Solutions Solutions: Simple Comparative Experiments Simple Comparative Experiments
Solutions Solutions: Simple Comparative Experiments Simple Comparative Experiments
Solutions Solutions: Simple Comparative Experiments Simple Comparative Experiments
http://www.book4me.xyz/solutions-manual-design-and-analysis-of-experiments-montgomery/
Chapter 2
Simple Comparative Experiments
Solutions
2.1. Computer output for a random sample of data is shown below. Some of the quantities are missing.
Compute the values of the missing quantities.
2.2. Computer output for a random sample of data is shown below. Some of the quantities are missing.
Compute the values of the missing quantities.
Y 16 ? 0.159 ? 399.851
2.3. Suppose that we are testing H 0 : µ = µ 0 versus H 1 : µ ≠ µ 0 . Calculate the P-value for the following
observed values of the test statistic:
2.4. Suppose that we are testing H 0 : µ = µ 0 versus H 1 : µ > µ 0 . Calculate the P-value for the following
observed values of the test statistic:
2-1
2.5. Consider the computer output shown below.
One-Sample Z
Test of mu = 30 vs not = 30
(a) Fill in the missing values in the output. What conclusion would you draw?
Two-sided.
(c) Use the output and the normal table to find a 99 percent CI on
on the mean.
mean.
CI = 30.42725, 31.97275
(d) What is the P-value if the alternative hypothesis is H 1 : µ > 30
P-value = 0.00003
2.6. Suppose that we are testing H 0 : µ 1 = µ 2 versus H 1 : µ 1 = µ 2 with a sample size of n 1 = n 2 = 12.
Both sample variances are unknown but assumed
assumed equal. Find bounds on the P-value for the following
observed values of the test statistic:
(a) t 0 = 2.30 Table P-value = 0.02, 0.05 Computer P-value = 0.0313
(b) t 0 = 3.41 Table P-value = 0.002, 0.005 Computer P-value = 0.0025
(c) t 0 = 1.95 Table P-value = 0.1, 0.05 Computer P-value = 0.0640
(d) t 0 = -2.45 Table P-value = 0.05, 0.02 Computer P-value = 0.0227
Note that the degrees of freedom is (12 +12) – 2 = 22. This is a two-sided test
2.7. Suppose that we are testing H 0 : µ 1 = µ 2 versus H 1 : µ 1 > µ 2 with a sample size of n 1 = n 2 = 10.
Both sample variances are unknown but assumed
assumed equal. Find bounds on the P-value for the following
observed values of the test statistic:
(a) t 0 = 2.31 Table P-value = 0.01, 0.025 Computer P-value = 0.01648
(b) t 0 = 3.60 Table P-value = 0.001, 0.0005 Computer P-value = 0.00102
(c) t 0 = 1.95 Table P-value = 0.05, 0.025 Computer P-value = 0.03346
2-2
(d) t 0 = 2.19 Table P-value = 0.01, 0.025 Computer P-value = 0.02097
Note that the degrees of freedom is (10 +10) – 2 = 18. This is a one-sided test.
Minitab Output
Summ
Summary for Sample Data
A nderson-Darling N ormality Test
A -Squared 0.33
P -V al
alue 0.435
M ean 10.952
S tD ev 1.993
V ariance 3.974
Skewness
Skewness -0.45
0.4513131
1
K ur
urto si
si s -1 .0
.0 67
674 6
N 10
M in
inim um 7. 750
1 st
st Q u ar
art ilil e 9 .0
.0 80
80
M ed
edian 11.345
3rd
3rd Quartil
Quartile e 13.06
13.067 7
8 9 10 11 12 13
M ax
axim um
um 13.210
95% C onfidence
onfidence Interv
Interv al for Mean
9. 526 12. 378
95% C onfidence
onfidence Interv
Interv al for Median
8. 973 13. 078
95% C onfidence
onfidence Interv
Interv al for StD ev
95 % Confidence
Confidence Intervals
1. 371 3. 639
Mean
Median
9 10 11 12 13
According to the output, the Anderson-Darling Normality Test has a P-Value of 0.435. The data can be
considered normal. The 95% confidence
confidence interval on the mean is (9.526,12.378). This confidence interval
contains 10, therefore there is evidence that the population mean is 10.
