, the
sample standard deviation s.
50. The approximate standard error of the point estimate of the population total is .
51. Samples of exam scores for employees before and after a training class would be examples of paired
data
52. If two samples contain the same number of observations, then the data must be paired.
53. If we cannot make the strong assumption that the variances of two samples are equal, then we must use
the pooled standard deviation in calculating the standard error of a difference between the means.
SHORT ANSWER
54. You are told that a random sample of 150 people from Iowa has been given cholesterol tests, and 60 of
these people had levels over the “safe” count of 200. Construct a 95% confidence interval for the
population proportion of people in Iowa with cholesterol levels over 200.
ANS:
= (0.3216, 0.4784).
PTS: 1 MSC: AACSB: Analytic | AACSB: Interval Estimation
55. You are trying to estimate the average amount a family spends on food during a year. In the past, the
standard deviation of the amount a family has spent on food during a year has been approximately
$1200. If you want to be 99% sure that you have estimated average family food expenditures within
$60, how many families do you need to survey?
ANS:
=1200, z - multiple = 2.575, B = 60. The sample size for a mean is given by
56. You have been assigned to determine whether more people prefer Coke to Pepsi. Assume that roughly
half the population prefers Coke and half prefers Pepsi. How large a sample would you need to take to
ensure that you could estimate, with 95% confidence, the proportion of people preferring Coke within
3% of the actual value?
ANS:
= 0.50, z - multiple = 1.96, B = 0.03. The sample size for a proportion is given by
NARRBEGIN: SA_57_59
A marketing research consultant hired by Coca-Cola is interested in determining the proportion of
customers who favor Coke over other soft drinks. A random sample of 400 consumers was selected
from the market under investigation and showed that 53% favored Coca-Cola over other brands.
NARREND
57. (A) Compute a 95% confidence interval for the true proportion of people who favor Coke. Do the
results of this poll convince you that a majority of people favors Coke?
(B) Suppose 2,000 (not 400) people were polled and 53% favored Coke. Would you now be convinced
that a majority of people favor Coke? Why might your answer be different than in (A)?
(C) How many people would have to be surveyed to be 95% confident that you can estimate the
fraction of people who favor Coca-Cola within 1%?
ANS:
(A) n =400, = 0.53
0.53 ± 0.0489 = (0.4811, 0.5789).

Since confidence interval ranges from about 48% to 57.9%, it is difficult to conclude that a majority of
people favors Coke. It could be below 50%.
(B) n = 2,000, = 0.53
0.53 ± 0.0219 = (0.5081, .5519).

In this case the 95% confidence interval is entirely above 50%, the data is now more convincing than it
was previously.
(C) = 0.53, z – multiple = 1.96, B = .01. The sample size for proportion is given by:
NARRBEGIN: SA_60_61
The employee benefits manager of a medium size business would like to estimate the proportion of
full-time employees who prefer adopting plan A of three available health care plans in the coming
annual enrollment period. A reliable frame of the company’s employees and their tentative health care
preferences are available. Using Excel, the manager chose a random sample of size 50 from the frame.
There were 17 employees in the sample who preferred plan A.
NARREND
58. (A) Construct a 99% confidence interval for the proportion of company employees who prefer plan A.
Assume that the population consists of the preferences of all employees in the frame.
(B) Interpret the 99% confidence interval constructed in (A).
ANS:
(A)
= (0.1675), 0.5125)
(B) We are 99% confident that the true proportion of all employees who prefer plan A is within this
confidence interval; that is, between 0.1675 and 0.5125.
NARRBEGIN: SA_62_64
Q-Mart is interested in comparing its male and female customers. Q-Mart would like to know if its
female charge customers spend more money, on average, than its male charge customers. They have
collected random samples of 25 female customers and 22 male customers. On average, women charge
customers spend $102.23 and men charge customers spend $86.46. Some information are shown
below.
Summary statistics for two samples

