Anda di halaman 1dari 38

Chapter 13

Analysis of Variance and

Experimental Design
Learning Objectives
1.

Understand how the analysis of variance procedure can be used to determine if the means of more
than two populations are equal.

2.

Know the assumptions necessary to use the analysis of variance procedure.

3.

Understand the use of the F distribution in performing the analysis of variance procedure.

4.

Know how to set up an ANOVA table and interpret the entries in the table.

5.

Be able to use output from computer software packages to solve analysis of variance problems.

6.

Know how to use Fishers least significant difference (LSD) procedure and Fishers LSD with the
Bonferroni adjustment to conduct statistical comparisons between pairs of populations means.

7.

Understand the difference between a completely randomized design, a randomized block design, and
factorial experiments.

8.

Know the definition of the following terms:

comparisonwise Type I error rate
experimentwise Type I error rate
factor
level
treatment
replication

partitioning
blocking
main effect
interaction

13 - 1

Chapter 13

Solutions:
1.

a.

x = (30 + 45 + 36)/3 = 37
k

SSTR = n j x j x
j =1

k

b.

j =1

c.

F = MSTR /MSE = 285/5.5 = 51.82

Using F table (2 degrees of freedom numerator and 12 denominator), p-value is less than .01
Actual p-value = .0000
Because p-value = .05, we reject the null hypothesis that the means of the three populations are
equal.

d.
Source of Variation
Treatments
Error
Total
2.

a.

Sum of Squares
570
66
636

Degrees of Freedom
2
12
14

Mean Square
285
5.5

F
51.82

k

SSTR = n j x j x
j =1

k

b.

j =1

c.

F = MSTR /MSE = 268/92 = 2.91

Using F table (2 degrees of freedom numerator and 9 denominator), p-value is greater than .10
Actual p-value = .1060
Because p-value > = .05, we cannot reject the null hypothesis.

13 - 2

Analysis of Variance and Experimental Design

d.
Source of Variation
Treatments
Error
Total
3.

a.

x=

Sum of Squares
536
828
1364

Degrees of Freedom
2
9
11

Mean Square
268
92

F
2.91

= 87
15
k

SSTR = n j x j x
j =1

k

b.

j =1

c.

F = MSTR /MSE = 510/38.17 = 13.36

Using F table (2 degrees of freedom numerator and 12 denominator), p-value is less than .01
Actual p-value = .0009
Because p-value = .05, we reject the null hypothesis that the means of the three populations are
equal.

d.

4.

Source of Variation
Treatments
Error
Total

Sum of Squares
1020
458
1478

Degrees of Freedom
2
12
14

Mean Square
510
38.17

F
13.36

Source of Variation
Treatments
Error
Total

Sum of Squares
1200
300
1500

Degrees of Freedom
3
60
63

Mean Square
400
5

F
80

a.

b.

Using F table (3 degrees of freedom numerator and 60 denominator), p-value is less than .01
Actual p-value = .0000
Because p-value = .05, we reject the null hypothesis that the means of the 4 populations are
equal.

5.

a.
Source of Variation
Treatments
Error
Total

Sum of Squares
120
216
336

Degrees of Freedom
2
72
74

13 - 3

Mean Square
60
3

F
20

Chapter 13

b.

Using F table (2 numerator degrees of freedom and 72 denominator), p-value is less than .01
Actual p-value = .0000
Because p-value = .05, we reject the null hypothesis that the 3 population means are equal.

6.
Manufacturer 1
23
6.67

Sample Mean
Sample Variance

Manufacturer 2
28
4.67

Manufacturer 3
21
3.33

x = (23 + 28 + 21)/3 = 24
k

SSTR = n j x j x
j =1

k

j =1

MSE = SSE /(nT - k) = 44.01/(12 - 3) = 4.89

F = MSTR /MSE = 52/4.89 = 10.63
Using F table (2 degrees of freedom numerator and 9 denominator), p-value is less than .01
Actual p-value = .0043
Because p-value = .05, we reject the null hypothesis that the mean time needed to mix a batch
of material is the same for each manufacturer.
7.
Superior
5.75
1.64

Sample Mean
Sample Variance

Peer
5.5
2.00

Subordinate
5.25
1.93

k

SSTR = n j x j x
j =1

k

j =1

MSE = SSE /(nT - k) = 38.99/21 = 1.86

F = MSTR /MSE = 0.5/1.86 = 0.27

13 - 4

Analysis of Variance and Experimental Design

Using F table (2 degrees of freedom numerator and 21 denominator), p-value is greater than .10
Actual p-value = .7660
Because p-value > = .05, we cannot reject the null hypothesis that the means of the three
populations are equal; thus, the source of information does not significantly affect the dissemination
of the information.
8.
Marketing
Managers
5
.8

Sample Mean
Sample Variance

Marketing
Research
4.5
.3

6
.4

k

SSTR = n j x j x
j =1

k

j =1

MSE = SSE /(nT - k) = 7.50/(18 - 3) = .5

F = MSTR /MSE = 3.5/.50 = 7.00
Using F table (2 degrees of freedom numerator and 15 denominator), p-value is less than .01
Actual p-value = .0071
Because p-value = .05, we reject the null hypothesis that the mean perception score is the same
for the three groups of specialists.
9.
Real Estate
Agent
67.73
117.72

Sample Mean
Sample Variance

Architect
61.13
180.10

Stockbroker
65.80
137.12

k

SSTR = n j x j x
j =1

k

j =1

13 - 5

Chapter 13

MSE = SSE /(nT - k) = 6089.16/(45-3) = 144.98

F = MSTR /MSE = 172.74/144.98 = 1.19
Using F table (2 degrees of freedom numerator and 42 denominator), p-value is greater than .10
Actual p-value = .3143
Because p-value > = .05, we cannot reject the null hypothesis that the job stress ratings are the
same for the three occupations.
10.

