Essentials of Business Statistics Communicating With Numbers 1st Edition Jaggia Solutions Manual

Essentials of Business Statistics
Communicating with Numbers 1st

Edition Jaggia Solutions Manual
Visit to Download in Full: https://testbankdeal.com/download/essentials-of-business-st
atistics-communicating-with-numbers-1st-edition-jaggia-solutions-manual/
Chapter 07 - Sampling and Sampling Distributions
Chapter 7. Sampling and Sampling

Distributions
Solutions
1.
a. To get a simple random sample of all iPad purchasers, you could randomly select
customers from all of the stores that sell iPads and ask their ages and whether or
not they have purchased an iPad.
b. To get a stratified random sample you could first create strata based on ethnicity;
for example under white, black, Hispanic, Asian, and then randomly select people in
each group and ask his/her age when he/she purchased an iPad.
c. To get a cluster sample, you could make clusters of stores that sell iPads, and
randomly select certain stores and gather data about the ages of people who
purchase iPads there.
2. There could be some nonresponse bias due to certain people choosing not to stop at the
booth. There could also be some selection bias since the booth is only open during the
weekend. Some people may avoid the mall on the weekend and therefore the booth will
not get information about people who shop and eat at the food courts during the week.
Also, since they are trying to determine the monthly expenditure, and they only gather
information from one weekend, this could cause some bias as well.
3. Natalie’s analysis is not based on a representative sample because she is only surveying
students in her accounting class. Since students other than accounting students are likely
to apply to the MBA program, Natalie should include other majors/concentrations as well.
In addition, since she will likely be competing with students from other schools as well, she
should also include information from students at other schools who are likely to apply to
the Berkeley MBA program.
4.
a. This could lead to nonresponse bias since people who do not mail the envelope
back may have different preferences than people who do respond.
b. This could lead to selection bias, since customers who frequent the store in the
morning are more likely to prefer an earlier opening time than people who
frequent the store at other times of the day.
c. This could also lead to selection bias and nonresponse bias since not everyone in
Grover Beach reads the newspaper and not everyone can get online to the store’s
website.
7-1
© 2014 by McGraw-Hill Education. This is proprietary material solely for authorized instructor use. Not authorized for sale or distribution in any
manner. This document may not be copied, scanned, duplicated, forwarded, distributed, or posted on a website, in whole or part.
5.
a. For a simple random sample you could have the clerks randomly select customers
who visit the Vons in Grover Beach throughout the day. The clerks could ask them
and record how likely they would be to visit the store between 6am and 7am.
b. For a stratified random sample, you could create groups of customers based on age
or time of day visiting the store and randomly sample customers in each group to
ask how likely they would be to shop at Vons between 6am and 7am.
c. For cluster sampling, you could cluster people by neighborhood that they live in
and randomly select neighborhoods to do door-to-door surveys.
6.
a. Both sample means have a normal distribution because the population is normally
distributed.