Female Male
Sample sizes 25 22
Sample means 102.23 86.46
Sample standard deviations 93.393 59.695
Confidence interval for difference between means
Sample mean difference 15.77

Pooled standard deviation 79.466
Std error of difference 23.23
NARREND
59. (A) Use a t - value of 2.014 to calculate a 95% confidence interval for the difference between the
average female purchase and the average male purchase. Would you conclude that there is a significant
difference between females and males in this case? Explain.
(B) What are the degrees of freedom for the t-multiple in this calculation? Explain how you would
calculate the degrees of freedom in this case.
(C) What is the assumption in this case that allows you to use the pooled standard deviation for this
confidence interval?
ANS:
(A)
15.77±46.785 = (-31.015, 62.555). Since the
range includes 0, it does not appear to be a significant difference between the means of the two groups.
(B) – 2 = 45
(C) In order to use the pooled standard deviation for this confidence interval, we must assume that the
two populations standard deviations are equal ( ).
NARRBEGIN: SA_65_66
A company employs two shifts of workers. Each shift produces a type of gasket where the thickness is
the critical dimension. The average thickness and the standard deviation of thickness for shift 1, based
on a random sample of 40 gaskets, are 10.85 mm and 0.16 mm, respectively. The similar figures for
shift 2, based on a random sample of 30 gaskets, are 10.90 mm and 0.19 mm. Let be the
difference in thickness between shifts 1 and 2, and assume that the population variances are equal.
NARREND
60. (A) Construct a 95% confidence interval for .
(B) Based on your answer to (A), are you convinced that the gaskets from shift 2 are, on average,
wider than those from shift 1? Why or why not?
ANS:
(A)
The pooled standard deviation is = 0.1734. Then
=(-0.1336, .0336).
(B) The confidence interval extends from a negative number (indicating shift 2 thickness is larger) to a
positive number (indicating shift 2 thickness is smaller). So we are not absolutely sure which mean is
greater.
NARRBEGIN: SA_67_68
A sample of 9 production managers with over 15 years of experience has an average salary of $71,000
and a sample standard deviation of $18,000.
NARREND
61. (A) You can be 95% confident that the mean salary for all production managers with at least 15 years
of experience is between what two numbers (the t-multiple with 8 degrees of freedom is 2.306)? What
assumption are you making about the distribution of salaries?
(B) What sample size would be needed to ensure that we could estimate the true mean salary of all
production managers with more than 15 years of experience and have only 5 chances in 100 of being
off by more than $4200?
ANS:
(A)
n = 9, = 71,000, s = 18,000. Then
$71,000 ± $13,836 = ($57,164, $84,836).
The assumption is that the population is normal or near normal. This is particularly important since the
sample size is so small (n = 9). However, the t-distribution is rather robust to violations of normality.
(B)
z-multiple = 1.96, =18,000, B = 4200
NARRBEGIN: SA_69_70
A sample of 40 country CD recordings of Willie Nelson has been examined. The average playing time
of these recordings is 51.3 minutes, and the standard deviation is 5.8 minutes.
NARREND
62. (A) Construct a 95% confidence interval for the mean playing time of all Willie Nelson recordings.
(B) Interpret the confidence interval you constructed in (A).
ANS:
(A)
n = 10, = 51.3, s = 5.8,
(B) We are 95% confident that the mean playing time of all Willie Nelson recordings is approximately
between. 49.4 and 53.2 minutes.
NARRBEGIN: SA_71_72
A department store is interested in the average balance that is carried on its store’s credit card. A
sample of 40 accounts reveals an average balance of $1,250 and a standard deviation of $350.
NARREND
63. (A) Find a 95% confidence interval for the mean account balance on this store’s credit card (the t-
multiple with 39 degrees of freedom is 2.0227).
(B) What sample size would be needed to ensure that we could estimate the true mean account balance
and have only 5 chances in 100 of being off by more than $100?
ANS:
(A)
n = 40, = 1250, s = 350,
(B)
z – multiple = 1.96, = 350, B = 100. Then,
NARRBEGIN: SA_73_74
A market research consultant hired by Coke Classic Company is interested in estimating the difference
between the proportions of female and male customers who favor Coke Classic over Pepsi Cola in
Chicago. A random sample of 200 consumers from the market under investigation showed the
following frequency distribution.
Male Female
Coke 72 38 110
Pepsi 58 32 90
130 70 200
NARREND
64. (A) Construct a 95% confidence interval for the difference between the proportions of male and female
customers who prefer Coke Classic over Pepsi Cola.
(B) Interpret the constructed confidence interval.
ANS:
(A)
, ,
= 0.0436 + 0.0595 = 0.1031
= (-0.1912, 0.213)
(B) We are 95% confident that the population difference between these proportions is between –
19.12% and 21.30%.
NARRBEGIN: SA_75_78
The percent defective for parts produced by a manufacturing process is targeted at 4%. The process is
monitored daily by taking samples of sizes n = 160 units. Suppose that today’s sample contains 14
defectives.
NARREND
65. (A) Determine a 95% confidence interval for the proportion defective for the process today.
(B) Based on your answer to (A), is it still reasonable to think the overall proportion defective
produced by today’s process is actually the targeted 4%? Explain your reasoning.
(C) The confidence interval in (A) is based on the assumption of a large sample size. Is this sample
size sufficiently large in this example? Explain how you arrived at your answer.
(D) How many units would have to be sampled to be 95% confident that you can estimate the fraction
of defective parts within 2% (using the information from today’s sample)?
ANS:
(A)
n = 160, = 0.0875,
(B) No, since 4% falls outside of this confidence interval.
(C) Yes. Because are all greater than 5.0.
(D) z- multiple = 1.96, = 0.0875