2.9. A computer program has produced the following output for the hypothesis testing problem:
2-3
y1 − y 2 2.35
t 0 = = = 2.01
1 1 StdError
S p +
n1 n2
StdError = 2.35 / 2.01 = 1.169
(b) Is this a two-sided or one-sided test? One-sided test for a t 0 = 2.01 is a P-value of 0.0298.
(c) If α =0.05,
=0.05, what are your conclusions? Reject the null hypothesis
hypothesis and conclude
conclude that there is a
difference in the two samples.
(d) Find a 90% two-sided CI on the difference in the means.
1 1 1 1
y1 − y2 − tα 2, n1 + n2 − 2 S p + ≤ µ1 − µ 1 ≤ y1 − y2 + t α 2 , n1 + n2 − 2
S p +
n1 n2 n1 n2
1 1 1 1
y1 − y2 − t0.05,18 S p + ≤ µ1 − µ 1 ≤ y1 − y2 + t0.05,18 S p +
n1 n2 n1 n2
2.35 − 1.734 (1.169) ≤ µ1 − µ 1 ≤ 2.35 + 1.734( 1.169)
0.323 ≤ µ1 − µ 1 ≤ 4.377
2.10. A computer program has produced the following output for the hypothesis testing problem:
(b) Is this a two-sided or one-sided test? Two-sided test for a t 0 = -1.88 is a P-value of 0.0723.
(c) If α =
=0.05,
0.05, what are your conclusions?
conclusions? Accept the null hypothesis, there is no difference in the
means.
(d) Find a 90% two-sided CI on the difference in the means.
2-4
1 1 1 1
y1 − y2 − tα 2, n1 + n2 − 2 S p + ≤ µ1 − µ 1 ≤ y1 − y2 + t α 2 , n1 + n2 − 2
S p +
n1 n2 n1 n2
1 1 1 1
y1 − y2 − t0.05,24 S p + ≤ µ1 − µ 1 ≤ y1 − y2 + t0.05,24 S p +
n1 n2 n1 n2
−11.5 −1.711( 6.12) ≤ µ1 − µ 1 ≤ −11.5 + 1.711( 6.12)
−21.97 ≤ µ1 − µ 1 ≤ −1.03
(a) t 0 = 2.35 Table P-value = 0.01, 0.025 Computer P-value = 0.01698
(b) t 0 = 3.55 Table P-value = 0.001, 0.0025 Computer P-value = 0.00160
(c) t 0 = 2.00 Table P-value = 0.025, 0.005 Computer P-value = 0.03264
(d) t 0 = 1.55 Table P-value = 0.05, 0.10 Computer P-value = 0.07172
(a) t 0 = 2.48 Table P-value = 0.02, 0.05 Computer P-value = 0.03499
(b) t 0 = -3.95 Table P-value = 0.002, 0.005 Computer P-value = 0.00335
(c) t 0 = 2.69 Table P-value = 0.02, 0.05 Computer P-value = 0.02480
(d) t 0 = 1.88 Table P-value = 0.05, 0.10 Computer P-value = 0.09281
(e) t 0 = -1.25 Table P-value = 0.20, 0.50 Computer P-value = 0.24282
One-Sample T: Y
(a) Fill in the missing values in the output. Can the null hypothesis be rejected at the 0.05 level? Why?
2-5
Two-sided.
(c) If the hypothesis had been H 0 : µ = 90 versus H 1 : µ ≠ 90 would you reject the null hypothesis at the
0.05 level?
Yes.
CI = 91.2735, 93.8875
(e) What is the P-value if the alternative hypothesis is H 1 : µ > 91?
P-value = 0.001.
One-Sample T: Y
Test of mu = 25 vs > 25
(N-1) = (12 – 1) = 11
Two-sample T for Y1 vs Y2
P-Value = 0.01
0.01 DF = 38
(a) Can the null hypothesis be rejected at the 0.05 level? Why?
2-6
Yes, the P-Value of 0.001 is much less than 0.05.
Two-sided.
(c) If the hypothesis had been H 0 : µ 1 - µ 2 = 2 versus H 1 : µ 1 - µ 2 ≠ 2 would you reject the null
hypothesis at the 0.05 level?
Yes.
(d) If the hypothesis had been H 0 : µ 1 - µ 2 = 2 versus H 1 : µ 1 - µ 2 < 2 would you reject the null
hypothesis at the 0.05 level? Can you answer this question question without doing any additional
calculations? Why?
Yes, no additional calculations are required because the test is naturally becoming more significant
with the change from -2.33341 to -4.33341.