b. Yes, because for both sample sizes 𝑋̅ is normally distributed.
c.
12.5−12
i. For 𝑛 = 20, 𝑃(𝑋̅ < 12.5) = 𝑃 (𝑍 < ) = 𝑃(𝑍 < 1.49) = 0.9319.
1.5/√20
12.5−12
ii. For 𝑛 = 40, 𝑃(𝑋̅ < 12.5) = 𝑃 (𝑍 < ) = 𝑃(𝑍 < 2.11) = 0.9826.
1.5/√40
7.
a. Since we do not know whether or not the population is normally distributed and
𝑛 = 16 < 30, we cannot assume that the distribution of 𝑋̅ with 𝑛 = 16 is
approximately normal. However, for 𝑛 = 36, by the Central Limit Theorem, 𝑋̅ is
approximately normal.
b. We can only use the normal approximation for the sample mean with 𝑛 = 36.
66−66 68−66
c. 𝑃(66 ≤ 𝑋̅ ≤ 68) = 𝑃 ( 5.5 ≤𝑍≤ 5.5 ) = 𝑃(0 ≤ 𝑍 ≤ 2.18) = 0.9854 − 0.5 =
√36 √36
0.4854.
8.
14
a. 𝐸(𝑋̅) = 𝐸(𝑋) = 80; 𝑠𝑒(𝑋̅) = = 1.4
√100
77−80 85−80
b. 𝑃(77 ≤ 𝑋̅ ≤ 85) = 𝑃 (14/ 100 ≤ 𝑍 ≤ 14/ 100) = 𝑃(−2.14 ≤ 𝑍 ≤ 3.57) = 0.9998 −
√ √
0.0162 = 0.9836.
84−80
c. 𝑃(𝑋̅ > 84) = 𝑃 (𝑍 > 14/ 100) = 𝑃(𝑍 > 2.86) = 1 − 0.9979 = 0.0021.
√
9.
2
a. 𝐸(𝑋̅) = 𝐸(𝑋) = −9.5; 𝑠𝑒(𝑋̅) = = 0.2828.
√50
−10−(−9.5)
b. 𝑃(𝑋̅ < −10) = 𝑃 (𝑍 < 2/ 50 ) = 𝑃(𝑍 < −1.77) = 0.0384.
√
−10−(−9.5) −9−(−9.5)
c. 𝑃(−10 ≤ 𝑋̅ ≤ −9) = 𝑃 ( 2/ 50 ≤ 𝑍 ≤ 2/ 50 ) = 𝑃(−1.77 ≤ 𝑍 ≤ 1.77) =
√ √
0.9616 − 0.0384 = 0.9232.
7-2
10.
105−100
a. 𝑃(𝑋̅ > 105) = 𝑃 (𝑍 > 20/ 50 ) = 𝑃(𝑍 > 1.77) = 1 − 0.9616 = 0.0384.
√
95−100
b. 𝑃(𝑋̅ < 95) = 𝑃 (𝑍 < 20/√50
) = 𝑃(𝑍 < −1.77) = 0.0384.
c. 𝑃(95 ≤ 𝑋̅ ≤ 105) = 0.9616 − 0.0384 = 0.9232.
11.
325−330
a. 𝑃(𝑋 < 325) = 𝑃 (𝑍 < 4
) = 𝑃(𝑍 < −1.25) = 0.1056.
325−330
b. 𝑃(𝑋̅ < 325) = 𝑃 (𝑍 < 4/√6
) = 𝑃(𝑍 < −3.06) = 0.0011.
325−330
c. 𝑃(𝑋̅ < 325) = 𝑃 (𝑍 < ) = 𝑃(𝑍 < −4.33) = 0.0001.
4/√12
d. The probability that the mean weight of a 12-pack of beer is less than 325 ml is
much less than that of a single bottle because the variation in 𝑋̅ is less when the
sample size is bigger.
12.
18−16
a. 𝑃(𝑋̅ ≥ 18) = 𝑃 (𝑍 ≥ 7/ 42 ) = 𝑃(𝑍 ≥ 1.85) = 1 − 0.9678 = 0.0322.
√
17.5−16
b. 𝑃(𝑋̅ ≥ 17.5) = 𝑃 (𝑍 ≥ 7/√90
) = 𝑃(𝑍 ≥ 2.03) = 1 − 0.9788 = 0.0212.
c. Janice. A higher probability found in part a suggests that Janice's findings are more
likely if a representative sample is used.
13. The probability that the total weight of the 16 selected persons is greater than 3,200
pounds is the same as the probability that their average weight is greater than 3,200/16 =
200 pounds per person.
200−180
𝑃(𝑋̅ > 200) = 𝑃 (𝑍 > ) = 𝑃(𝑍 > 2.86) = 1 − 0.9979 = 0.0021.
28/√16
Therefore, there is a 0.0021 probability that a random sample of 16 individuals will exceed
the weight limit of 3,200 pounds.
14.
a. The sample mean with 𝑛 = 16 has a normal distribution because the sample is
taken from a normally distributed population; in addition, 𝐸(𝑋̅) = 22; 𝑠𝑒(𝑋̅) =
5
= 1.25.
√16
25−22
b. 𝑃(𝑋̅ > 25) = 𝑃 (𝑍 > 5/ 16 ) = 𝑃(𝑍 > 2.4) = 1 − 0.9918 = 0.0082.
√
18−22 24−22
c. 𝑃(18 ≤ 𝑋̅ ≤ 24) = 𝑃 ( 5/ 16 ≤𝑍≤ 5/√16
) = 𝑃(−3.20 ≤ 𝑍 ≤ 1.60) = 0.9452 −
√
0.0007 = 0.9445.
7-3
15.
35−30
a. 𝑃(𝑋̅ ≥ 35) = 𝑃 (𝑍 ≥ 6/ 5 ) = 𝑃(𝑍 ≥ 1.86) = 1 − 0.9686 = 0.0314. Therefore,
√
there is a 3.14% chance of getting a sample average of 35 or more without a
discount.
b. We feel reasonably confident that the manager's discount strategy has worked
since there is only a small chance of 3.14% of getting 35 or more customers if the
manager had not offered the discount.
16.
25,000−27,200
a. 𝑃(𝑋̅ > 25,000) = 𝑃 (𝑍 > ) = 𝑃(𝑍 > −0.63) = 1 − 0.2643 = 0.7357
7000/√4
30,000−27,200
b. 𝑃(𝑋̅ > 30,000) = 𝑃 (𝑍 > ) = 𝑃(𝑍 > 0.80) = 1 − 0.7881 = 0.2119
7000⁄√4
17.
5,000 125−115
𝑃(∑40 ̅
𝑖=1 𝑋𝑖 > 5,000) = 𝑃 (𝑋 > 40
) = 𝑃 (𝑍 > 35⁄√40
) = 𝑃(𝑍 > 1.80) = 1 − 0.9641 =
0.0359
18.
1,000,000−800,000
a. 𝑃(𝑋 > 1,000,000) = 𝑃 (𝑍 > ) = 𝑃(𝑍 > 0.80) = 1 − 0.7881 =
250,000
0.2119
1,000,000−800,000
b. P(∑4𝑖=1 𝑋𝑖 > 4,000,000) = P(𝑋̅ > 1,000,000) = 𝑃 (𝑍 > 250,000⁄√4