NARRBEGIN: SA_79_80
Auditors of Independent Bank are interested in comparing the reported value of all 1775 customer
savings account balances with their own findings regarding the actual value of such assets. Rather than
reviewing the records of each savings account at the bank, the auditors randomly selected a sample of
100 savings account balances from the frame. The sample mean and sample standard deviations were
$505.75 and 360.95, respectively.
NARREND
66. (A) Construct a 90% confidence interval for the total value of all savings account balances within this
bank. Assume that the population consists of all savings account balances in the frame.
ANS:
(A)
N = 1775, n = 100, = 505.75, s = 360.95, t – multiple = 1.6604,
= ($719,326.70, $1,004,085.8)
(B) We are 90% confident that the total balance of all 1775 savings account balances within the bank is
approximately between $791,327 and $1,004,086.
NARRBEGIN: SA_81_82
A real estate agent has collected a random sample of 40 houses that were recently sold in Grand
Rapids, Michigan. She is interested in comparing the appraised value and recent selling price (in
thousands of dollars) of the houses in this particular market. The values of these two variables for each
of the 40 randomly selected houses are shown below.
House Value Price House Value Price

1 140.93 140.24 21 136.57 135.35
2 132.42 129.89 22 130.44 121.54
3 118.30 121.14 23 118.13 132.98
4 122.14 111.23 24 130.98 147.53
5 149.82 145.14 25 131.33 128.49
6 128.91 139.01 26 141.10 141.93
7 134.61 129.34 27 117.87 123.55
8 121.99 113.61 28 160.58 162.03
9 150.50 141.05 29 151.10 157.39
10 142.87 152.90 30 120.15 114.55
11 155.55 157.79 31 133.17 139.54
12 128.50 135.57 32 140.16 149.92
13 143.36 151.99 33 124.56 122.08
14 119.65 120.53 34 127.97 136.51
15 122.57 118.64 35 101.93 109.41
16 145.27 149.51 36 131.47 127.29
17 149.73 146.86 37 121.27 120.45
18 147.70 143.88 38 143.55 151.96
19 117.53 118.52 39 136.89 132.54
20 140.13 146.07 40 106.11 114.33
NARREND
67. (A) Use the sample data to generate a 95% confidence interval for the mean difference between the
appraised values and selling prices of the houses sold in Grand Rapids.
(B) Interpret the constructed confidence interval fin (A) for the real estate agent.
ANS:
(A) We applied the paired sample analysis with , where: D =
Difference = Appraised value – selling price.
= (-3.785, 0.561) in thousands of dollars.
(B) We are 95% confident that the actual mean difference between the appraised values and selling
prices of all the houses sold in Grand Rapids is between -$3785 and $561.
NARRBEGIN: SA_83_87
THE FOLLOWING ITEMS REQUIRE THE USE OF EXCEL:
NARREND
68. (A) Compute has a t-distribution with 15 degrees of freedom.
(B) Compute has a t-distribution with 150 degrees of freedom.
(C) How do you explain the difference between the results obtained in (A) and (B)?
(D) Compute where Z is a standard normal random variable.
(E) Compare the results of (D) to the results obtained in (A) and (B). How do you explain the
difference in these probabilities?
ANS:
(A) By using the Excel function = TDIST (2, 15, 1) we get 0.03197.