The Minitab output is shown below:

Analysis of Variance
Source
DF
SS
Factor
2
226.1
Error
21
640.8
Total
23
866.9
Level
Small
Medium
Large

N
8
8
8

Pooled StDev =

Mean
92.200
89.650
84.800

MS
113.0
30.5

StDev
6.365
4.973
5.129

5.524

F
3.70

P
0.042

Individual 95% CIs For Mean

Based on Pooled StDev
---------+---------+---------+------(-------*--------)
(-------*-------)
(--------*-------)
---------+---------+---------+------85.0
90.0
95.0

Because p-value = .05, we reject the null hypothesis that the mean service ratings are equal.

11. a

1 1
1 1
LSD = t / 2 MSE + = t.025 5.5 + = 2.179 2.2 = 3.23
ni n j
5 5

x1 x2 = 30 45 = 15 > LSD; significant difference

x1 x3 = 30 36 = 6 > LSD; significant difference
x2 x3 = 45 36 = 9 > LSD; significant difference

b.

1 1
x1 x2 t / 2 MSE +
n1 n2
1 1
(30 45) 2.179 5.5 +
5 5
-15 3.23 = -18.23 to -11.77

12. a.
Sample Mean
Sample Variance

Sample 1
51
96.67

Sample 2
77
97.34

13 - 6

Sample 3
58
81.99

Analysis of Variance and Experimental Design

x = (51 + 77 + 58)/3 = 62
k

SSTR = n j x j x
j =1

k

j =1

MSE = SSE /(nT - k) = 828/(12 - 3) = 92

F = MSTR /MSE = 724/92 = 7.87
Using F table (2 degrees of freedom numerator and 9 denominator), p-value is between .01 and .025
Actual p-value = .0106
Because p-value = .05, we reject the null hypothesis that the means of the three populations are
equal.

b.

1 1
1 1
LSD = t / 2 MSE + = t.025 92 + = 2.262 46 = 15.34
ni n j
4 4

x1 x2 = 51 77 = 26 > LSD; significant difference

x1 x3 = 51 58 = 7 < LSD; no significant difference
x2 x3 = 77 58 = 19 > LSD; significant difference

13.

1 1
1 1
LSD = t / 2 MSE + = t.025 4.89 + = 2.262 2.45 = 3.54
4 4
n1 n3
Since x1 x3 = 23 21 = 2 < 3.54, there does not appear to be any significant difference between
the means of population 1 and population 3.

14.

x1 x2 LSD
23 - 28 3.54
-5 3.54 = -8.54 to -1.46

15.

Since there are only 3 possible pairwise comparisons we will use the Bonferroni adjustment.

= .05/3 = .017
t.017/2 = t.0085 which is approximately t.01 = 2.602

13 - 7

Chapter 13

1 1
1 1
BSD = 2.602 MSE + = 2.602 .5 + = 1.06
ni n j
6 6

x1 x2 = 5 4.5 = .5 < 1.06; no significant difference

x1 x3 = 5 6 = 1 < 1.06; no significant difference
x2 x3 = 4.5 6 = 1.5 > 1.06; significant difference
16. a.
Machine 1
7.1
1.21

Sample Mean
Sample Variance

Machine 2
9.1
.93

Machine 3
9.9
.70

Machine 4
11.4
1.02

k

SSTR = n j x j x
j =1

k

j =1

MSE = SSE /(nT - k) = 19.30/(24 - 4) = .97

F = MSTR /MSE = 19.26/.97 = 19.86
Using F table (3 degrees of freedom numerator and 20 denominator), p-value is less than .01
Actual p-value = .0000
Because p-value = .05, we reject the null hypothesis that the mean time between breakdowns is
the same for the four machines.
b.

Note: t /2 is based upon 20 degrees of freedom

1 1
1 1
LSD = t / 2 MSE + = t.025 0.97 + = 2.086 .3233 = 1.19
ni n j
6 6

17.

= .05/6 = .008 and /2 = .004

Since the smallest value for /2 in the t table is .005, we will use t.005 = 2.845 as an approximation
for t.004 (20 degrees of freedom)

13 - 8

Analysis of Variance and Experimental Design

1 1
BSD = 2.845 0.97 + = 1.62
6 6
Thus, if the absolute value of the difference between any two sample means exceeds 1.62, there is
sufficient evidence to reject the hypothesis that the corresponding population means are equal.
Means
| Difference |
Significant ?
18.

(1,2)
2
Yes

(1,3)
2.8
Yes

(1,4)
4.3
Yes

(2,3)
0.8
No

(2,4)
2.3
Yes

(3,4)
1.5
No

n1 = 8 n2 = 8 n3 = 8
t /2 is based upon 21 degrees of freedom
1 1
LSD = t.025 30.5 + = 2.080 7.6250 = 5.74
8 8
Comparing Small and Medium
92.20 89.65 = 2.55 < LSD; no significant difference
Comparing Small and Large
92.20 84.80 = 7.40 > LSD; significant difference
Comparing Medium and Large
89.65 84.80 = 4.85 < LSD; no significant difference

19. a.

k

SSTR = n j x j x
j =1

b.

c.

s12 = 164.4

s22 = 131.2

s32 = 110.4

j =1

d.

e.

F = MSTR /MSE = 744/135.3 = 5.50

Using F table (2 degrees of freedom numerator and 15 denominator), p-value is between .01 and .
025

13 - 9

Chapter 13

Actual p-value = .0162

Because p-value = .05, we reject the hypothesis that the means for the three treatments are
equal.
20. a.
Source of Variation
Treatments
Error
Total

b.

Sum of Squares
1488
2030
3518

Degrees of Freedom
2
15
17

Mean Square
744
135.3

F
5.50

Mean Square
75
5.33

F
14.07

1 1
1 1
LSD = t / 2 MSE + = 2.131 135.3 + = 14.31
ni n j
6 6

|156-142| = 14 < 14.31; no significant difference

|156-134| = 22 > 14.31; significant difference
|142-134| = 8 < 14.31; no significant difference

21.
Source of Variation
Treatments
Error
Total
22. a.

Sum of Squares
300
160
460

Degrees of Freedom
4
30
34

H0: u1 = u2 = u3 = u4 = u5
Ha: Not all the population means are equal

b.