)
= P(𝑍 > 1.60) = 1 − 0.9452 = 0.0548
19.
35−33.8
a. 𝑃(𝑋 > 35) = 𝑃 (𝑍 > 3.5
) = 𝑃(𝑍 > 0.34) = 1 − 0.6331 = 0.3669
35−33.8
b. 𝑃(𝑋̅ > 35) = 𝑃 (𝑍 > 3.5⁄ 4 ) = 𝑃(𝑍 > 0.69) = 1 − 0.7549 = 0.2451
√
c. Since the mileages of the cars are independently distributed, the probability that all
four cars get more than 35 mpg is (𝑃(𝑋 > 35))4 = (0.3669)4 = 0.0181
20.
90−100
a. 𝑃(𝑋 < 90) = 𝑃 (𝑍 < 16
) = 𝑃(𝑍 < −0.63) = 0.2643
90−100
b. 𝑃(𝑋̅ < 90) = 𝑃 (𝑍 < 16⁄ 4 ) = 𝑃(𝑍 < −1.25) = 0.1056
√
c. Since the IQ scores are independently distributed, the probability that all four
people have an IQ score of less than 90 is:
(𝑃(𝑋 < 90))4 = (0.2643)4 = 0.0049
7-4
21.
0.68(1−0.68)
a. 𝐸(𝑃̅) = 𝑝 = 0.68; 𝑠𝑒(𝑃̅) = √ 20
= 0.1043. It is appropriate to use the
normal distribution approximation because the approximation conditions are
satisfied: 𝑛𝑝 = 20(0.68) = 13.6 > 5 and 𝑛(1 − 𝑝) = 20(1 − 0.68) = 6.4 > 5.
0.68(1−0.68)
b. 𝐸(𝑃̅) = 𝑝 = 0.68; 𝑠𝑒(𝑃̅) = √ 50
= 0.066. It is appropriate to use the normal
distribution approximation because the approximation conditions are satisfied:
𝑛𝑝 = 50(0.68) = 34 > 5 and 𝑛(1 − 𝑝) = 50(1 − 0.68) = 16 > 5.
22.
0.12(1−0.12)
a. With 𝑛 = 20, 𝐸(𝑃̅) = 𝑝 = 0.12; 𝑠𝑒(𝑃̅) = √ 20
= 0.0727. With 𝑛 =
0.12(1−0.12)
50, 𝐸(𝑃̅) = 𝑝 = 0.12; 𝑠𝑒(𝑃̅) = √ = 0.0460. The sampling distribution of
50
the sample proportion is approximately normal when 𝑛 = 50 but not when 𝑛 = 20.
When 𝑛 = 20 the approximation condition is not satisfied because 𝑛𝑝 =
20(0.12) = 2.4 < 5. When 𝑛 = 50 the approximation conditions are satisfied: 𝑛𝑝 =
50(0.12) = 6 > 5 and 𝑛(1 − 𝑝) = 50(1 − 0.12) = 44 > 5.
b. As shown in part (a), you can only use the normal approximation for the sample
size of 50. For the sample with n = 20, you cannot assume that 𝑃̅ is approximately
normally distributed.
0.10−0.12 0.12−0.12
c. 𝑃(0.10 ≤ 𝑃̅ ≤ 0.12) = 𝑃 ( ≤𝑍≤ ) = 𝑃(−0.44 ≤ 𝑍 ≤ 0) =
0.12(1−0.12) 0.12(1−0.12)
√ √
50 50
0.5 − 0.3300 = 0.1700.
23.
0.75(1−0.75)
a. 𝐸(𝑃̅) = 𝑝 = 0.75; 𝑠𝑒(𝑃̅) = √ 200
= 0.0306.
0.70−0.75 0.80−0.75
b. 𝑃(0.70 ≤ 𝑃̅ ≤ 0.80) = 𝑃 ( ≤𝑍≤ ) = 𝑃(−1.63 ≤ 𝑍 ≤ 1.63) =
0.75(1−0.75) 0.75(1−0.75)
√ √
200 200
0.9484 − 0.0516 = 0.8968.
0.70−0.75
c. 𝑃(𝑃̅ < 0.70) = 𝑃 (𝑍 < ) = 𝑃(𝑍 < −1.63) = 0.0516.
0.75(1−0.75)
√
200
24.
0.82(1−0.82)
a. The sampling distribution of 𝑃̅ has 𝐸(𝑃̅) = 𝑝 = 0.82 and 𝑠𝑒(𝑃̅) = √ 100
=
0.038; it is approximately normal because the approximation conditions are met:
𝑛𝑝 = 100(0.82) = 82 > 5 and 𝑛(1 − 𝑝) = 100(1 − 0.82) = 18 > 5.
7-5
0.80−0.82
b. 𝑃(𝑃̅ < 0.80) = 𝑃 (𝑍 < ) = 𝑃(𝑍 < −0.52) = 0.3015.
0.82(1−0.82)
√
100
0.80−0.82 0.84−0.82
c. 𝑃(0.80 ≤ 𝑃̅ ≤ 0.84) = 𝑃 ( ≤𝑍≤ ) = 𝑃(−0.52 ≤ 𝑍 ≤ 0.52) =
0.82(1−0.82) 0.82(1−0.82)
√ √
100 100
0.6985 − 0.3015 = 0.3970
25.
0.17(1−0.17)
a. The sampling distribution of 𝑃̅ has 𝐸(𝑃̅) = 𝑝 = 0.17 and 𝑠𝑒(𝑃̅) = √ 200
=
0.0266; The normal approximation criteria are met because 𝑛𝑝 = 200(0.17) =
34 > 5 and 𝑛(1 − 𝑝) = 200(1 − 0.17) = 166 > 5. Therefore, it is appropriate to
use the normal distribution approximation for the sample proportion.
0.20−0.17
b. 𝑃(𝑃̅ > 0.20) = 𝑃 (𝑍 > ) = 𝑃(𝑍 > 1.13) = 1 − 0.8708 = 0.1292
0.17(1−0.17)
√
200
26.
26
a. 𝑝̅ = 40 = 0.65, 𝑝 = 0.70 and 𝑛 = 40. Therefore 𝑃(𝑃̅ < 0.65) = 𝑃 (𝑍 <
0.65−0.70
) = 𝑃(𝑍 < −0.69) = 0.2451, so there is a 25.41% chance of getting fined.
0.70(1−0.70)
√
40
22 0.55−0.70
b. 𝑝̅ = 40 = 0.55. Therefore 𝑃(𝑃̅ < 0.55) = 𝑃 (𝑍 < ) = 𝑃(𝑍 < −2.07) =
0.70(1−0.70)
√
40
0.0192, so there is a 1.92% chance of getting dissolved.
27.
60
a. 𝑝̅ = 200 = 0.30, 𝑝 = 0.26 and 𝑛 = 200; Therefore 𝑃(𝑃̅ < 0.30) = 𝑃 (𝑍 <
0.30−0.26
) = 𝑃(𝑍 < 1.29) = 0.9015
0.26(1−0.26)
√
200
b. 𝑝 = 0.26 is the proportion of French people who approve, therefore 1 − 0.26 =