(e) Use the output and the t table to find a 95 percent upper confidence bound on the difference in
means?
(f) What is the P-value if the alternative hypotheses are H 0 : µ 1 - µ 2 = 2 versus H 1 : µ 1 - µ 2 ≠ 2?
P-value = 1.4E-07.
H 0: µ =
= 150 H 1: µ >
> 150
y − µ o 148.75 − 15
150 −1.25
zo = = = = −0.8333
σ 3 3
n 4 2
2-7
The 95% confidence interval is
σ σ
y − z α ≤ µ ≤ y + z α
2 2
n n
148.75 − 1.96 3 2 ( )( ) ≤ µ ≤ 148.75 + (1.96)(3 2)
145. 81 ≤ µ ≤ 151. 69
2.17. The viscosity of a liquid detergent is supposed to average 800 centistokes at 25 °C. A random
sample of 16 batches of detergent is collected, and the average viscosity is 812. Suppose we know that the
standard deviation of viscosity is σ = 25 centistokes.
2.18. The diameters of steel shafts produced by a certain manufacturing process should have a mean
mean
diameter of 0.255 inches. The diameter is known to have a standard
standard deviation of σ = 0.0001 inch. A
random sample of 10 shafts has an average diameter of 0.2545 inches.
2-8
y − µ o 0.2545 − 0.255
0.2545 0.255
zo = = = −15.81
σ 0.0001
n 10
Since z0.025 = 1.96, reject H 0.
n 2
n
(
0.2545 − 1.96 ) 0.0001 ≤ µ ≤ 0.2545 + (1.96) 0.0001
10 10
0. 254438 ≤ µ ≤ 0. 254562
2.19. A normally distributed random variable has an unknown mean µ and a known variance = 9. Find σ 2
the sample size required to construct a 95 percent confidence interval on the mean that has total length of
1.0.
Since y ∼ N(µ ,9),
,9), a 95% two-sided confidence interval on µ is
is
If the total interval is to have width 1.0, then the half-interval is 0.5. Since zα /2
/2 = z0.025 = 1.96,
(1.96)(3 n )= 0.5
n = (1.96 )(3 0.5) = 11.76
2.20. The shelf life of a carbonated beverage is of interest. Ten bottles are randomly selected and
and tested,
and the following results are obtained:
Days
108 138
124 163
124 159
106 134
115 139
2-9
y = 131
S 2 = 3438 / 9 = 382
S =
= 382
382 = 19.5
19.54
4
Minitab Output
T-Test of the Mean
Test of mu = 120.
120. 00 vs mu > 120.
120. 00
Var
Var i abl e N Mean St Dev SE Mean T P
Shel
Shel f Li f e 10 131.
131. 00 19.
19. 54 6. 18 1. 78 0. 054
054
T Confidence Intervals
(c) Find the P-value for the test in part (b). P=0.054
(
131 − 3.250 ) 19.54 ≤ µ ≤ 131 + (3.250 ) 19.54
10 10
110.91 ≤ µ ≤ 151.08
A normal probability plot, obtained from Minitab, is shown. There is no reason to doubt the adequacy of
the normality assumption. If shelf life is not normally
normally distributed, then the impact of this on the t -test
-test in
problem 2.20 is not too serious unless the departure from nor mality is severe.
http://www.book4me.xyz/solutions-manual-design-and-analysis-of-experiments-montgomery/
2-10
Normal Probability Plot
.999
.99
.95
y
t
i
l .80
i
b
a .50
b
o
r .20
P
.05
.01
.001
2.22. The time to repair an electronic instrument is a normally distributed random variable measured
measured in
hours. The repair time for 16 such instruments chosen at random are
are as follows:
Hours
159 280 101 212
224 379 179 264
222 362 168 250
149 260 485 170
(a) You wish to know if the mean repair time exceeds 225 hours. Set up appropriate
appropriate hypotheses for
investigating this issue.
H 0: µ =
= 225 H 1: µ > 225
y = 241.50
S 2 =146202 / (16 - 1) = 9746.80
S =
= 9746.8 = 98.7
9746.8 98.73
3
Minitab Output
T-Test of the Mean
Test of mu = 225.
225. 0 vs mu > 225.
225. 0
Var
Var i abl e N Mean St Dev SE Mean T P
Hour
our s 16 241.
241. 5 98.
98. 7 24.