Using F table (4 degrees of freedom numerator and 30 denominator), p-value is less than .01
Actual p-value = .0000
Because p-value = .05, we reject H0

23.
Source of Variation
Treatments
Error
Total

Sum of Squares
150
250
400

Degrees of Freedom
2
16
18

Mean Square
75
15.63

F
4.80

Using F table (2 degrees of freedom numerator and 16 denominator), p-value is between .01 and .
025
Actual p-value = .0233
Because p-value = .05, we reject the null hypothesis that the means of the three treatments are
equal.

13 - 10

Analysis of Variance and Experimental Design

24.
Source of Variation
Treatments
Error
Total

Sum of Squares
1200
600
1800

Degrees of Freedom
2
44
46

Mean Square
600
13.64

F
43.99

Using F table (2 degrees of freedom numerator and 44 denominator), p-value is less than .01
Actual p-value = .0000
Because p-value = .05, we reject the hypothesis that the treatment means are equal.
25.
A
119
146.89

Sample Mean
Sample Variance

x=

B
107
96.43

C
100
173.78

= 107.93
28
k

SSTR = n j x j x
j =1

k

j =1

MSE = SSE /(nT - k) = 3,460 /(28 - 3) = 138.4

F = MSTR /MSE = 809.95 /138.4 = 5.85
Using F table (2 degrees of freedom numerator and 25 denominator), p-value is less than .01
Actual p-value = .0082
Because p-value = .05, we reject the null hypothesis that the means of the three treatments are
equal.
26. a.
Source of Variation
Treatments
Error
Total

b.

Sum of Squares
4560
6240
10800

Degrees of Freedom
2
27
29

Mean Square
2280
231.11

F
9.87

Using F table (2 degrees of freedom numerator and 27 denominator), p-value is less than .01
Actual p-value = .0006

13 - 11

Chapter 13
Because p-value = .05, we reject the null hypothesis that the means of the three assembly
methods are equal.
27.
Source of Variation
Between
Error
Total

Sum of Squares
61.64
23.41
85.05

Degrees of Freedom
3
20
23

Mean Square
20.55
1.17

F
17.56

Using F table (3 degrees of freedom numerator and 20 denominator), p-value is less than .01
Actual p-value = .0000
Because p-value = .05, we reject the null hypothesis that the mean breaking strength of the four
cables is the same.
28.
50
33
32

Sample Mean
Sample Variance

60
29
17.5

70
28
9.5

x = (33 + 29 + 28)/3 = 30
k

SSTR = n j x j x
j =1

k

j =1

MSE = SSE /(nT - k) = 236 /(15 - 3) = 19.67

F = MSTR /MSE = 35 /19.67 = 1.78
Using F table (2 degrees of freedom numerator and 12 denominator), p-value is greater than .10
Actual p-value = .2104
Because p-value > = .05, we cannot reject the null hypothesis that the mean yields for the three
temperatures are equal.
29.

Sample Mean
Sample Variance

Direct
Experience
17.0
5.01

13 - 12

Indirect
Experience
20.4
6.26

Combination
25.0
4.01

SSTR = n j x j x
j =1

k

j =1

MSE = SSE /(nT - k) = 91.68 /(21 - 3) = 5.09

F = MSTR /MSE = 112.84 /5.09 = 22.17
Using F table (2 degrees of freedom numerator and 18 denominator), p-value is less than .01
Actual p-value = .0000
Because p-value = .05, we reject the null hypothesis that the means for the three groups are
equal.
30.
Paint 1
13.3
47.5

Sample Mean
Sample Variance

Paint 2
139
.50

Paint 3
136
21

Paint 4
144
54.5

k

SSTR = n j x j x
j =1

k

j =1

MSE = SSE /(nT - k) = 692 /(20 - 4) = 43.25

F = MSTR /MSE = 110 /43.25 = 2.54
Using F table (3 degrees of freedom numerator and 16 denominator), p-value is between .05 and .10
Actual p-value = .0931
Because p-value > = .05, we cannot reject the null hypothesis that the mean drying times for the
four paints are equal.
31.
Sample Mean
Sample Variance

A
20
1

B
21
25

x = (20 + 21 + 25)/3 = 22

13 - 13

C
25
2.5

Chapter 13

SSTR = n j x j x
j =1

k

j =1

MSE = SSE /(nT - k) = 24 /(15 - 3) = 2

F = MSTR /MSE = 35 /2 = 17.5
Using F table (2 degrees of freedom numerator and 12 denominator), p-value is less than .01
Actual p-value = .0003
Because p-value = .05, we reject the null hypothesis that the mean miles per gallon ratings are
the same for the three automobiles.
32.

Note: degrees of freedom for t /2 are 18

1 1
1 1
LSD = t / 2 MSE + = t.025 5.09 + = 2.101 1.4543 = 2.53
ni n j
7 7

x1 x2 = 17.0 20.4 = 3.4 > 2.53; significant difference

x1 x3 = 17.0 25.0 = 8 > 2.53; significant difference
x2 x3 = 20.4 25 = 4.6 > 2.53; significant difference

33.

Note: degrees of freedom for t /2 are 12

1 1
1 1
LSD = t / 2 MSE + = t.025 2 + = 2.179 .8 = 1.95
ni n j
5 5

x1 x2 = 20 21 = 1 < 1.95; no significant difference

x1 x3 = 20 25 = 5 > 1.95; significant difference
x2 x3 = 21 25 = 4 > 1.95; significant difference

34.