0.74 is the population proportion of disapproval.
150 0.75−0.74
𝑝̅ = 200 = 0.75; 𝑃(𝑃̅ > 0.75) = 𝑃 (𝑍 > ) = 𝑃(𝑍 > 0.32) = 1 −
0.74(1−0.74)
√
200
0.6255 = 0.3745.
7-6
28. You would choose 50 balls because with larger sample sizes the standard deviation of 𝑃̅ is
reduced. The proportion of green balls is 60/100 = 0.6. Therefore your probability of
getting 70% green balls is slightly higher with a smaller sample because of the increased
standard deviation. If you are unsure about this you can calculate 𝑃(𝑃̅ > 0.70) with 𝑛 =
50 and 𝑛 = 100 to confirm that the probability is higher for 𝑛 = 50.
0.70−0.60
For 𝑛 = 50, 𝑃(𝑃̅ > 0.70) = 𝑃 (𝑍 > ) = 𝑃(𝑍 > 1.44) = 1 − 0.9251 = 0.0749.
0.60(1−0.60)
√
50
0.70−0.60
For 𝑛 = 100, 𝑃(𝑃̅ > 0.70) = 𝑃 (𝑍 > ) = 𝑃(𝑍 > 2.04) = 1 − 0.9793 = 0.0207.
0.60(1−0.60)
√
100
29.
0.20−0.23
a. For 𝑛 = 50, 𝑃(𝑃̅ > 0.20) = 𝑃 (𝑍 > ) = 𝑃(𝑍 > −0.50) = 1 − 0.3085 =
0.23(1−0.23)
√
50
0.6915.
0.20−0.23
b. For 𝑛 = 200, 𝑃(𝑃̅ > 0.20) = 𝑃 (𝑍 > ) = 𝑃(𝑍 > −1.01) = 0.1562 =
0.23(1−0.23)
√
200
0.8438.
c. The difference between the two probabilities is from the sample size. The standard
deviation is reduced with larger sample sizes, which brings the sample proportion
closer to the population proportion. Therefore, when 𝑛 = 200 there is a greater
probability that the sample proportion will be closer to the population proportion
of 23%, so there is a greater probability that the sample proportion will be more
than 20%.
30.
81
a. 𝐸(𝑋̅) = 𝜇 = −45; 𝑠𝑒(𝑋̅) = √100 = 0.90. It is not necessary to apply the finite
population correction because the sample constitutes less than 5 percent of the
population: n = 100 < 125 = 2500(0.05).
−47−(−45) −43−(−45)
b. 𝑃(−47 ≤ 𝑋̅ ≤ −43) = ( ≤𝑍≤ ) = 𝑃(−2.22 ≤ 𝑍 ≤ 2.22) =
0.90 0.90
0.9868 − 0.0132 = 0.9736.
−44−(−45)
c. 𝑃(𝑋̅ > −44) = 𝑃 (𝑍 > 0.90
) = 𝑃(𝑍 > 1.11) = 1 − 0.8665 = 0.1335.
7-7
31.
324 500−70
a. 𝐸(𝑋̅) = 220; 𝑠𝑒(𝑋̅) = √ 70 √ 500−1 = 1.997. Yes, it is necessary to apply the finite
population correction because the sample constitutes for more than 5 percent of
the population: n = 70 > 25 = 500(0.05).
210−220
b. 𝑃(𝑋̅ < 210) = 𝑃 (𝑍 < 1.997 ) = 𝑃(𝑍 < −5.01) = 0 (approximately)
215−220) 230−220
c. 𝑃(215 ≤ 𝑋̅ ≤ 230) = ( 1.997 ≤ 𝑍 ≤ 1.997 ) = 𝑃(−2.50 ≤ 𝑍 ≤ 5.01) =
𝑃(𝑍 ≤ 5.01) − 𝑃(𝑍 < −2.50) = 1 − 0.0062 = 0.9938
32.
0.34(1−0.34)
a. 𝐸(𝑃̅) = 𝑝 = 0.34; 𝑠𝑒(𝑃̅) = √ = 0.047; there is no need to apply the finite
100
population correction because the sample size accounts for less than 5 percent of
the population size: 𝑛 = 100 < 150 = 0.05(3,000).
0.37−0.34
b. 𝑃(𝑃̅ > 0.37) = 𝑃 (𝑍 > ) = 𝑃(𝑍 > 0.63) = 1 − 0.7357 = 0.2643.
0.047
33.
0.46(1−0.46) 600−80
a. 𝐸(𝑃̅) = 0.46; 𝑠𝑒(𝑃̅) = √ 80
√
600−1
= 0.0519. Yes, it is necessary because
the sample size accounts for more than 5 percent of the population size: 𝑛 = 80 >
0.05(600) = 30.
0.40−0.46
b. 𝑃(𝑃̅ < 0.40) = 𝑃 (𝑍 < ) = 𝑃(𝑍 < −1.16) = 0.1230.
0.0519
34.
a. Yes, it is necessary because the sample size is greater than 5 percent of the
population size: 𝑛 = 32 > 0.05(500) = 25.
b. The sampling distribution of the sample mean is approximately normal because the
sample size is greater than 30.
9.78 500−32
c. 𝐸(𝑋̅) = 10.32; 𝑠𝑒(𝑋̅) = 32 √ 500−1 = 1.6743.
√
12−10.32
d. 𝑃(𝑋̅ > 12) = 𝑃 (𝑍 > ) = 𝑃(𝑍 > 1.00) = 1 − 0.8413 = 0.1587.
1.6743
35.
a. No, it is not necessary because the sample size accounts for less than 5 percent of
the population size: 𝑛 = 12 < 0.05(500) = 25.
b. We cannot assume the sampling distribution of the sample mean is approximately
normally distributed because we do not know if the population has a normal
distribution and the sample size is not sufficiently large to assume so (𝑛 < 30).
9.78
c. 𝐸(𝑋̅) = 10.32; 𝑠𝑒(𝑋̅) = = 2.8232.
√12
d. The normal approximation is not justified (see part b).
7-8
36.
a. 𝑛𝑝 = 20(0.60) = 12 and 𝑛(1 − 𝑝) = 20(1 − 0.60) = 8. Therefore, since 𝑛𝑝 ≥ 5
and 𝑛(1 − 𝑝) ≥ 5, the sampling distribution of the sample proportion is
approximately normal. Since the sample accounts for more than 5 percent of the
population size (𝑛 = 20 > 0.05(250) = 12.5), we need to apply the finite
0.60(1−0.60) 250−20
population correction; 𝐸(𝑃̅ ) = 0.46 and 𝑠𝑒(𝑃̅) = √ 20
√
250−1
= 0.1053.
0.50−0.60
b. 𝑃(𝑃̅ < 0.50) = 𝑃 (𝑍 < 0.1053 ) = 𝑃(𝑍 < −0.95) = 0.1711.
2
37. First define: 𝑛 = 120, 𝑁 = 1,000, 𝑝 = 3 = 0.67.