(B) By using the Excel function = TDIST (2, 150, 1) we get 0.02365.
(C) The smaller the degrees of freedom, the larger the variance of t, and so the larger the tail
probabilities are.
(D) By using the Excel function = 1 – NORMDIST (2) we get 0.02275.

(E) First, the variance of t with a small degree of freedom is larger than a t with a large degree of
freedom, which is larger than for a Z. This explains why the probabilities in (A), (B) and (D) increases.
Second, when the sample size is large, the degrees of freedom of t are large; and that the t-distribution
and the standard normal distribution are practically indistinguishable. This explains why the
probabilities in Questions 83 and 84 are close.
NARRBEGIN: SA_88_90
Senior management of a consulting services firm is concerned about a growing decline in the firm’s
weekly number of billable hours. The firm expects each professional employee to spend at least 40
hours per week on work. In an effort to understand this problem better, management would like to
estimate the standard deviation of the number of hours their employees spend on work-related
activities in a typical week. Rather than reviewing the records of all the firm’s full-time employees, the
management randomly selected a sample of size 50 from the available frame. The sample mean and
sample standard deviations were 48.5 and 7.5 hours, respectively.
NARREND
69. (A) Construct a 99% confidence interval for the standard deviation of the number of hours this firm’s
employees spend on work-related activities in a typical week.
(C) Given the target range of 40 to 60 hours of work per week, should senior management be
concerned about the number of hours their employees are currently devoting to work? Explain why or
why not.
ANS:
(A) . The confidence interval formulas for (not presented in the book)
are:
Lower limit = =5.936
Upper limit =
(B) We are 99% confident that the population standard deviation is between 5.936 and 10.057.
(C) The best guess for the population mean is 48.5 hours per week, and about 95% of all employees
are within 2 standard deviations of this, where we are almost sure (99% sure) that this standard
deviation is between 5.9 and 10.1. But even if the standard deviation is only 5.9, then 48.5 standard
deviations will produce the range 36.7 to 60.3. Maybe management should be concerned.
NARRBEGIN: SA_91_94
Q-Mart is interested in comparing customer who used its own charge card with those who use other
types of credit cards. Q-Mart would like to know if customers who use the Q-Mart card spend more
money per visit, on average, than customers who use some other type of credit card. They have
collected information on a random sample of 38 charge customers as shown below. On average, the
person using a Q-Mart card spends $192.81 per visit and customers using another type of card spend
$104.47 per visit.
Summary statistics for two samples
Q-Mart Other Charges
Sample sizes 13 25
Sample means 192.81 104.47
Sample standard deviations 115.243 71.139
Confidence interval for difference between means
Sample mean difference 88.34