Treatment Means:
xg1 = 13.6 xg2 = 11.0

xg3 = 10.6

13 - 14

Analysis of Variance and Experimental Design

Block Means:
x1g = 9 x2 g = 7.67

x3g = 15.67

x4 g = 18.67

x5g = 7.67

Overall Mean:
x = 176/15 = 11.73
Step 1

SST = xij x
i

Step 2

SSTR = b xgj x
j

Step 3
SSBL = k xi g x
i

= 3 [ (9 - 11.73) 2 + (7.67 - 11.73) 2 + (15.67 - 11.73) 2 +

(18.67 - 11.73) 2 + (7.67 - 11.73) 2 ] = 312.32

Step 4
SSE = SST - SSTR - SSBL = 354.93 - 26.53 - 312.32 = 16.08
Source of Variation
Treatments
Blocks
Error
Total

Sum of Squares
26.53
312.32
16.08
354.93

Degrees of Freedom
2
4
8
14

Mean Square
13.27
78.08
2.01

F
6.60

Using F table (2 degrees of freedom numerator and 8 denominator), p-value is between .01 and .025
Actual p-value = .0203
Because p-value = .05, we reject the null hypothesis that the means of the three treatments are
equal.
35.
Source of Variation
Treatments
Blocks
Error
Total

Sum of Squares
310
85
35
430

Degrees of Freedom
4
2
8
14

Mean Square
77.5
42.5
4.38

F
17.69

Using F table (4 degrees of freedom numerator and 8 denominator), p-value is less than .01

13 - 15

Chapter 13

Actual p-value = .0005

Because p-value = .05, we reject the null hypothesis that the means of the treatments are equal.
36.
Source of Variation
Treatments
Blocks
Error
Total

Sum of Squares
900
400
500
1800

Degrees of Freedom
3
7
21
31

Mean Square
300
57.14
23.81

F
12.60

Using F table (3 degrees of freedom numerator and 21 denominator), p-value is less than .01
Actual p-value = .0001
Because p-value = .05, we reject the null hypothesis that the means of the treatments are equal.
37.

Treatment Means:
xg1 = 56 xg2 = 44
Block Means:
x1g = 46 x2 g = 49.5

x3g = 54.5

Overall Mean:
x = 300/6 = 50
Step 1

SST = xij x
i

Step 2

SSTR = b xgj x
j

Step 3
SSBL = k xi g x
i

13 - 16

Analysis of Variance and Experimental Design

Step 4
SSE = SST - SSTR - SSBL = 310 - 216 - 73 = 21
Source of Variation
Treatments
Blocks
Error
Total

Sum of Squares
216
73
21
310

Degrees of Freedom
1
2
2
5

Mean Square
216
36.5
10.5

F
20.57

Using F table (1 degree of freedom numerator and 2 denominator), p-value is between .025 and .05
Actual p-value = .0453
Because p-value = .05, we reject the null hypothesis that the mean tune-up times are the same
for both analyzers.
38.
Source of Variation
Treatments
Blocks
Error
Total

Sum of Squares
45
36
19
100

Degrees of Freedom
4
3
12
19

Mean Square
11.25
12
1.58

F
7.12

Using F table (4 degrees of freedom numerator and 12 denominator), p-value is less than .01
Actual p-value = .0035
Because p-value = .05, we reject the null hypothesis that the mean total audit times for the five
auditing procedures are equal.
39.

Treatment Means:
xg1 = 16 xg2 = 15

xg3 = 21

Block Means:
x1g = 18.67 x2 g = 19.33

x3g = 15.33

x4 g = 14.33

x5g = 19

Overall Mean:
x = 260/15 = 17.33
Step 1

SST = xij x
i

Step 2

SSTR = b xgj x
j

13 - 17

Chapter 13

Step 3

SSBL = k xi g x
i

= 3 [ (18.67 - 17.33) 2 + (19.33 - 17.33) 2 + + (19 - 17.33) 2 ] = 64.75

Step 4
SSE = SST - SSTR - SSBL = 175.33 - 103.33 - 64.75 = 7.25
Source of Variation
Treatments
Blocks
Error
Total

Sum of Squares
100.33
64.75
7.25
175.33

Degrees of Freedom
2
4
8
14

Mean Square
51.67
16.19
.91

F
56.78

Using F table (2 degrees of freedom numerator and 8 denominator), p-value is less than .01
Actual p-value = .0000
Because p-value = .05, we reject the null hypothesis that the mean times for the three systems
are equal.
40.

The Minitab output for these data is shown below:

Analysis of Variance
Source
DF
SS
Factor
3
19805
Error
36
10744
Total
39
30550
Level
ArmCrank
Shovel
SnowThro

N
10
10
10
10

Pooled StDev =

Mean
178.00
171.00
175.00
123.60

MS
6602
298

StDev
16.77
18.89
14.80
18.36

17.28

F
22.12

P
0.000

Individual 95% CIs For Mean

Based on Pooled StDev
-----+---------+---------+---------+(---*----)
(---*----)
(---*---)
(---*----)
-----+---------+---------+---------+125
150
175
200

Because p-value =.05, we reject the null hypothesis that the mean heart rate for the four methods
are equal.
41.

13 - 18

Factor B
Level 1

Level 2

Factor A
Level 3

Means

Level 1

x11 = 150

x12 = 78

x13 = 84

x1g = 104

Level 2

x21 = 110

x22 = 116

x23 = 128

x2 g = 118

xg1 = 130

xg2 = 97

xg3 = 106

x = 111

Factor A

Factor B Means

Step 1

SST = xijk x
i

Step 2

SSA = br x j g x
i

Step 3
SSB = ar xgj x
j

Step 4

i

= 2 [ (150 - 104 - 130 + 111) 2 + (78 - 104 - 97 + 111) 2 +

+ (128 - 118 - 106 + 111) 2 ] = 4,392

Step 5
SSE = SST - SSA - SSB - SSAB = 9,028 - 588 - 2,328 - 4,392 = 1,720
Source of Variation
Factor A
Factor B
Interaction
Error
Total

Sum of Squares
588
2328
4392
1720
9028

Degrees of Freedom
1
2
2
6
11

Mean Square
588
1164
2196
286.67

F
2.05
4.06
7.66

Factor A: F = 2.05
Using F table (1 degree of freedom numerator and 6 denominator), p-value is greater than .10
Actual p-value = .2022
Because p-value > = .05, Factor A is not significant