Since the sample accounts for more than 5 percent of the population size (n =120 >
0.05(1000) = 50), we need to apply the finite population correction; we then find E(𝑃̅ ) =
0.67(1−0.67) 1,000−120 75
0.67 and 𝑠𝑒(𝑃̅) = √ √ = 0.0403. Then, = 0.625, so we compute
120 1,000−1 120
0.625−0.67
𝑃(𝑃̅ > 0.625) = 𝑃 (𝑍 > 0.0403 ) = 𝑃(𝑍 > −1.12) = 1 − 0.1314 = 0.8686.
38.
a.
110
100
90
Centerline
80
UCL
70 LCL
60
50
1 2 3 4 5
Centerline: µ = 80
14
UCL = 80 + 3 ( 5) = 98.78
√
14
LCL = 80 − 3 ( 5) = 61.22
√
b.
7-9
100
90
Centerline
80
UCL
LCL
70
60
1 2 3 4 5
Centerline: µ = 80
14
UCL = 80 + 3 ( ) = 80 + 13.28 = 93.28
√10
14
LCL = 80 − 3 ( 10) = 80 − 13.28 = 66.72
√
c. The larger sample size gives narrower control limits due to the smaller standard
deviation.
39.
a.
35
30
25
Centerline
20
UCL
15 LCL
10
1 2 3 4 5 6
Sample Number
Centerline: µ = 20
10
UCL = 20 + 3 ( ) = 20 + 6 = 26
√25
10
LCL = 20 − 3 ( 25) = 20 − 6 = 14
√
b.
7-10
35
30
25 Centerline
20 UCL
LCL
15
Sample Mean
10
1 2 3 4 5 6
Sample Number
c. The last two points are outside the upper control limit. There is also an upward
trend, suggesting the process is becoming increasingly out of control. The process
should be adjusted.
40.
a.
190
180
170
160
150 Centerline
140 UCL
130 LCL
120
110
1 2 3 4 5
Sample Number
Centerline: µ = 150
42
UCL = 150 + 3 ( ) = 150 + 21 = 171
√36
42
LCL = 150 − 3 ( 36) = 150 − 21 = 129
√
b.
7-11
190
180
170
Centerline
160
UCL
150
LCL
140
Sample Mean
130
120
1 2 3 4 5
Sample Number
c. There are no points outside of the control limits. However, there is a positive trend
suggesting that the process may soon have a mean outside of the upper control
limit if it is not adjusted.
41.
a.
0.1
0.08
0.06
Centerline
0.04
UCL
0.02
LCL
0
1 2 3 4 5
Sample Number
Centerline: 𝑝 = 0.04
0.04(1−0.04)
UCL = 0.04 + 3√ = 0.04 + 0.037 = 0.077
250
0.04(1−0.04)
LCL = 0.04 − 3√ 250
= 0.04 − 0.037 = 0.003
b.
7-12
0.1
0.09
0.08
0.07
0.06
0.05 Centerline
0.04 UCL
0.03
0.02 LCL
0.01
0
1 2 3 4 5
Sample Number
0.04(1−0.04)
UCL = 0.04 + 3√ = 0.04 + 0.048 = 0.088
150
0.04(1−0.04)
LCL = 0.04 − 3√ 150
= 0.04 − 0.048 = −0.008, so LCL = 0.
c. The control limits have a larger spread with smaller sample sizes due to the
increased standard deviation for the smaller sample size.
42.
a.
0.16
0.14
0.12
0.1
0.08 Centerline
0.06 UCL
0.04
LCL
0.02
0
1 2 3 4 5 6
Sample Number
0.10(1−0.10))
UCL = 0.10 + 3√ 400
= 0.10 + 0.045 = 0.145
0.10(1−0.10))
LCL = 0.10 − 3√ 400
= 0.10 − 0.045 = 0.055
b.
7-13
0.2
0.15
Centerline
0.1
UCL
0.05 LCL
Sample Proportions
0
1 2 3 4 5 6
Sample Number
c. No, the production process seems to be out of control due to the 6th sample
proportion, which is above the upper control limit.
43.
a.
0.45
0.4
0.35
Centerline
0.3
UCL
0.25
LCL
0.2
1 2 3 4 5 6
Sample Number
0.34(1−0.34)
UCL = 0.34 + 3√ = 0.34 + 0.064 = 0.404
500
0.34(1−0.34)
LCL = 0.34 − 3√ 500
= 0.34 − 0.064 = 0.276
b.
7-14
0.45
0.4
Centerline
0.35
UCL
0.3 LCL
Sample Proportions
0.25
0.2
1 2 3 4 5 6
c. There are no points outside the control limits, so the process is under control.
However, the positive trend suggests that the process may become out of control if
the upward trend continues.
44.
a.
17
16
15
14 Centerline
13 UCL
12 LCL
11 Sample Mean
10
1 2 3 4 5 6
Sample Number
Centerline: µ = 14
2
UCL = 14 + 3 ( ) = 14 + 1.897 = 15.897
√10
2
LCL = 14 − 3 ( 10) = 14 − 1.897 = 12.103
√
b. All the sample means randomly lie within the control limits. Therefore, we can
conclude that the production process is in control and operating properly.
7-15
45.
a.
5.3
5.25
5.2
5.15
5.1 Centerline
5.05 UCL
5
LCL
4.95
4.9 Sample Mean
4.85
1 2 3 4 5 6
Sample Number
Centerline: µ = 5.125
UCL = 5.25
LCL = 5
b. There are no points outside the control limits. It appears that the process is under
control, but the positive trend suggests the process may become out of control if
the trend continues.
46.
a.
98
96
94
Centerline
92
UCL
90
LCL
88 Sample Mean
86
1 2 3 4
Over Number
Centerline: µ = 94
2.8
UCL = 94 + 3 ( 6 ) = 94 + 3(1.14) = 97.43
√
2.8
LCL = 94 − 3 ( 6) = 94 − 3(1.14) = 90.57
√
To plot the average speed, take the average of each over:
96.8+99.5+88.8+81.9+100.1+96.8
Over 1: x = 6
= 93.98
7-16
Similarly,
Over 2: x = 97.23
Over 3: x = 90.70
Over 4: x = 89.85