24. 7 0. 67 0. 26
2-11
T Confidence Intervals
S S
The 95% confidence interval is y − t α
2
,n −1
≤ µ ≤ y + t α
2
,n −1
n n
(
241.50 − 2.131 ) 98.73 ≤ µ ≤ 241.50 + (2.131) 98.73
16 16
188.9 ≤ µ ≤ 294.1
2.23. Reconsider the repair time data in Problem 2.22. Can repair time, in your opinion, be adequately
adequately
modeled by a normal distribution?
The normal probability plot below does not reveal any serious problem with the normality assumption.
.999
.99
.95
y
t
i
l .80
i
b
a .50
b
o
r .20
P
.05
.01
.001
2.24. Two machines are used for filling plastic bottles with a net volume of 16.0 ounces.ounces. The filling
deviat ion of σ 1 = 0.015 and σ 2 = 0.018. The quality
processes can be assumed to be normal, with standard deviation quality
engineering department suspects that both machines fill to the same net volume, whether or not this volume
is 16.0 ounces. An experiment is performed by taking a random sample from the output
output of each machine.
Machine 1 Machine 2
16.03 16.01 16.02 16.03
16.04 15.96 15.97 16.04
16.05 15.98 15.96 16.02
16.05 16.02 16.01 16.01
16.02 15.99 15.99 16.00
2-12
(a) State the hypotheses
hypotheses that should
should be tested in this experiment.
n1 = 10 n2 = 10
(d) Find a 95 percent confidence interval on the difference in the mean fill volume for the two machines.
2.25. Two types of plastic are suitable for use by an electronic calculator manufacturer. The breaking
strength of this plastic is important. It is known that σ 1 = σ 2 = 1.0 psi. From random samples of n1 = 10
and n2 = 12 we obtain y 1 = 162.5 and y 2 = 155.0. The company will not adopt plastic 1 unless unless its
breaking strength exceeds that of plastic 2 by at least 10 psi. Based on the sample infor mation, should they
use plastic 1? In answering this questions, set up and and test appropriate hypotheses using α = 0.01.
Construct a 99 percent confidence interval on the true mean difference in breaking strength.
n1 = 10 n2 = 10
2-13
The 99 percent confidence interval is
2 2 2 2
σ 1 σ 2 σ 1 σ 2
y1 − y 2 − z α + ≤ µ 1 − µ 2 ≤ y1 − y 2 + z α +
2
n1 n2 2
n1 n2
12 12 12 12
(162.5 − 155.0) − (2.575) + ≤ µ 1 − µ 2 ≤ (162.5 − 155.0) + (2.575) +
10 12 10 12
2.26. The following are the burning times (in minutes) of chemical flares of two different formulations.
The design engineers are interested in both the means and variance of the burning times.
Type 1 Type 2
65 82 64 56
81 67 71 69
57 59 83 74
66 75 59 82
82 70 65 79
(a) Test the hypotheses that the two variances are equal. Use α =
= 0.05.
H 0 : σ 1 = σ 22
2
H 1 : σ 1 ≠ σ 2
2 2
Do not reject.
(b) Using the results of (a), test the hypotheses that the mean burning times are equal. Use α =
= 0.05.
What is the P-value for this test?
Do not reject.
Minitab Output
Two Sample T-Test
T-Test and Confidence Interval
95%
95% CI f or mu Type
Type 1 - mu Type
Type 2: ( - 8. 6, 9. 0)
T- Test mu Type 1 = mu Type 2 ( vs not =) : T = 0. 05 P = 0. 96 DF = 18
Bot h use
use Pool
Pool ed StDev
StDev = 9. 32
2-14
(c) Discuss the role of the normality assumption in this problem. Check the assumption of normality
normality for
both types of flares.
.999
.99
.95
y
t
i
l .80
i
b
a .50
b
o
r .20
P
.05
.01
.001
60 70 80
Type 1
Average: 70.4
70.4 Anderson-Darling
Anderson-Darling Normality Test
StDev: 9.26403 A-Squared: 0.344
0.344
N: 10 P-Value: 0.409
.999
.99
.95
y
t
i
l .80
i
b
a .50
b
o
r .20
P
.05
.01
.001
60 70 80
Type 2
Average: 70.2
70.2 Anderson-Darling
Anderson-Darling Normality Test
StDev: 9.36661 A-Squared: 0.186
0.186
N: 10 P-Value: 0.876
2.27. An article in Solid State Technology, "Orthogonal Design of Process Optimization and Its
Application to Plasma Etching" by G.Z. Yin and D.W. Jillie (May, 1987) describes an experiment to
determine the effect of C 2 F 6 flow rate on the uniformity of the etch on a silicon wafer used in integrated
circuit manufacturing. Data for two flow rates are as follows:
2-15
(a) Does the C 2 F 6 flow rate affect average etch
etch uniformity? Use α =
= 0.05.