13 - 19

Chapter 13

Factor B: F = 4.06
Using F table (2 degrees of freedom numerator and 6 denominator), p-value is between .05 and .10
Actual p-value = .0767
Because p-value > = .05, Factor B is not significant
Interaction: F = 7.66
Using F table (2 degrees of freedom numerator and 6 denominator), p-value is between .01 and .025
Actual p-value = .0223
Because p-value = .05, Interaction is significant
42.
Source of Variation
Factor A
Factor B
Interaction
Error
Total

Sum of Squares
26
23
175
56
280

Degrees of Freedom
3
2
6
24
35

Mean Square
8.67
11.50
29.17
2.33

F
3.72
4.94
12.52

Using F table for Factor A (3 degrees of freedom numerator and 24 denominator), p-value is .025
Because p-value = .05, Factor A is significant.
Using F table for Factor B (2 degrees of freedom numerator and 24 denominator), p-value is
between .01 and .025
Actual p-value = .0160
Because p-value = .05, Factor B is significant.
Using F table for Interaction (6 degrees of freedom numerator and 24 denominator), p-value is less
than .01
Actual p-value = .0000
Because p-value = .05, Interaction is significant
43.

13 - 20

Factor B
Small

Factor B

Large

Means

x11 = 10

x12 = 10

x1g = 10

Factor A B

x21 = 18

x22 = 28

x2 g = 23

x31 = 14

x32 = 16

x3g = 15

Means

xg1 = 14

xg2 = 18

x = 16

Factor B

Step 1

SST = xijk x
i

Step 2

SSA = br xi g x
i

Step 3
SSB = ar xgj x
j

Step 4

i

= 2 [ (10 - 10 - 14 + 16) 2 + + (16 - 15 - 18 +16) 2 ] = 56

Step 5
SSE = SST - SSA - SSB - SSAB = 544 - 344 - 48 - 56 = 96
Source of Variation
Factor A
Factor B
Interaction
Error
Total

Sum of Squares
344
48
56
96
544

Degrees of Freedom
2
1
2
6
11

Mean Square
172
48
28
16

F
172/16 = 10.75
48/16 = 3.00
28/16 = 1.75

Using F table for Factor A (2 degrees of freedom numerator and 6 denominator), p-value is
between .01 and .025
Actual p-value = .0104

13 - 21

Chapter 13

Because p-value = .05, Factor A is significant; there is a difference due to the type of
Using F table for Factor B (1 degree of freedom numerator and 6 denominator), p-value is greater
than .01
Actual p-value = .1340
Because p-value > = .05, Factor B is not significant; there is not a significant difference due to
Using F table for Interaction (2 degrees of freedom numerator and 6 denominator), p-value is greater
than .10
Actual p-value = .2519
Because p-value > = .05, Interaction is not significant.

44.
Roller
Coaster

Factor B
Screaming
Demon

Log
Flume

Factor A
Means

Method 1

x11 = 42

x12 = 48

x13 = 48

x1g = 46

Method 2

x21 = 50

x22 = 48

x23 = 46

x2 g = 48

Means

xg1 = 46

xg2 = 48

xg3 = 47

x = 47

Factor A

Factor B

Step 1

SST = xijk x
i

Step 2

SSA = br xi g x
i

Step 3
SSB = ar xgj x
j

Step 4

13 - 22

i

= 2 [ (41 - 46 - 46 + 47) 2 + + (44 - 48 - 47 + 47) 2 ] = 56

Step 5
SSE = SST - SSA - SSB - SSAB = 136 - 12 - 8 - 56 = 60
Source of Variation
Factor A
Factor B
Interaction
Error
Total

Sum of Squares
12
8
56
60
136

Degrees of Freedom
1
2
2
6
11

Mean Square
12
4
28
10

F
12/10 = 1.2
4/10 = .4
28/10 = 2.8

Using F table for Factor A (1 degree of freedom numerator and 6 denominator), p-value is greater
than .10
Actual p-value = .3153
Because p-value > = .05, Factor A is not significant
Using F table for Factor B (2 degrees of freedom numerator and 6 denominator), p-value is greater
than .10
Actual p-value = .6870
Because p-value > = .05, Factor B is not significant
Using F table for Interaction (2 degrees of freedom numerator and 6 denominator), p-value is greater
than .10
Actual p-value = .1384
Because p-value > = .05, Interaction is not significant
45.
Factor B

Male
Factor A
Female
Factor B

Means

Factor A

Under junior
high school

Senior high
school

above college

x11 =29.5114

x12 =35.2943

x13 =50.3092

x 21 =20.6957

x 1 =25.1036

x 22 =27.1644

x2 =31.2294

x 23 = 42.053

Step 1

13 - 23

x3 =46.1809

Means

x1 =38.3716

x 2 =29.9709
x =34.1713

Chapter 13

SST = xijk x
i

= 3,270.7354

Step 2

SSA = br xi g x
i

Step 3
SSB = ar xgj x
j

= 940.4381

Step 4

i

= 2[(29.5114 - 38.3716 - 25.1036 + 34.1713 ) 2 + + (42.0526

- 29.9709 - 46.1809 + 34.1713) 2] = 0.2663

13 - 24

Analysis of Variance and Experimental Design

Step 5
SSE = SST - SSA - SSB - SSAB = 2,118.3143
Source of Variation
Factor A
Factor B
Interaction
Error
Total

Sum of
Squares
211.7166
940.4381
0.2663
2,118.3143
3,270.7353

Degrees of Freedom
1
2
2
24
29

Mean Square
211.7166
470.2191
0.1332
88.2631

F
2.3987
5.3275
0.0015

Using F table for Factor A (1 degree of freedom numerator and 24 denominator), p-value is greater
than .1
Actual p-value = .1345
Because p-value > = .05, Factor A (gender) is not significant
Using F table for Factor B (2 degrees of freedom numerator and 24 denominator), p-value is less
than .05
Actual p-value = .0122
Because p-value = .05, Factor B (educational level) is significant
Using F table for Interaction (2 degrees of freedom numerator and 24 denominator), p-value is
greater than .9
Actual p-value =.9985
Because p-value > = .05, Interaction is not significant
46.