b. Kalwant’s average speed is out of the control limits on 1 out of 4 of his overs, which
rather justifies his coach’s concern that he is not very consistent.
47.
a.
0.07
0.06
Centerline
0.05
0.04
UCL
0.03
0.02 LCL
0.01
Sample
0 Proportions
1 2 3 4 5 6 7 8 9 10
Sample Number
0.04(1−0.04)
UCL = 0.34 + 3√ 500
= 0.04 + 0.026 = 0.066
0.04(1−0.04)
LCL = 0.34 − 3√ 500
= 0.04 − 0.026 = 0.014
b. All sample proportions are within the control limits and there is no apparent trend,
which suggests that the machine is operating properly.
48.
a.
0.08
0.07
0.06
0.05
0.04 Centerline
0.03 UCL
0.02
0.01 LCL
0
1 2 3 4
Sample Number
7-17
0.05(1−0.05)
UCL = 0.05 + 3√ 1,000
= 0.05 + 3(0.007) = 0.071
0.05(1−0.05)
LCL = 0.05 − 3√ 1,000
= 0.05 − 3(0.007) = 0.029
b. Since 0.062 is within the control limits, the process is in control.
49.
a.
0.31
0.29
0.27
0.25
Centerline
0.23
UCL
0.21
0.19 LCL
0.17
1 2 3 4 5
Sample Number
0.25(1−0.25)
UCL = 0.25 + 3√ = 0.25 + 0.047 = 0.297
750
0.25(1−0.25)
LCL = 0.25 − 3√ 750
= 0.25 − 0.047 = 0.203
b. 240/750 = 0.32, which is above the upper control limit of 0.297, so the university
should be concerned about this.
50.
a.
0.35
0.3
Centerline
0.25
0.2 UCL
0.15
0.1 LCL
0.05
0 Sample
1 2 3 4 5 6 Proportions
Month
7-18
0.15(1−0.15)
UCL = 0.15 + 3√ = 0.15 + 0.12 = 0.27
80
0.15(1−0.15)
LCL = 0.15 − 3√ = 0.15 − 0.12 = 0.03
80
b. Find the proportion of complaints each month:
Month Sample Proportion
1 20/80 = 0.25
2 12/80 = 0.15
3 24/80 = 0.30
4 14/80 = 0.175
5 25/80 = 0.3125
6 22/80 = 0.275
We plot each sample proportion on the control chart (shown above) to see that 3
out of 6 months were out of the control limits, which is a good justification for why
Dell chose to direct customers away from India call centers.
51.
a. There could be selection bias because people who go to the beach often go there to
walk or exercise and are more likely to follow a consistent walking regimen.
b. There could be nonresponse bias due to differences in people who mail back the
envelope.
c. There could be selection bias because some people may not have access to the
website, or may not be computer savvy. Also nonresponse bias because people
who choose not to respond may have different preferences than the people who do
choose to respond.
d. There could be selection bias because patients in hospitals are probably ill and may
not be capable of walking vigorously three times a week.
52.
a. As one example, use a random number table or a random number generator (in
Excel, for instance) to randomly select individuals into the sample from the list of
all residents of Miami. Then conduct the survey by contacting those selected.
b. To get a stratified random sample, you could create strata based on ethnicity; for
example under white, black, Hispanic, Asian, and then randomly select adults in
each group and ask whether or not they walk regularly.
c. To get a cluster sample, you could choose a number of representative
neighborhoods in Miami and randomly select adults within these neighborhoods
and ask whether or not they walk regularly.
53.
20−22
a. 𝑃(𝑋̅ > 20) = 𝑃 (𝑍 > 2/ 8 ) = 𝑃(𝑍 > −2.83) = 1 − 0.0023 = 0.9977.
√
15−16
b. 𝑃(𝑋̅ < 15) = 𝑃 (𝑍 < 2/√20
) = 𝑃(𝑍 < −2.24) = 0.0125.
7-19
54.
52,000−50,402
a. 𝑃(𝑋̅ > 52,000) = 𝑃 (𝑍 > 6,000/ 100 ) = 𝑃(𝑍 > 2.66) = 1 − 0.9961 = 0.0039.
√
52,000−49,703
b. 𝑃(𝑋̅ > 52,000) = 𝑃 (𝑍 > ) = 𝑃(𝑍 > 2.30) = 1 − 0.9893 = 0.0107.
10,000/√100
c. Even though finance graduates have a lower mean starting salary than accounting
graduates, there is greater standard deviation for salary for finance graduates.
Therefore, 100 finance graduates have a slightly higher probability (1.07%
compared to 0.39%) of earning an average salary of more than $52,000.
55.
79−80
a. 𝑃(𝑋 < 79) = 𝑃 (𝑍 < 2
) = 𝑃(𝑍 < −0.5) = 0.3085.
79−80
b. 𝑃(𝑋̅ < 79) = 𝑃 (𝑍 < ) = 𝑃(𝑍 < −1.58) = 0.0571.
2/√10
79−80
c. 𝑃(𝑋̅ < 79) = 𝑃 (𝑍 < ) = 𝑃(𝑍 < −2.74) = 0.0031.
2/√30
56.
a. The sampling distribution of the sample mean will be approximately normal for
any sample size because the population is normally distributed; in addition, 𝐸(𝑋̅) =
𝜎
𝜇; 𝑠𝑒(𝑋̅) = 𝑛.
√
16−16.01
b. 𝑃(𝑋̅ < 16) = 𝑃 (𝑍 < 0.08/ 20 ) = 𝑃(𝑍 < −0.56) = 0.2877.
√
48 16−16.01
c. 𝑥̅ = 3
= 16; 𝑃(𝑋̅ ≥ 16) = 𝑃 (𝑍 ≥ 0.08/ 3 ) = 𝑃(𝑍 ≥ −0.22) = 1 − 0.4129 =
√
0.5871.
57.
500−470.73