Pooled standard deviation 88.323
Std error of difference 30.201
NARREND
70. (A) Using a t - value of 2.0281, calculate a 95% confidence interval for the difference between the
average Q-Mart charge and the average charge on another type of credit card.
(B) What are the degrees of freedom for the t - multiple in this calculation? Explain how you would
calculate the degrees of freedom in this case.
(C) What is the assumption in this case that allows you to use the pooled standard deviation for this
confidence interval?
(D) Would you conclude that there is a significant difference between the two types of customers in
this case? Explain.
ANS:
(A) 88.34±61.25= (27.09, 149.59)
(B) – 2 = 13 + 25 – 2 = 36
(C) In order to use the pooled standard deviation for this confidence interval, we must assume that the
two population standard deviations are equal ( ).
(D) Since the confidence interval in (B) does not include 0, there appears to be a significant difference
between the means of the two groups. In this case, it appears as though the Q-Mart charge card holders
spend more money than those who use other types of charge cards.
NARRBEGIN: SA_95_96
An automobile dealer wants to estimate the proportion of customers who still own the cars they
purchased six years ago. A random sample of 200 customers selected from the automobile dealer’s
records indicates that 88 still own cars that were purchased six years earlier.
NARREND
71. (A) Construct a 95% confidence interval estimate of the population proportion of all customers who
still own the cars they purchased six years ago
(B) How can the result in (A) be used by the automobile dealer to study satisfaction with cars
purchased at the dealership?
ANS:
(A)
= 88/200 = 0.44, z – multiple = 1.96
(B) The dealer can infer that the proportion of all customers who still own the cars they purchased at
the dealership 6 years earlier is somewhere between 03712 and 0.5088 with a 95% level of confidence.
NARRBEGIN: SA_97_100
The average annual household income levels of citizens of selected U.S. cities are shown below.
City Household City Household City Household

Index Income Index Income Index Income
1 $54,300 21 $53,500 41 $61,500
2 $61,800 22 $45,600 42 $53,000
3 $61,400 23 $70,100 43 $51,000
4 $50,800 24 $108,700 44 $55,600
5 $56,200 25 $46,400 45 $51,600
6 $48,300 26 $56,700 46 $57,200
7 $61,600 27 $59,100 47 $54,300
8 $63,200 28 $46,300 48 $51,500
9 $55,200 29 $52,900 49 $53,500
10 $58,000 30 $56,300 50 $61,800
11 $77,600 31 $67,300 51 $44,800
12 $47,600 32 $63,800 52 $57,400
13 $62,700 33 $70,600 53 $48,100
14 $46,200 34 $49,800 54 $52,700
15 $64,300 35 $51,300 55 $57,400
16 $56,000 36 $56,600 56 $65,500
17 $53,400 37 $49,600 57 $59,600
18 $56,800 38 $67,400 58 $62,000
19 $51,200 39 $53,700 59 $49,700
20 $59,000 40 $48,700 60 $54,400
NARREND
72. (A) Use Excel to obtain a simple random sample of size 10 from this frame.
(B) Using the sample generated in (A), construct a 95% confidence interval for the mean average
annual household income level of citizens in the selected U.S. cities. Assume that the population
consists of all average annual household income levels in the given frame.
(C) Interpret the 95% confidence interval constructed in (B).
(D) Does the 95% confidence interval contain the actual population mean? If not, explain why not.
What proportion of many similarly constructed confidence intervals should include the true population
mean value?
ANS:
(A) The StatTools Random Sample tool under Data Utilities is used to generate a sample of size 10,
then the VLOOKUP function is used to get the corresponding incomes. The following sample is
obtained:
City Household City Household City Household