Minitab Output
Two Sample T-Test
T-Test and Confidence Interval
(b) What is the P-value for the test in part (a)? From the Minitab output, P=0.21
(c) Does the C 2 F 6 flow rate affect the wafer-to-wafer variability in etch uniformity? Use α = 0.05.
H 0 : σ 1
2
= σ 22
H 1 : σ 1 ≠ σ 2
2 2
F 0.025,5,5 = 7.15
F 0.975,5,5 = 0.14
0.5776
F 0 = = 0.86
0.6724
The box plots shown below indicate that there is little difference in uniformity at the two gas flow rates.
Any observed difference is not statistically significant. See the t -test
-test in part (a).
y
t
i 4
m
r
o
f
i
n
U
125 200
Flow Rate
2-16
2.28. A new filtering device is installed in a chemical unit. Before its installation, a random sample
2
yielded the following information about the percentage of impurity: y 1 = 12.5, S 1 =101.17, and n 1 = 8.
2
After installation, a random sample yielded y 2 = 10.2, S 2 = 94.73, n 2 = 9.
S 12 101.17
F 0 = = = 1.07
S 22 94.73
Do not reject. Assume that the variances
variances are equal.
(b) Has the filtering device reduced the percentage of impurity significantly? Use α =
= 0.05.
H 0 : µ1 = µ 2
H 1 : µ1 > µ 2
n1 + n2 − 2 8+9− 2
S p = 9.89
y1 − y2 12.5 − 10.2
12.5 10.2
t 0 = = = 0.479
1 1 1 1
S p + 9.89 +
n1 n2 8 9
t 0.05,15 = 1.753
95 ºC 100 ºC
11.176 5.623
7.089 6.748
8.097 7.461
11.739 7.015
11.291 8.133
10.759 7.418
6.467 3.772
8.315 8.963
http://www.book4me.xyz/solutions-manual-design-and-analysis-of-experiments-montgomery/
2-17
(a) Is there evidence to support the claim that the higher baking temperature results in wafers with a lower
mean photoresist thickness? Use α = = 0.05.
H 0 : µ1 = µ 2
H 1 : µ1 > µ 2
(n − 1) S12 + (n2 − 1)S 22 (8 − 1)(4
1)(4.41) + (8 − 1)(2
.41 1)(2.5
.54)
4)
S p = 1 = = 3.48
2
n1 + n2 − 2 8+ 8− 2
S p = 1.86
y1 − y2 9.37 − 6.89
9.37 6.89
t 0 = = = 2.65
1 1 1 1
S p + 1.86 +
n1 n2 8 8
t 0.05,14 = 1.761
Since t 0.05,14
0.05,14 = 1.761, reject H 0 . There appears to be a lower mean thickness at the higher temperature.
This is also seen in the computer output.
Minitab Output
Two-Sample
Two-Sample T-Test
T-Test and CI: Thickness, Temp
Di f f er ence = mu Thi
Thi ck@95 - mu Thi ck@100
Est i mat e f or di f f er ence:
ce: 2. 475
95% l ower boun
boundd f or di f f erence:
erence: 0. 833
833
T- Test of di f f er ence = 0 ( vs >) : T- Val ue = 2. 65 P- Val ue = 0. 009 DF = 14
Bot h use
use Pool
Pool ed StDev
StDev = 1. 86
(b) What is the P-value for the test conducted in part (a)? P = 0.009
(c) Find a 95% confidence interval on the difference in means. Provide a practical interpretation of this
interval.
From the computer output the 95% lower confidence bound is 0.833 ≤ µ1 − µ 2 . This lower confidence
confidence
bound is greater than 0; therefore, there is a difference
di fference in the two temperatures on the thickness of the
photoresist.
2-18
(d) Draw dot diagrams to assist in interpreting the results from this experiment.