a. 𝑃(𝑋 > 500) = 𝑃 (𝑍 > 50
) = 𝑃(𝑍 > 0.59) = 1 − 0.7224 = 0.2776
500−470.73
b. 𝑃(𝑋̅ > 500) = 𝑃 (𝑍 > 50⁄ 4 ) = 𝑃(𝑍 > 1.17) = 1 − 0.8790 = 0.1210
√
c. Since the amounts spent on the lottery are independent, the probability of all four
randomly selected Georgia residents spending more than $500 on lottery is:
(𝑃(𝑋 > 500))4 = (0.2776)4 = 0.0059.
58.
200−210
a. 𝑃(𝑋̅ > 200) = 𝑃 (𝑍 > 48/ 100 ) = 𝑃(𝑍 > −2.08) = 1 − 0.0188 = 0.9812
√
600−608
b. 𝑃(𝑋̅ > 600) = 𝑃 (𝑍 > 132/√100
) = 𝑃(𝑍 > −0.61) = 1 − 0.2709 = 0.7291
59.
120 0.80−0.77
a. = 0.80; 𝑃(𝑃̅ > 0.80) = 𝑃 (𝑍 > ) = 𝑃(𝑍 > 0.87) = 1 − 0.8078 =
150 0.77(1−0.77)
√
150
0.1922.
7-20
0.70−0.77
b. 𝑃(𝑃̅ < 0.70) = 𝑃 (𝑍 < ) = 𝑃(𝑍 < −2.04) = 0.0207.
0.77(1−0.77)
√
150
60.
0.40−0.35
a. 𝑃(𝑃̅ > 0.40) = 𝑃 (𝑍 > ) = 𝑃(𝑍 > 1.05) = 1 − 0.8531 = 0.1469
0.35(1−0.35)
√
100
0.40−0.35
b. 𝑃(𝑃̅ > 0.40) = 𝑃 (𝑍 > ) = 𝑃(𝑍 > 1.48) = 1 − 0.9306 = 0.0694
0.35(1−0.35)
√
200
c. There is a smaller probability that more than 40 percent have pushed back their
retirement date with the larger sample size of 200 adults (0.0694 compared to
0.1469 with a sample size of 100). This is because with larger sample sizes, the
standard deviation of ̅𝑃 is reduced; 𝑃̅ is more likely to be closer to the population
proportion of 0.35, and therefore has less probability of being greater than 0.40.
61.
0.50−0.45
a. 𝑃(𝑃̅ > 0.50) = 𝑃 (𝑍 > ) = 𝑃(𝑍 > 0.71) = 1 − 0.7611 = 0.2389.
0.45(1−0.45)
√
50
0.50−0.48
b. 𝑃(𝑃̅ > 0.50) = 𝑃 (𝑍 > ) = 𝑃(𝑍 > 0.28) = 1 − 0.6103 = 0.3897.
0.48(1−0.48)
√
100
62.
a.
440
430
420
410 Centerline
400 UCL
390 LCL
380 Sample Mean
370
1 2 3 4 5 6
Sample Number
Centerline: µ = 410
25
UCL = 410 + 3 ( ) = 410 + 15 = 425
√25
25
LCL = 410 − 3 ( 25) = 410 − 15 = 395
√
7-21
b. Two of the sample means are above the upper control limit, indicating that the
advertised amount of sodium content is not accurate.
63.
a.
5.6
5.4
5.2
5 Centerline
4.8 UCL
4.6 LCL
4.4 Sample Mean
4.2
1 2 3 4 5
Week
Centerline: µ = 5
0.3
UCL =5 + 3 ( ) = 5 + 0.32 = 5.32
√8
0.3
LCL =5 − 3( ) = 5 − 0.32 = 4.68
√8
The five weekly sample means are:
Week 1 4.97
Week 2 5.16
Week 3 5.12
Week 4 5.35
Week 5 5.50
b. The last two points are outside the upper control limit, and there is a positive
trend, suggesting that the process is out of control and will continue getting out of
control and needs to be adjusted.
64.
0.15−0.10
a. 𝑃(𝑃̅ > 0.15) = 𝑃 (𝑍 > ) = 𝑃(𝑍 > 1.18) = 1 − 0.8810 = 0.1190.
0.10(1−0.10)
√
50
0.15−0.10
b. 𝑃(𝑃̅ > 0.15) = 𝑃 (𝑍 > ) = 𝑃(𝑍 > 1.67) = 1 − 0.9525 = 0.0475.
0.10(1−0.10)
√
100
7-22
65.
a.
0.25
0.2 Centerline
0.15
UCL
0.1
0.05 LCL
0
Sample
1 2 3 4 5 6 Proportions
Week
0.10(1−0.10)
UCL = 0.10 + 3√ 50
= 0.10 + 0.127 = 0.227
0.10(1−0.10)
LCL = 0.10 − 3√ 50
= 0.10 − 0.127 = −0.027, so LCL = 0
All of the sample proportions appear randomly within the control limits, and
therefore no adjustments are needed.
b.
0.25
0.2 Centerline
0.15
UCL
0.1
0.05 LCL
0
1 2 3 4 5 6 Sample
Proportions
Week
0.10(1−0.10)
UCL = 0.10 + 3√ 100
= 0.10 + 0.09 = 0.19
0.10(1−0.10)
LCL = 0.10 − 3√ 100
= 0.10 − 0.09 = 0.01
Week 5 is slightly above the upper control limit when the sample size is larger. In
week 6 the sample proportion is almost below the lower control limit. The firm may
want to inspect the machine.
7-23
Case Study 7.1
4.48−4.20
1. 𝑃(𝑋̅ ≥ 4.48) = 𝑃 (𝑍 ≥ 1.4/ 160 ) = 𝑃(𝑍 ≥ 2.53) = 1 − 0.9943 = 0.0057.
√
4.60 − 4.20
𝑃(𝑋̅ ≥ 4.60) = 𝑃 (𝑍 ≥ ) = 𝑃(𝑍 ≥ 1.81) = 1 − 0.9649 = 0.0351.
1.4/√40
2. The 1st sample with 𝑛 = 160 will be more representative of the U.S. as a whole due to the
smaller standard deviation.
Case Study 7.2
1. We are given that 8% of college graduates, under the age of 25, and 24.5% of high school
graduates are unemployed. Therefore, the expected number of college graduates who are
employed is 220(1 − 0.08) = 202 (approximately). Similarly, the expected number of high
school graduates who are employed is 140(1 − 0.245) = 106 (approximately).
200 0.91−0.92
2. 220
= 0.91; 𝑃(𝑃̅ ≥ 0.91) = 𝑃 (𝑍 ≥ ) = 𝑃(𝑍 ≥ −0.55) = 1 − 0.2912 = 0.7088.
0.92(1−0.92)
√
220
100 0.71−0.755
= 0.71; 𝑃(𝑃̅ ≥ 0.71) = 𝑃 (𝑍 ≥ ) = 𝑃(𝑍 ≥ −1.24) = 1 − 0.1075 = 0.8925.
140 0.755(1−0.755)
√
140
Case Study 7.3
1.
175
170
165
Centerline
160
UCL
155
LCL
150
Sample Mean
145
1 2 3 4 5 6 7 8
Sample Number
Centerline: µ = 163 grams