Index Income Index Income Index Income
1 $54,300 21 $53,500 41 $61,500
2 $61,800 22 $45,600 42 $53,000
3 $61,400 23 $70,100 43 $51,000
4 $50,800 24 $108,700 44 $55,600
5 $56,200 25 $46,400 45 $51,600
6 $48,300 26 $56,700 46 $57,200
7 $61,600 27 $59,100 47 $54,300
8 $63,200 28 $46,300 48 $51,500
9 $55,200 29 $52,900 49 $53,500
10 $58,000 30 $56,300 50 $61,800
11 $77,600 31 $67,300 51 $44,800
12 $47,600 32 $63,800 52 $57,400
13 $62,700 33 $70,600 53 $48,100
14 $46,200 34 $49,800 54 $52,700
15 $64,300 35 $51,300 55 $57,400
16 $56,000 36 $56,600 56 $65,500
17 $53,400 37 $49,600 57 $59,600
18 $56,800 38 $67,400 58 $62,000
19 $51,200 39 $53,700 59 $49,700
20 $59,000 40 $48,700 60 $54,400
(B) t-multiple = 2.2622
(C) We are 95% confident that the average annual household income level of all citizens is
approximately between $52,737 and $66,243.
(D) This confidence interval easily captures the true population mean of $57,043. Approximately 95%
of the confidence intervals constructed in this way should contain the true population mean.
The personnel department of a large corporation wants to estimate the family dental expenses of its
employees to determine the feasibility of providing a dental insurance plan. A random sample of 12
employees in 2004 reveals the following family dental expenses (in dollars): 115, 370, 250, 93, 540,
225, 177, 425, 318, 182, 275, and 228. Use StatTools for your calculations.
NARREND
73. (A) Construct a 90% confidence interval estimate of the mean family dental expenses for all
employees of this corporation.
(B) What assumption about the population distribution must be made to answer (A)?
(C) Interpret the 90% confidence interval constructed in (A).

(D) Suppose you used a 95% confidence interval in (A). What would be your answer?
(E) Suppose the fourth value were 593 instead of 93. What would be your answer to (A)? What effect
does this change have on the confidence interval?
(F) Construct a 90% confidence interval estimate for the standard deviation of family dental expenses
for all employees of this corporation.
(G) Interpret the 90% confidence interval constructed in (E).
ANS:
(A)
(B) The population of dental expenses must be approximately normally distributed.
(C) We are 90% confident that the mean family dental expenses for all employees of this corporation is
between $199.26 and $333.74.
(D)
(E)
The additional $500 in dental expenses, divided across the sample of 12, raises the mean by $41.67
and increases the standard deviation by nearly $18.20. The interval width increases over $23 in the
process.
(F)
The following values have been calculated using the TDIST and TINV functions in Excel. These
values come from a t- distribution with 15 degrees of freedom.
These values represent the probability to the right of the given positive values.
Value t
-probability
1.00 0.1666
1.20 0.1244
1.40 0.0909
These values represent the positive t- value for a given probability in both tails (sum of both tails).
Probability t -value
0.20 1.3406
0.10 1.7531
0.05 2.1315
NARREND
74. What is the probability of a t-value smaller than 1.00?
ANS:
0.8334
75. What is the probability of a t-value larger than 1.20?
ANS:
0.1244
76. What is the probability of a t-value between –1.40 and +1.40?
ANS:
0.8182
77. What would be the t-value where 0.05 of the values are in the upper tail?
ANS:
+ 1.7531
78. What would be the t-values where 0.10 of the values are in both tails (sum of both tails)?
ANS:
- 1.7531 and + 1.7531
79. What would be the t-values where 0.95 of the values would fall within this interval?
ANS:
- 2.1315 and + 2.1315

80. You are trying to estimate the average amount a family spends on food during a year. In the past the
standard deviation of the amount a family has spent on food during a year has been approximately
$800. If you want to be 95% sure that you estimated average family food expenditures within $50,
how many families do you need to survey?
ANS:
= 800, B = 50, z – multiple = 1.96
81. In past years, approximately 25% of all U.S. families purchased potato chips at least once a month. We
are interested in determining the fraction of all U.S. families that currently purchase potato chips at
least once a month. How many families must we survey if we want to be 99% sure that our estimate of
the fraction of U.S. families currently purchasing potato chips at least once a month is accurate within
2%?
ANS:
= 0.25, z – multiple = 2.575, B = 0.02
82. In constructing confidence interval estimate for the difference between the means of two populations,
where the unknown population variances are assumed not to be equal, summary statistics computed
from two independent samples are as follows: , , , , , and
. Construct 90% confidence interval for .
ANS:
t-multiple = 1.6646, = 6.776