x1g = (1.13 + 1.56 + 2.00)/3 = 1.563

x2 g = (0.48 + 1.68 + 2.86)/3 = 1.673
xg1 = (1.13 + 0.48)/2 = 0.805
xg2 = (1.56 + 1.68)/2 = 1.620
xg3 = (2.00 + 2.86)/2 = 2.43
x = (1.13 + 1.56 + 2.00 + 0.48 + 1.68 + 2.86)/6 = 1.618
Step 1
SST = 327.50 (given in problem statement)
Step 2

13 - 25

Chapter 13

SSA = br xi g x
i

Step 3
SSB = ar xgj x
j

Step 4

i

= 25[(1.13 - 1.563 - 0.805 + 1.618) 2 + (1.56 - 1.563 - 1.62

+ 1.618) 2 + + (2.86 - 1.673 - 2.43 + 1.618) 2] = 14.2525

Step 5
SSE = SST - SSA - SSB - SSAB = 327.50 - 0.4538 - 66.0159 - 14.2525
Source of Variation
Factor A
Factor B
Interaction
Error
Total

Sum of Squares
0.4538
66.1059
14.2525
246.7778
327.5000

Degrees of Freedom
1
2
2
144
149

Mean Square
0.4538
33.0080
7.1263
1.7137

F
0.2648
19.2608
4.1583

Factor A: Actual p-value = .6076. Because p-value > = .05, Factor A is not significant.
Factor B: Actual p-value = .0000. Because p-value = .05, Factor B is significant.
Interaction: Actual p-value = .0176. Because p-value = .05, Interaction is significant.
47. a.
Area 1
96
50

Sample Mean
Sample Variance

Area 2
94
40

x = (96 + 94)/2 = 95
k

SSTR = n j x j x
j =1

k

j =1

MSE = SSE /(nT - k) = 270 /(8 - 2) = 45

F = MSTR/MSE = 8/45 = .18
Using F table (1 degree of freedom numerator and 6 denominator), p-value is greater than .10

13 - 26

Actual p-value = .6862

Because p-value > = .05, the means are not significantly different.
b.
Area 1
96
50

Sample Mean
Sample Variance

Area 2
94
40

Area 3
83
42

x = (96 + 94 + 83)/3 = 91
k

SSTR = n j x j x
j =1

k

j =1

MSTR = SSE /(nT - k) = 396 /(12 - 3) = 44

F = MSTR /MSE = 196 /44 = 4.45
Using F table (2 degrees of freedom numerator and 6 denominator), p-value is between .05 and .10
Actual p-value = .0653
Because p-value > = .05, we cannot reject the null hypothesis that the mean asking prices for all
three areas are equal.
48.

The Minitab output for these data is shown below:

Analysis of Variance
Source
DF
SS
Factor
2
753.3
Error
27
546.9
Total
29
1300.2
Level
SUV
Small
FullSize

N
10
10
10

Pooled StDev =

Mean
58.600
48.800
60.100

MS
376.6
20.3

StDev
4.575
4.211
4.701

4.501

F
18.59

P
0.000

Individual 95% CIs For Mean

Based on Pooled StDev
---------+---------+---------+------(-----*-----)
(-----*----)
(-----*-----)
---------+---------+---------+------50.0
55.0
60.0

Because the p-value = .000 < = .05, we can reject the null hypothesis that the mean resale value is
the same. It appears that the mean resale value for small pickup trucks is much smaller than the
mean resale value for sport utility vehicles or full-size pickup trucks.

13 - 27

Chapter 13

49.

The Minitab output is shown below:

Analysis of Variance
Source
DF
SS
Factor
3
2.603
Error
36
10.612
Total
39
13.215
Level
Midcap
Smallcap
Hybrid
Specialt

N
10
10
10
10

Mean
1.2800
1.6200
1.6000
2.0000

Pooled StDev =

MS
0.868
0.295

F
2.94

P
0.046

Individual 95% CIs For Mean

Based on Pooled StDev
-------+---------+---------+--------(--------*--------)
(-------*--------)
(--------*--------)
(--------*--------)
-------+---------+---------+--------1.20
1.60
2.00

StDev
0.2394
0.3795
0.7379
0.6583

0.5429

Because p-value = .05, we reject the null hypothesis that the mean expense ratios are equal.
50.

Sample Mean
Sample Variance
x=

Physical
Therapist
63.7
164.68

Lawyer
50.0
124.22

Cabinet
Maker
69.1
105.88

Systems
Analyst
61.2
136.62

= 61
4
k

SSTR = n j x j x
j =1

k

j =1

MSE = SSE /(nT - k) = 4782.6 /(40 - 4) = 132.85

F = MSTR /MSE = 646.47 /132.85 = 4.87
Using F table (3 degrees of freedom numerator and 36 denominator), p-value is less than .01
Actual p-value = .0061
Because p-value = .05, we reject the null hypothesis that the mean job satisfaction rating is the
same for the four professions.

13 - 28

51.

The Minitab output for these data is shown below:

Analysis of Variance
Source
DF
SS
Factor
2
4339
Error
27
15991
Total
29
20330
Level
West
South
NE

N
10
10
10

Pooled StDev =

Mean
108.00
91.70
121.10

MS
2169
592

F
3.66

P
0.039

Individual 95% CIs For Mean

Based on Pooled StDev
---+---------+---------+---------+--(-------*-------)
(-------*-------)
(-------*------)
---+---------+---------+---------+--80
100
120
140

StDev
23.78
19.62
28.75

24.34

Because the p-value = .039 < = .05, we can reject the null hypothesis that the mean rate for the
three regions is the same.
52.