7.5
UCL = 163 + 3 ( ) = 163 + 7.12 = 170.12
√10
7.5
LCL = 163 − 3 ( 10) = 163 − 7.12 = 155.88
√
7-24
2. Sample Means are:

#1 163.1
2 163.73
3 158.4
4 166.52
5 166.37
6 170.75
7 172.54
8 172.28
The process does not appear to be in control; the last 3 sample means are above the upper
control limit, and there is a positive trend, suggesting that the process is going to continue
to get further out of control. The process should be adjusted.
7-25

Essentials of Business Statistics Communicating With Numbers 1st Edition Jaggia Solutions Manual

Uploaded by

Document Information

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

Essentials of Business Statistics Communicating With Numbers 1st Edition Jaggia Solutions Manual

Uploaded by

Copyright:

Available Formats

Essentials of Business Statistics

Communicating with Numbers 1st

Chapter 7. Sampling and Sampling

0.5 − 0.3300 = 0.1700.

0.9484 − 0.0516 = 0.8968.

0.6985 − 0.3015 = 0.3970

0.0192, so there is a 1.92% chance of getting dissolved.

b. 𝑝 = 0.26 is the proportion of French people who approve, therefore 1 − 0.26 =

Case Study 7.1

Case Study 7.2

Case Study 7.3

Centerline: µ = 163 grams

2. Sample Means are:

You might also like