= 20 1.6646(6.776) = 20 11.28 = (8.72, 31.28)
The widths of 100 elevator rails have been measured. The sample mean and standard deviation of the
elevator rails are 2.05 inches and 0.01 inch.
NARREND
83. (A) Construct a 95% confidence interval for the average width of an elevator rail. Do we need to
assume that the width of elevator rails follows a normal distribution?
(B) How large a sample of elevator rails would we have to measure to ensure that we could estimate,
with 95% confidence, the average diameter of an elevator rail within 0.01 inch?
ANS:
(A) n = 100,
The assumption of normality is not crucial here because the sample size is fairly large.
(B) = 0.01, B = 0.01, z – multiple = 1.96
A large regional department store is evaluating the effectiveness of its credit card program, which costs
it approximately $1m per year to administer. The store believes that for the credit card program to be
worthwhile, the administrative costs should be no more than 10% of the total of the average annual
account balances. Rather than reviewing each of the 15,000 individual accounts, the store’s analysts
randomly selected a sample of 500 average annual balances from the frame. The sample mean and
sample standard deviation were $215.75 and $55.90, respectively.
NARREND
84. (A) Construct a 95% confidence interval for the mean of the average annual credit account balances.
(C) Use the confidence interval constructed for (A) to help the store evaluate its criteria for whether or
not the credit card program is worthwhile.
ANS:
(A) n = 500, = 215.75, s = $55.90, t–multiple = 1.965,
(B) The store can be 95% confident that the typical average annual balance of all 15,000 credit card
accounts will be somewhere between $210.84 and $220.66.
(C) The point estimate of the total of the annual credit account balances is 15,000
($215.75)=$3,236,250. Similarly, the upper and lower confidence interval limits for the total are
15,000($210.84)=$3,162,575 and 15,000($220.66) = $3,309,925, respectively. The store’s current
administrative costs ($1,000,000) are over 30% of the best possible case for the total of the average
annual account balances (the upper limit of the interval, $3,309,925), therefore the credit card program
is not worthwhile.
A defensive driving training company is interested in evaluating the relative effectiveness of its two
main modes of training; online and traditional classroom. The company has collected a random sample
of 300 customers in a particular area, with the following results:
Online Classroom
Pass 143 128
Fail 17 12
Total 160 140
NARREND
85. (A) Construct a 95% confidence interval for the difference between the proportions of online and
classroom customers who pass the final exam.
(B) Interpret the confidence interval obtained in (A).
ANS:
(A)
, , , and
= 0.0244 + 0.0237 = 0.048
(B) We are 95% confident that the population difference between these proportions is between –0.1147
and 0.0736. Since the interval includes both negative differences and positive differences, the company
really cannot conclude whether one form of training is better than the other or vice-versa.

Chapter - 08 Con Interval

Uploaded by

Copyright:

Available Formats

Chapter - 08 Con Interval

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Chapter - 08 Con Interval

Uploaded by

Copyright:

Available Formats

CHAPTER 8: Confidence Interval Estimation

2. If you increase the confidence level, the confidence interval _____.

5. Confidence intervals are a function of which of the following three things?

19. The general form of a confidence interval is:

ANS: D PTS: 1 MSC: AACSB: Analytic | AACSB: Interval Estimation

23. The shape of a chi-square distribution

ANS: T PTS: 1 MSC: AACSB: Analytic | AACSB: Interval Estimation

ANS: F PTS: 1 MSC: AACSB: Analytic | AACSB: Interval Estimation

ANS: F PTS: 1 MSC: AACSB: Analytic | AACSB: Interval Estimation

ANS: T PTS: 1 MSC: AACSB: Analytic | AACSB: Interval Estimation

ANS: F PTS: 1 MSC: AACSB: Analytic | AACSB: Interval Estimation

ANS: F PTS: 1 MSC: AACSB: Analytic | AACSB: Interval Estimation

ANS: F PTS: 1 MSC: AACSB: Analytic | AACSB: Interval Estimation

ANS: T PTS: 1 MSC: AACSB: Analytic | AACSB: Interval Estimation

ANS: F PTS: 1 MSC: AACSB: Analytic | AACSB: Interval Estimation

ANS: T PTS: 1 MSC: AACSB: Analytic | AACSB: Interval Estimation

ANS: T PTS: 1 MSC: AACSB: Analytic | AACSB: Interval Estimation

ANS: F PTS: 1 MSC: AACSB: Analytic | AACSB: Interval Estimation

ANS: T PTS: 1 MSC: AACSB: Analytic | AACSB: Interval Estimation

ANS: T PTS: 1 MSC: AACSB: Analytic | AACSB: Interval Estimation

ANS: T PTS: 1 MSC: AACSB: Analytic | AACSB: Interval Estimation

ANS: T PTS: 1 MSC: AACSB: Analytic | AACSB: Interval Estimation

ANS: F PTS: 1 MSC: AACSB: Analytic | AACSB: Interval Estimation

ANS: T PTS: 1 MSC: AACSB: Analytic | AACSB: Interval Estimation

ANS: T PTS: 1 MSC: AACSB: Analytic | AACSB: Interval Estimation

ANS: T PTS: 1 MSC: AACSB: Analytic | AACSB: Interval Estimation

ANS: F PTS: 1 MSC: AACSB: Analytic | AACSB: Interval Estimation

ANS: T PTS: 1 MSC: AACSB: Analytic | AACSB: Interval Estimation

ANS: F PTS: 1 MSC: AACSB: Analytic | AACSB: Interval Estimation

ANS: T PTS: 1 MSC: AACSB: Analytic | AACSB: Interval Estimation

ANS: T PTS: 1 MSC: AACSB: Analytic | AACSB: Interval Estimation

ANS: F PTS: 1 MSC: AACSB: Analytic | AACSB: Interval Estimation

ANS: F PTS: 1 MSC: AACSB: Analytic | AACSB: Interval Estimation

PTS: 1 MSC: AACSB: Analytic | AACSB: Interval Estimation

PTS: 1 MSC: AACSB: Analytic | AACSB: Interval Estimation

PTS: 1 MSC: AACSB: Analytic | AACSB: Interval Estimation

(A) n =400, = 0.53

0.53 ± 0.0489 = (0.4811, 0.5789).

(B) n = 2,000, = 0.53

0.53 ± 0.0219 = (0.5081, .5519).

PTS: 1 MSC: AACSB: Analytic | AACSB: Interval Estimation

(B) Interpret the 99% confidence interval constructed in (A).

PTS: 1 MSC: AACSB: Analytic | AACSB: Interval Estimation

Summary statistics for two samples

Confidence interval for difference between means

Sample mean difference 15.77

PTS: 1 MSC: AACSB: Analytic | AACSB: Interval Estimation

The pooled standard deviation is = 0.1734. Then

PTS: 1 MSC: AACSB: Analytic | AACSB: Interval Estimation

PTS: 1 MSC: AACSB: Analytic | AACSB: Interval Estimation

(B) Interpret the confidence interval you constructed in (A).

PTS: 1 MSC: AACSB: Analytic | AACSB: Interval Estimation

PTS: 1 MSC: AACSB: Analytic | AACSB: Interval Estimation

(B) Interpret the constructed confidence interval.

= 0.0436 + 0.0595 = 0.1031

PTS: 1 MSC: AACSB: Analytic | AACSB: Interval Estimation

(B) No, since 4% falls outside of this confidence interval.

(C) Yes. Because are all greater than 5.0.

(D) z- multiple = 1.96, = 0.0875