The Minitab output is shown below:

ANALYSIS OF VARIANCE
SOURCE
DF
SS
FACTOR
3
1271.0
ERROR
36
1744.2
TOTAL
39
3015.2

LEVEL
West
South
N.Cent
N.East

N
10
10
10
10

POOLED STDEV =

MEAN
60.000
45.400
47.300
52.100

MS
423.7
48.4

STDEV
7.218
7.610
6.778
6.152

6.961

F
8.74

p
0.000

INDIVIDUAL 95 PCT CI'S FOR MEAN

BASED ON POOLED STDEV
--+---------+---------+---------+---(------*-----)
(------*-----)
(------*-----)
(-----*------)
--+---------+---------+---------+---42.0
49.0
56.0
63.0

Because p-value = 0.05, we reject the null hypothesis that that the mean base salary for art
directors is the same for each of the four regions.
53.

The Minitab output for these data is shown below:

Analysis of Variance
Source
DF
SS
Factor
2
12.402
Error
37
24.596
Total
39
36.998
Level
Guard
Tackle

N
15
13
12

Pooled StDev =

Mean
7.4133
6.1077
7.0583

MS
6.201
0.665

StDev
0.8855
0.7399
0.8005

0.8153

13 - 29

F
9.33

P
0.001

Individual 95% CIs For Mean

Based on Pooled StDev
------+---------+---------+---------+
(-------*------)
(-------*------)
(-------*-------)
------+---------+---------+---------+
6.00
6.60
7.20
7.80

Chapter 13
Because p-value = .05, we reject the null hypothesis that the mean rating for the three positions
is the same. It appears that wide receivers and tackles have a higher mean rating than guards.
54.
x
92
30

Sample Mean
Sample Variance

y
97
6

z
84
35.33

x = (92 + 97 + 44) /3 = 91
k

SSTR = n j x j x
j =1

k

j =1

MSE = SSE /(nT - k) = 213.99 /(12 - 3) = 23.78

F = MSTR /MSE = 172 /23.78 = 7.23
Using F table (2 degrees of freedom numerator and 9 denominator), p-value is between .01 and .025
Actual p-value = .0134
Because p-value = .05, we reject the null hypothesis that the mean absorbency ratings for the
three brands are equal.

55.

The EXCEL output is shown below:

Analysis of Variance
Summary
Level

Sum

Average Variation

Professionals

1241.8

155.225

20.66211

Services

1183.9

147.9875

1.64185

Manufacturing

1147.78

143.4725

2.296879

SS

DF

MS

Factor

562.3677

281.1839

34.28954

Error

172.2059

21

8.200281

Total

734.5736

23

ANOVA
Source

13 - 30

P-value
0.000000243

Analysis of Variance and Experimental Design

Because p-value = .05, we reject the null hypothesis that the mean consumption is the same for
the three sectors.

13 - 31

Chapter 13

56.
Method A
90
98.00

Sample Mean
Sample Variance

Method B
84
168.44

Method C
81
159.78

x = (90 + 84 + 81) /3 = 85
k

SSTR = n j x j x
j =1

k

j =1

MSE = SSE /(nT - k) = 3,836 /(30 - 3) = 142.07

F = MSTR /MSE = 210 /142.07 = 1.48
Using F table (2 degrees of freedom numerator and 27 denominator), p-value is greater than .10
Actual p-value = .2455
Because p-value > = .05, we can not reject the null hypothesis that the means are equal.
57.
Type A
32,000
2,102,500

Sample Mean
Sample Variance

Type B
27,500
2,325,625

Type C
34,200
2,722,500

Type D
30,300
1,960,000

k

SSTR = n j x j x
j =1

= 30(32,000 - 31,000) 2 + 30(27,500 - 31,000) 2 + 30(34,200 - 31,000) 2 +

30(30,300 - 31,000) 2 = 719,400,000

k

SSE = (n j 1) s 2j = 29(2,102,500) + 29(2,325,625) + 29(2,722,500) + 29(1,960,000)

j =1

= 264,208,125
MSE = SSE /(nT - k) = 264,208,125 /(120 - 4) = 2,277,656.25
F = MSTR /MSE = 239,800,000 /2,277,656.25 = 105.28
Using F table (3 degrees of freedom numerator and 116 denominator), p-value is less than .01
Actual p-value = .0000
Because p-value = .05, we reject the null hypothesis that the population means are equal.

13 - 32

58.
Design A
90
82.67

Sample Mean
Sample Variance

Design B
107
68.67

Design C
109
100.67

k

SSTR = n j x j x
j =1

k

j =1

MSE = SSE /(nT - k) = 756.03 /(12 - 3) = 84

F = MSTR /MSE = 436 /84 = 5.19
Using F table (2 degrees of freedom numerator and 9 denominator), p-value is between .025 and .05
Actual p-value = .0317
Because p-value = .05, we reject the null hypothesis that the mean lifetime in hours is the same
for the three designs.
59. a.
Nonbrowser
4.25
1.07

Sample Mean
Sample Variance

Light Browser
5.25
1.07

Heavy Browser
5.75
1.36

k

SSTR = n j x j x
j =1

k

j =1

MSW = SSW /(nT - k) = 24.5 /(24 - 3) = 1.17

F = MSB /MSW = 4.67 /1.17 = 3.99
Using F table (2 degrees of freedom numerator and 21 denominator), p-value is between .025 and .
05
Actual p-value = .0340
Because p-value = .05, we reject the null hypothesis that the mean comfort scores are the same

13 - 33

Chapter 13

for the three groups.

b.

1 1
1 1
LSD = t / 2 MSW + = 2.080 1.17 + = 1.12
ni n j
8 8

Since the absolute value of the difference between the sample means for nonbrowsers and light
browsers is 4.25 5.25 = 1 , we cannot reject the null hypothesis that the two population means are
equal.

60. a.