Temp
95
3.6 4.8 6.0 7.2 8.4 9.6 10.8 12.0 100
Thickness
.999
.99
.95
y
t
i
l .80
i
b
a .50
b
o
r .20
P
.05
.01
.001
7 8 9 10 11 12
Thick@95
Average: 9.36662
9.36662 Anderson-Darling
Anderson-Darling Normality Test
StDev: 2.09956 A-Squared: 0.483
0.483
N: 8 P-Value: 0.161
2-19
Normal Probability Plot
.999
.99
.95
y
t
i
l .80
i
b
a .50
b
o
r .20
P
.05
.01
.001
4 5 6 7 8 9
Thick@100
Average: 6.89163
6.89163 Anderson-Darling
Anderson-Darling Normality Test
StDev: 1.59509 A-Squared: 0.316
0.316
N: 8 P-Value: 0.457
2- Sam
Sampl e t Test
Test
Sampl e
Di f f erence
erence Si ze Pow
Power
2. 5 8 0. 7056
7056
(g) What sample size would be necessary to detect an actual difference in means of 1.5 kÅ with a power
power of
at least 0.9?.
Minitab Output
Power and Sample Size
2- Sam
Sampl e t Test
Test
Sam
Sampl e Target Act ual
ual
Di f f erence Si ze Pow
Power Pow
Power
1. 5 34 0. 9000
9000 0. 9060
9060
This result makes intuitive sense. More samples are needed to detect a smaller
smaller difference.
2-20
two cool-down times, 10 seconds and 20 seconds, and 20 housings were evaluated at each level of cool-
down time. All 40 observations in this experiment were run in random order. The data are shown below.
10 Seconds 20 Seconds
1 3 7 6
2 6 8 9
1 5 5 5
3 3 9 7
5 2 5 4
1 1 8 6
5 6 6 8
2 8 4 5
3 2 6 8
5 3 7 7
(a) Is there evidence to support the claim that the longer cool-down time results in fewer appearance
defects? Use α =
= 0.05.
From the analysis shown below, there is evidence that the longer cool-down time results in fewer
appearance defects.
Minitab Output
Two-Sample T-Test and CI: 10 seconds, 20 seconds
Di f f er ence = mu 10 10 seconds
seconds - mu 2020 seconds
seconds
Est i mat e f or di
di f f er ence:
ce: - 3. 150
95% uppupper boun
boundd f or di f f erence:
erence: - 2. 196
T- Test of di f f er ence = 0 ( vs <) : T- Val ue = - 5. 57 P- Val ue = 0. 000 DF = 38
Bot h use
use Pool
Pool ed StDev
StDev = 1. 79
(b) What is the P-value for the test conducted in part (a)? From the Minitab output, P = 0.000
(c) Find a 95% confidence interval on the difference in means. Provide a practical interpretation of this
interval.
From the Minitab output, µ1 − µ 2 ≤ − 2.196 . This lower confidence bound is less than 0. The two samples
samples
are different. The 20 second cooling time gives a cosmetically better housing.
http://www.book4me.xyz/solutions-manual-design-and-analysis-of-experiments-montgomery/
2-21
Dotplot of Ranking vs C4
C4
10 sec
20 sec
2 4 6 8
Ranking
(e) Check the assumption of normality for the data from this experiment.
.999
.99
.95
y
t
i
l .80
i
b
a .50
b
o
r .20
P
.05
.01
.001
1 2 3 4 5 6 7 8
10 seconds
Average: 3.35 Anderson-Darling
Anderson-Darling Normality Test
StDev: 2.00722 A-Squared: 0.748
0.748
N: 20 P-Value: 0.043
.999
.99
.95
y
t
i
l .80
i
b
a .50
b
o
r .20
P
.05
.01
.001
4 5 6 7 8 9
20 seconds
Average: 6.5 Anderson-Darling
Anderson-Darling Normality Test
StDev: 1.53897 A-Squared: 0.457
0.457
N: 20 P-Value: 0.239
2-22
There are no significant departures from normality.
2.31. Twenty observations on etch uniformity on silicon wafers are taken during a qualification
experiment for a plasma etcher. The data are as follows:
Etch Uniformity
5.34 6.65 4.76 5.98 7.25
6.00 7.55 5.54 5.62 6.21
5.97 7.35 5.44 4.39 4.98
5.25 6.35 4.61 6.00 5.32
(n − 1)S 2
≤ σ 2
≤
(n −1)S 2
2 2
χα χ
, n −1 (1−α ),n −1
2 2
(20 − 1)(0.88907 ) 2
≤ σ 2
≤
(20 −1)(0.88907 )2
32.852 8.907
0.457 ≤ σ 2
≤ 1.686
(c) Discuss the normality assumption and its role in this problem.