Treatment Means:
x1 = 22.8 x2 = 24.8

x3 = 25.80

Block Means:
x1 = 19.67 x2 = 25.67

x3 = 31

x4 = 23.67

x5 = 22.33

Overall Mean:
x = 367 /15 = 24.47
Step 1

SST = xij x
i

Step 2

SSTR = b xgj x
j

Step 3
SSBL = k xi g x
i

= 3 [ (19.67 - 24.47) 2 + (25.67 - 24.47) 2 + + (22.33 - 24.47) 2 ] = 217.02

Step 4
SSE = SST - SSTR - SSBL = 253.73 - 23.33 - 217.02 = 13.38
Source of Variation
Treatment
Blocks
Error
Total

Sum of Squares
23.33
217.02
13.38
253.73

Degrees of Freedom
2
4
8
14

Mean Square
11.67
54.26
1.67

F
6.99
32.49

Using F table (2 degrees of freedom numerator and 8 denominator), p-value is between .01 and .025
Actual p-value = .0175

13 - 34

Analysis of Variance and Experimental Design

Because p-value = .05, we reject the null hypothesis that the mean miles per gallon ratings for
the three brands of gasoline are equal.
b.
I
22.8
21.2

Sample Mean
Sample Variance

II
24.8
9.2

III
25.8
27.2

k

SSTR = n j x j x
j =1

k

j =1

MSE = SSE /(nT - k) = 230.4 /(15 - 3) = 19.2

F = MSTR /MSE = 11.67 /19.2 = .61
Using F table (2 degrees of freedom numerator and 12 denominator), p-value is greater than .10
Actual p-value = .4406
Because p-value > = .05, we cannot reject the null hypothesis that the mean miles per gallon
ratings for the three brands of gasoline are equal.
Thus, we must remove the block effect in order to detect a significant difference due to the brand of
gasoline. The following table illustrates the relationship between the randomized block design and
the completely randomized design.

Sum of Squares
SST
SSTR
SSBL
SSE

Randomized
Block Design
253.73
23.33
217.02
13.38

Completely
Randomized Design
253.73
23.33
does not exist
230.4

Note that SSE for the completely randomized design is the sum of SSBL (217.02) and SSE (13.38)
for the randomized block design. This illustrates that the effect of blocking is to remove the block
2
effect from the error sum of squares; thus, the estimate of for the randomized block design is
substantially smaller than it is for the completely randomized design.
61.

The blocks correspond to the 15 items in the market basket (Product) and the treatments correspond
to the grocery chains (Store).
The Minitab output is shown below:
Analysis of Variance for Price

13 - 35

Chapter 13
Source
Product
Store
Error
Total

DF
14
2
28
44

SS
217.236
2.728
6.769
226.734

MS
15.517
1.364
0.242

F
64.18
5.64

P
0.000
0.009

Because the p-value for Store = .009 is less than = .05, there is a significant difference in the
mean price per item for the three grocery chains.
62.

The Minitab output for these data is shown below:

Analysis of Variance
Source
DF
SS
Factor
2
731.75
Error
63
247.42
Total
65
979.17
Level
UK
US
Europe

N
22
22
22

Mean
12.052
14.957
20.105

Pooled StDev =

MS
365.88
3.93

StDev
1.393
1.847
2.536

1.982

F
93.16

P
0.000

Individual 95% CIs For Mean

Based on Pooled StDev
---+---------+---------+---------+--(--*--)
(--*--)
(--*--)
---+---------+---------+---------+--12.0
15.0
18.0
21.0

Because the p-value = .05, we can reject the null hypothesis that the mean download time is the
same for Web sites located in the three countries. Note that the mean download time for Web sites
located in the United Kingdom (12.052 seconds) is less than the mean download time for Web sites
in the United States (14.957) and Web sites located in Europe (20.105).
63.
Factor B
Spanish

French

Factor A
German

Means

System 1

x11 = 10

x12 = 12

x13 = 14

x1g = 12

System 2

x21 = 8

x22 = 15

x23 = 19

x2 g = 14

xg1 = 9

xg2 = 13.5

xg3 = 16.5

x = 13

Factor A

Factor B Means

Step 1

SST = xijk x
i

Step 2

SSA = br xi g x
i

Step 3

13 - 36

SSB = ar xgj x
j

Step 4

i

= 2 [(8 - 12 - 9 + 13) 2 + + (22 - 14 - 16.5 +13) 2] = 26

Step 5
SSE = SST - SSA - SSB - SSAB = 204 - 12 - 114 - 26 = 52
Source of Variation
Factor A
Factor B
Interaction
Error
Total

Sum of Squares
12
114
26
52
204

Degrees of Freedom
1
2
2
6
11

Mean Square
12
57
12
8.67

F
1.38
6.57
1.50

Factor A: Actual p-value = .2846. Because p-value > = .05, Factor A (translator) is not significant.
Factor B: Actual p-value = .0308. Because p-value = .05, Factor B (language translated) is
significant.
Interaction: Actual p-value = .2963. Because p-value > = .05, Interaction is not significant.
64.
Factor B
Manual
Machine 1

Automatic

Factor B
Means

x11 = 32

x12 = 28

x1g = 36

Machine 2 x21 = 21

x22 = 26

x2 g = 23.5

xg2 = 27

x = 26.75

Factor A

Factor B Means

xg1 = 26.5

Step 1

SST = xijk x
i

Step 2

SSA = br xi g x
i

Step 3

13 - 37

Chapter 13

SSB = ar xgj x
j

Step 4

i

= 2[(30 - 30 - 26.5 + 26.75) 2 + + (28 - 23.5 - 27 + 26.75) 2]

= 40.5

Step 5
SSE = SST - SSA - SSB - SSAB = 151.5 - 84.5 - 0.5 - 40.5 = 26
Source of Variation
Factor A
Factor B
Interaction
Error
Total

Sum of Squares
84.5
.5
40.5
26
151.5

Degrees of Freedom
1
1
1
4
7

Mean Square
84.5
.5
40.5
6.5

F
13
.08
6.23

Factor A: Actual p-value = .0226. Because p-value = .05, Factor A (machine) is significant.
Factor B: Actual p-value = .7913. Because p-value > = .05, Factor B (loading system) is not
significant.
Interaction: Actual p-value = .0671. Because p-value > = .05, Interaction is not significant.

13 - 38