The normality assumption is much more important when analyzing variances then when analyzing means.
A moderate departure from normality could cause problems with both statistical tests and confidence
intervals. Specifically, it will cause the reported significance
significance levels to be incorrect.
(d) Check normality by constructing a normal probability plot. What are your conclusions?
conclusions?
The normal probability plot indicates that there is not a serious problem with the normality assumption.
2-23
Normal Probability Plot
.999
.99
.95
y
t
i
l .80
i
b
a .50
b
o
r .20
P
.05
.01
.001
2.32. The diameter of a ball bearing was measured by 12 inspectors, each using two different kinds of
calipers. The results were:
(a) Is there a significant difference between the means of the population of measurements represented by
the two samples?
samples? Use α = = 0.05.
Minitab Output
Paired T-Test
T-Test and Confi dence Interval
Pai
Pai r ed T f or Ca
Cal i per 1 - Cal i per 2
95% CI f or mean
ean di
di f f erence
erence:: ( - 0. 00102
1024, 0. 001
001524)
524)
T- Test of mean di f f er ence = 0 ( vs not = 0) : T- Val ue = 0. 43 P- Val ue = 0. 674
674
2-24
(b) Find the P-value for the test in part (a). P=0.674
(c) Construct a 95 percent confidence interval on the difference in the mean diameter measurements for
the two types of calipers.
Sd S d
d −t
α , n −1
≤ µ D (= µ1 − µ 2 ) ≤ d + t
α , n −1
2 n 2 n
0.002 0.002
0.00025 − 2.201 ≤ µ d ≤ 0.00025 + 2.201
12 12
−0.00102 ≤ µ d ≤ 0.00152
2-25
2.33. An article in the journal of Neurology (1998, Vol. 50, pp.1246-1252) observed that the monozygotic
twins share numerous physical, psychological
psychological and pathological
pathological traits. The investigators measured
measured an
intelligence score of 10 pairs of twins. The data are obtained as follows:
(a) Is the assumption that the difference in score is normally distributed reasonable?
Minitab Output
Summ
Summary for Difference
Dif ference
A nderson-Darling N ormality Test
A -Squared 0.19
P -V al
alue 0.860
M ea
ean -0.051000
S tD
tD ev
ev 0.440919
V ariance 0.194410
Skewness
Skewness -0.18
-0.1829
2965
65
Kurto
Kurtosi
siss -0.8
-0.817
1739
391
1
N 10
Minim
Minimum um -0.7
-0.7707000
0000
1st Q uartil
uartile
e -0.4625
-0.462500 00
M e di
di an
an -0 .0
.0 10
10 00
00 0
3rd Q uartile
uartile 0.36750
0.367500 0
-0.75 -0.50 -0.25 0.00 0.25 0.50
Maximu
Maximum m 0.60
0.60000000
00
95% C onfidence
onfidence Interval for Mean
-0.36
-0.3664
6415
15 0.264
0.26441
415
5
95% C onfidence
onfidence Interval for Median
-0.47
-0.4745
4505
05 0.373
0.37396
964
4
95% C onfidence
onfidence Interval for StDev
95 % C onfidence
onfidence Intervals
0.30
0.3032
3280
80 0.80
0.8049
4947
47
Mean
Median
By plotting the differences, the output shows that the Anderson-Darling Normality Test shows a
P-Value of 0.860. The data is assumed
assumed to be normal.
(b) Find a 95% confidence interval on the difference in the mean score. Is there any evidence that
mean score depends on birth order?
The 95% confidence interval on the difference in mean score is (-0.366415, 0.264415) contains
the value of zero. There is no difference in birth order.
2-26
(c) Test an appropriate set of hypothesis indicating that the mean score does not depend on birth
order.
Minitab Output
Pai
Pai r ed T f or Bi r t h Or der : 1 - Bi r t h Or der : 2
N Mean
ea n St Dev SE Mean
Bi r t h Or der
der : 1 10 6. 967
967 0. 810 0. 256
Bi r t h Or der
der : 2 10 7. 018
018 1. 053 0. 333
Di f f erence
erence 10 - 0. 051 0. 441 0. 139
95% CI f or mean di
di f f er ence:
ce: ( - 0. 366, 0. 264)
T- Test of mean di f f er ence = 0 ( vs not = 0) : T- Val ue = - 0. 37 P- Val ue = 0. 723
723
2.34. An article in the Journal of Strain Analysis (vol.18, no. 2, 1983) compares several procedures for
predicting the shear strength for steel plate girders. Data for nine girders in the form of the ratio of
predicted to observed load for two of these procedures, the Karl sruhe and Lehigh methods, are as follows:
n
1 1
d =
n
∑ d = (2.46
i =19
5 ) = 0.27
2.465 i0.274
4
1
n 2 1 n 2 2
1
∑ di − ∑ d i 0.82
0.8211
1151
51−
1
(2.4
(2.465
65)
)2 2
n i =1
sd = 9
i =1
= = 0.135
n −1 9 − 1
2-27
d 0.274
t 0 = = = 6.08
S d 0.135
n 9
tα
2,n −1 = t 0.025,8
= 2.306 , reject the null hypothesis.
Minitab Output
Paired T-Test and Confidence Interval
P=0.0002
(c) Construct a 95 percent confidence interval for the difference in mean predicted to observed
observed load.
S d S d
d − t α ≤ µ d ≤ d + t α
,n −1 ,n −1
2 n 2 n
0.135 0.135
0.274 − 2.306 ≤ µ d ≤ 0.274 + 2.306
9 9
0.17023 ≤ µ d ≤ 0.37777
The normal probability plots of the observations for each method follow. There are no serious concerns
with the normality assumption, but there is an indication of a possible outlier (1.178) in the Lehigh method
data.
.999
.99
.95
y
t .80
i
l
i
b .50
a
b
o
r .20
P
.05
.01
.001
Karlsruhe
Av erage: 1. 34011 Anderson-Darling Normalit y Test
StDev : 0.146031
0.146031 A-Squared: 0. 286
N: 9 P-Value: 0.537
2-28
Normal Probability Plot
.999
.99
.95
y
t .80
i
l
i
b .50
a
b
o .20
r
P
.05
.01
.001
Lehigh
Av erage: 1.06622 Anderson-Darling N ormality Test
StDev: 0.0493806
0.0493806 A-Squared: 0.772
N: 9 P-Value: 0.028
(e) Investigate the normality assumption for the difference in ratios for the two methods.
.999
.99
.95
y
t .80
i
l
i
b .50
a
b
o
r .20
P
.05
.01
.001
Difference
Av erage: 0. 273889 Anderson-Darling Normalit y Test
StDev : 0.135099
0.135099 A-Squared: 0. 318
N: 9 P-Value: 0.464
There is no issue with normality in the difference of ratios of the two methods.
(f) Discuss the role of the normality assumption in the paired t -test.
-test.
As in any t -test,
-test, the assumption of normality is of only moderate importance. In the paired t -test,
-test, the
assumption of normality applies to the distribution of the differences. That is, the individual sample
measurements do not have to be normally distributed, only their difference.
2.35. The deflection temperature under load for two different formulations of ABS plastic pipe is being
studied. Two samples of 12 observations each
each are prepared using
using each formulation, and the deflection
temperatures (in °F) are reported below:
Formulation 1 Formulation 2
206 193 192 177 176 198
188 207 210 197 185 188
205 185 194 206 200 189
187 189 178 201 197 203
2-29
(a) Construct normal probability plots for both samples. Do these plots support
support assumptions of normality
and equal variance for both samples?
.999
.99
.95
y
t
i
l .80
i
b
a .50
b
o
r .20
P
.05
.01
.001
.999
.99
.95
y
t .80
i
l
i
b .50
a
b
o
r .20
P
.05
.01
.001
(b) Do the data support the claim that the mean deflection temperature under load for formulation 1
exceeds that
that of formulation 2? Use α = 0.05.
Minitab Output
Two Sample T-Test
T-Test and Confidence Interval
Di f f erence
erence = mu ForFor m 1 - mu ForFor m 2
Est i mat e f or di f f er ence:
ce: 1. 42
95% l ower boubound f or di f f er ence
ence:: - 5. 64
T- Test of di f f er ence = 0 ( vs >) : T- Val ue = 0. 34 P- Val ue = 0. 367 DF = 22
Bot h use
use Pool
Pool ed StDev
StDev = 10.10. 1
2-30